Bank Capital, Competition and Loan Spreads

Transcription

1 Bank Capital, Competition and Loan Spreads Markus Fischer Julian Mattes Sascha Steffen January 31, 2011 Abstract This paper empirically investigates whether well-capitalized banks charge higher spreads using a dataset of all syndicated loans issued by public non-financial U.S. borrowers during the 1993 to 2007 period. We find evidence suggesting that well-capitalized banks charge higher spreads. We further investigate whether this result can be explained by banks holding-up their borrowers. Using various information asymmetry and competition proxies, we cannot reject the null hypothesis that all borrowers pay a premium borrowing from well-capitalized banks. In other words, this premium is not competed away even for the most transparent firms as well as in the most competitive lending environments. Our results might also suggest that an increase in regulatory capital requirements will lead to an increase in borrowing costs for all firms. JEL classification: G14; G21; G30 Keywords: Capital ratio; Competition; Information Asymmetry; Loan spreads Contact author; Goethe University Frankfurt, Frankfurt. Tel: +49(69) , Facsimile +49(69) , Fischer@finance.uni-frankfurt.de Goethe University Frankfurt, Frankfurt. Tel +49(69) , Facsimile +49(69) , mattes@wiwi.uni-frankfurt.de University of Mannheim, Mannheim. Tel: +49(621) , Facsimile +49(621) , steffen@bank.bwl.uni-mannheim.de

2 1 Introduction Do well-capitalized banks charge higher loan spreads than poorly-capitalized banks? This is an important question as since the mid-1990s capital ratios in U.S. banking organizations have substantially exceeded even the highest supervisory standards (so-called capital buffer ). Berger et al. (2008) document that as of June 2007, the 67 BHCs with assets exceeding $10 billion had a mean Tier 1 leverage ratio of 7.63%, a Tier 1 risk-based ratio of 9.38%, and a total risk-based capital ratio of 11.97%. However, only few studies study the impact of bank capital on the cost of corporate debt 1. Using a dataset of all syndicated loans issued by public non-financial U.S. borrowers during the 1993 to 2007 period, we empirically investigate whether banks charge higher loan spreads for having high capital ratios. The literature provides several arguments as to why banks hold more capital than required by regulators 2. First, banks might hold more capital due to reputational reasons. As pointed out by Holmstrom and Tirole (1997), the moral hazard problem in lending can be overcome or at least be attenuated through the amount of capital a bank has at stake since a bank s capital is a credible commitment to dutifully monitor a borrower. Further, Allen et al. (2009) argue that corporate borrowers may prefer to deal with well-capitalized banks because this increases the banks incentives to monitor. The banks monitoring effort determines the probability of success of the investment project. This suggests that in loan markets, an intermediate market structure between perfect competition and monopoly, borrowers prefer banks that hold high levels of capital because they can commit to high levels of monitoring and thereby providing value to the borrower (Leland and Pyle, 1977; Diamond, 1984; Allen, 1990). These banks, however, might charge higher loan rates as a certification premium 3. 1 Hubbard et al. (2002) is the notable exception. 2 Capital regulation in the banking industry is deemed as main rationale for the reduction of moral hazard by banks (Santos, 2001) because banks have incentives to take on excessive risk due to limited liability (Freixas and Rochet, 2008). This incentive is reduced if banks have their own capital at risk. However, traditionally capital adequacy standards are seen as binding because capital is seen as more expensive than other sources of funds (e.g. Pennacchi, 1988). 3 If higher capitalization improves a bank s reputation, this would be consistent with the literature (e.g. Allen, 1984; Fang, 2005) that reputable firms can charge higher prices for their services or products. 2

3 Additionally, borrowers may prefer to establish relationships with banks with higher capital buffers in order to secure future financing (e.g. Berger et al., 1995, 2008). Thus, we expect that banks with higher capital ratios can charge higher spreads ( spread premium ) compared to banks with lower capital ratios. Our results indicate that the All-in-Spread-Drawn (AISD) is positive correlated to the total risk-based capital ratio with an effect that ranges from 40 to 44 basis points, suggesting that well-capitalized banks charge higher spreads than their not so well-capitalized peers 4. In other words, we find evidence that well-capitalized 5 banks can indeed charge a spread premium. If banks price high capital ratios into loan spreads, it is an interesting question to ask whether loans are priced differently for transparent versus opaque firms with respect to this loan spread component. As argued in Rajan (1992), banks might be able to charge higher loan spreads to informationally opaque firms that do not have outside funding options and thereby hold-up opaque borrowers. If higher loan costs induced by high capital ratios are simply an artifact of hold-up, this higher spread can be competed away at least for informationally transparent firms. We proxy switching costs as it is commonly done with measures of borrower s information opacity (e.g. Sufi, 2007; Bharath et al., 2009; Saunders and Steffen, 2010) and employ six different measures of information opacity, namely: (i) no credit rating, (ii) prime dependency, (iii) low asset size, (iv) low market capitalization, (v) firm being young and (vi) no NYSE listing. Taken together, our results confirm that information opacity increases the spread significantly but this is independent of the bank s capitalization. We find no evidence that well-capitalized banks particularly lock-in informational opaque borrowers and therefore our previously identified spread premium exists for opaque as well as transparent firms. 4 If we replace the capital ratio level by a dummy indicating high capital ratio, this does not qualitatively alter our results. Further, it provides evidence that our level results are not driven by outliers with extremely high capital ratios. 5 We use the term well-capitalized not in the sense of the definition of the Federal Deposit Insurance Corporation (FDIC), i.e. total risk-based capital ratio > 10%, but rather to identify banks with higher capital ratios than other banks. 3

4 Petersen and Rajan (1994) argue that competition reduces a bank s lending fees. Therefore we analyze if competition affects the higher loan spreads demanded by well-capitalized banks. More specifically, we proxy competition with different concentration measures (Bikker and Haaf, 2002) calculated on a single metropolitan statistical areas (MSA) level, namely, the Herfindahl-Hirschman Index (HHI), Top-3 lender market share, Top-5 lender market share and number of banks present in a MSA. Additionally, we incorporate, broadly comparable to Carey and Nini (2007), a dummy for the activity of foreign banks. However, our results suggest that less competition does not explain the loan spread premium charged by a well-capitalized bank. Overall, even after controlling for different proxies for information asymmetry and competition, we cannot reject the hypothesis that all borrowers pay for banks having high capital ratios. Specifically, our paper contributes to the literature in three important ways. First, by focusing on a period during which banks continuously increased their capital ratios, we find contradicting evidence to a widely cited argument in the literature, that weak banks charge higher loan spreads (Hubbard et al., 2002). Second, we provide evidence for a loan spread component which is not competed away and might at least partially explain why loan spreads are higher in the U.S. than in Europe 6 (the loan spread puzzle according to Carey and Nini (2007)) 7. Third, our findings also have implications with respect to the current debate about increasing the capital requirements for banks, i.e. our results suggest that an increase in capital requirements will lead to an increase in borrowing costs for all firms. This paper proceeds as follows. Section 2 gives an overview of related literature resulting in the formulation of hypotheses. Section 3 describes the dataset and provides some descriptive statistics and univariate tests. In section 4, we set out our methodology, discuss our results and provide evidence for their robustness. Section 5 concludes. 6 Several studies show that risk-based capital ratios are higher in the United States than in Europe (e.g. Bikker and Metzemakers, 2007; Brewer III et al., 2008; Fonseca and González, 2010). 7 However, further analyses needs to be undertaken to demystify the loan spread puzzle. 4

5 2 Hypotheses The literature on how reputation has an impact on the level of prices and the quality of issues can be at least traced back to Klein and Leffler (1981) and Shapiro (1983). Among others, Allen (1984) shows that reputable firms can charge higher prices for their products. This relation seems also to hold for the financial service industry as suggested by Booth and Smith (1986), Chemmanur and Fulghieri (1994) and Fang (2005) who all find that reputable banks can charge higher prices (fees) for their services. There are several ways to seize reputation, for example, Coleman et al. (2006) approximate reputation with lender monitoring skills. The authors find that banks with superior monitoring skills can charge higher loan spreads because they are able to reduce moral hazard problems. The bank s incentive for diligently monitoring the borrower and thereby providing value to the borrower (Leland and Pyle, 1977; Diamond, 1984; Allen, 1990) is also positively related to the bank s capital, as argued by Allen et al. (2009) 8. There exists a further well-developed research line why borrowers should prefer banks with more capital than the minimum requirement because borrower-lender relationships are costly to replace in the event that a bank fails (Berger et al., 1995, 2008) or when a bank has to cut back lending in order to reduce the risk of capital inadequacy in the future (Chen, 2001; Van den Heuvel, 2009). In both cases a bank is not able to award a loan to a previous borrower which might put the borrower in difficulties to obtain financing (Slovin et al., 1999). Therefore capital ratios are probably the most credible measure of a bank s risk and quality (Berger et al., 2008) and thereby an effective signal of a bank s reputation. Related to this, as shown by Billett et al. (1995), a lender with higher reputation leads to larger abnormal stock-market returns for the borrower. These strands of reasoning suggests that borrowers prefer banks to hold high levels of capital as a way to commit to high levels of monitoring and in order to secure future financing 8 The loan rate might also provide incentive for the bank, however, a borrower s effort requirement as in Holmstrom and Tirole (1997) reduces the attractiveness of the loan rate as an incentive tool and increases thereby the importance of capital. 5

6 via a prospering relationship. Banks being aware of this reputational effect might motivate a certification bonus by charging higher loan spreads to these borrowers. We translate this into our first hypothesis: Hypothesis 1 (H1): Well-capitalized banks charge higher spreads for loans compared to their not so well-capitalized peers. If banks price high capital ratios into loan spreads, it is an interesting question to ask whether loans are priced differently for transparent versus opaque firms with respect to this loan spread component. As argued in Rajan (1992) and shown in Booth (1992), banks might be able to charge higher loan spreads to informational opaque firms which lack alternative outside funding options and thereby hold-up the borrower. If higher loan costs induced by high capital ratios are simply an artifact of hold-up, this higher spread can be competed away at least for informational transparent firms. However, if loans are not priced differently for opaque versus transparent firms, information transparency does not explain the loan spread premium charged by well-capitalized banks. Hypothesis 2 is formulated as the following: Hypothesis 2 (H2): Well-capitalized banks charge higher spreads to informationally opaque firms. Petersen and Rajan (1994) show that competition reduces a bank s lending fees. Further, Hubbard et al. (2002) mention that absent informational frictions, in a competitive loan market, the loan interest rate charged by a bank to a borrower should reflect the bank s cost of funds and the risk characteristics of the borrower. Regarding competition in the syndicated loan market Rhodes (2004) notes that the number of banks participating in the market and capable of arranging syndicated transactions continues to decline but competition remains fierce nonetheless. Since competition can hardly be measured directly, it is commonly approximated by concentration in the financial intermediation literature. For example, Bikker and Haaf (2002) conclude that their findings are providing support for the conventional view that concentration impairs competitiveness. However, according to Allen et al. (2009), 6

7 reputable banks might be able to charge higher spreads independent of the competition environment. This gives us hypothesis 3: Hypothesis 3 (H3): Less competition cannot explain the loan spread premium charged by well-capitalized banks. In the next section a description of the dataset is offered including a summary statistic and univariate tests. 3 Data and sample selection 3.1 Data Sources To empirically investigate whether or not well-capitalized banks charge higher spreads, we construct a dataset using three different data sources, namely the Loan Pricing Corporation Dealscan (henceforth LPC) database, the Compustat database and the Call Reports from the Bank Regulatory Database. First we merge the information on loan transactions from LPC with the borrowing firms financial statements from Compustat. LPC contains detailed information on worldwide syndicated loan originations, e.g. contract terms, lender identities and their roles within the syndicate, as well as borrower identity information (i.e. name, region, country, and SIC industry classification). Since LPC focuses primarily on information of the loan contract, we use Compustat to obtain financial information on publicly-listed borrowers, such as total assets or market capitalization. Unfortunately, LPC does not offer a unique identifier, so we have to hand-match the LPC s borrower names with the one of Compustat where we use a rather conservative approach in a similar fashion as Santos and Winton (2008). To guarantee that we use only accounting information that was publicly available at the time of the loan origination, we use the accounting data from the year before the loan was syndicated. Since Compustat data is only available for publicly listed firms, we label all other firms as privately 7

8 held firms. After appending the accounting information we exclude all loans from financial service firms (SIC codes between 6000 and 6799) we turn to the information provided in the Call Reports. Of course, with this approach our analysis is only possible for banks subject to the Basel accord for capital requirements. However, as pointed out by Calomiris and Pornrojnangkool (2009), the loan market is dominated by commercial banks 9. We start by comparing the LPC s bank names, i.e. lead arranger names, with the bank names stated in the Call Reports and are able to retrieve the RSSD number for the majority of banks and use this unique identifier to obtain financial statement information of these banks. However, only in a small number of cases we find exactly the same bank names in LPC and Call Reports 10. For those banks that we are not able to match directly due to a lack of appropriate names in the Call Reports, we used the Institution Search function provided on the website of the National Information Center 11. With the help of this routine, we are able to identify RSSD numbers for nearly all banks in the lender universe given in LPC. It is important to point out that our dataset also contains banks that are now (based on the year 2007 status) extinct due to merger activities or failures 12. The bank s data is from the quarter prior to the issuance of the loan to quickly grasp changes in the lending environment. 3.2 Descriptive statistics After matching the sample with the borrower and bank data, our final dataset consists of all U.S. syndicated loans originated during the 1993 to 2007 period, i.e. 21,051 loan agreements 9 More accurate, the authors show that approximately 95% of all loans in their sample period, 1992 to 2002, have commercial banks in lead arranger roles. 10 For example, referring to the LPC lender parent name for Citigroup Inc., we come across various lender subsidiary names such as Citi, Citibank and Citibank NA, whereas in the Call Reports this subsidiary is labeled Citibank NA. 11 In order to prevent hasty bank matches, we take advantage of the historical organization hierarchy for the bank holding companies provided by the National Information Center. 12 In other words, we have accounting information on now extinct banks, such as Chemical Bank and Fleet Bank. 8

9 with an average facility amount of USD 380 million 13. Table 1 shows a detailed summary statistics for borrower, bank and loan characteristics. The mean All-in-Spread-Drawn (AISD) is 141 basis points (bps). Further, borrowers are large by both asset size (mean USD 5,504 million) and market capitalization (mean USD 5,170 million). Regarding bank characteristics, it is clearly visible that the mean total riskbased capital ratio (henceforth: total capital ratio) is 11.6% and therefore much higher than the minimum requirement of 8% as demanded by the Federal Deposit Insurance Corporation (FDIC) and the Bank for International Settlements (BIS). The average tier-1 risk-based capital ratio (8.4%) is also above the minimum requirement of 4%. Our observations are consistent with those of Flannery and Rangan (2008) and Berger et al. (2008). [Table 1] Furthermore, the descriptive banks statistics, such as coverage ratio (i.e., ratio of loanloss allowance to non-current loans) and the ratio of loan-loss provision to (net) charge-offs rate are broadly in line with the figures in the Quarterly Banking Profiles by the FDIC 14. Clearly visible is the huge kurtosis of banks total assets (in constant 2000 USD). We further show in Table 1 the first and fourth quartile values of the aforementioned variables. 3.3 Univariate tests Table 2 splits the entire sample of Table 1 into banks with low total capital ratios and those with high total capital ratios. In order to obtain similar group sizes we divide the sample at the median of the total capital ratios 15. The first two columns report the mean values (medians in parentheses) for various loan contract terms as well as borrower and bank characteristics for the two separate groups. The last column provides t-statistics for difference in means (z-statistics for Wilcoxon Rank sum test in parentheses). 13 Most of the lost observations are due to a lack of information about the borrower or the bank. For example, we have to exclude all privately held companies due to lack of borrower information. 14 The Quarterly Banking Profile is a quarterly publication by the FDIC that provides an aggregate summary of financial characteristics for all FDIC-insured institutions. 15 Therefore banks with high total capital ratios have total risk-based capital ratios above the median. 9

10 [Table 2] Banks in the high total capital ratio group can on average charge 5 basis points higher spreads than banks with low total capital ratios. This difference is significant at the 1% level. Banks in the low total capital ratio group serve more loans for LBO or acquisition purposes. Again these differences are significant at the 1% level. Additionally, loans by well-capitalized banks have a shorter maturity but a larger loan amount on average. The Wilcoxon Rank sum tests support the results of the t-tests. Both banks groups have similar ratios of total loans to total assets, however banks in the low total capital ratio group have on average more total assets. Regarding borrower characteristics, banks with high total capital ratios do not only serve less non-investment grade rated and not-rated firms but also borrowers with (slightly) higher leverage. Further, borrowers from banks with high total capital ratios are larger in terms of total assets. Overall, the univariate results suggest that well-capitalized banks can charge higher spreads. However, they also show that characteristics potentially influencing the cost of loans differ across the two groups. Therefore, to separate the effect of a bank s capitalization from other influences, we turn to a multivariate framework in the next section. 4 Multivariate analysis and hypothesis tests For our multivariate analysis we start with the base analysis testing whether there is evidence that well-capitalized banks, as measured by the total capital ratio, charge higher spreads. Following this, we introduce a new dummy variable indicating a high total capital ratio for all further regressions. 10

11 4.1 Methodology Our base regression set-up for testing the impact of a bank s capitalization on the pricing of syndicated loans, is given by 16 : n AISD i = β 0 + β 1 Capital.Ratio.Measure i + β Lk (Loan.Characteristics i ) + + k=1 n β F k (F irm.characteristics i ) + k=1 n β Ck (Controls i ) + ɛ i k=1 n β Bk (Bank.Characteristics i ) ˆ AISD: The dependent variable All-in-Spread-Drawn is measured in basis points and is the spread plus annualized upfront fees above LIBOR. ˆ Capital.Ratio.M easure: Ratio of bank s total capital to total risk-weighted assets. We use two different set-ups: The absolute value of a bank s total capital ratio measured in logs. An indicator variable equal to one, if a bank s total capital ratio is greater than a specified cut-off point. ˆ Loan.Characteristics i : Two different loan characteristics are included, namely Facility Size (logs) and Maturity (logs). ˆ F irm.characteristics i : Various characteristics of the borrower, more specifically: Total Assets (logs), Leverage, Current Ratio, Tangibility Ratio, Coverage (logs), Marketto-Book, Profitability, Investment Grade Rating and Not Rated ˆ Bank.Characteristics i : Different risk measures of the lead arranging bank, namely: Bank Assets (logs), Bank Coverage Ratio, Bank Loss Prov. to Charge-offs (logs+1) and Bank Total Loans to Total Assets. ˆ Control k : Other control variables include loan purpose controls, loan type dummy variables, industry classification (one-digit SIC code) and calendar year dummies. 16 The included variables are explained in more detail in the appendix. k=1 11

12 4.2 Do banks charge for maintaining high capital ratios? (H1) The regression results for our base set-up on the individual bank level for the period 1993 to 2007 are given in Table 3. Loan, borrower and bank characteristics included in our regressions are discussed in detail below. Columns (A) to (E) differ with respect to the included bank characteristics, but all regressions include time fixed effects (year dummies) and industry fixed effects (one-digit SIC code dummies). Standard errors are heteroscedastic robust and clustered at the bank level. Summarizing, across all six specifications our results indicate that the All-in-Spread-Drawn (AISD) is positive correlated to the total capital ratio with an effect that ranges from 40 to 44 basis points (bps), suggesting that well-capitalized banks charge higher spreads than not so well-capitalized banks. In all specifications the capital ratio variable is significant at the 1% level. [Table 3] We include various bank characteristics in our analysis to control for that our capital ratio coefficient does not pick up other effects, such as loan portfolio quality. In column (A) we use the total capital ratio without any other bank characteristics. The total capital ratio (in logs) coefficient is positive (+40 bps) 17 and significant at the 1% level. We additionally include bank total assets (also measured in logs) which increases the total capital ratio coefficient to 44 bps (significant at the 1% level). Our loan portfolio risk measures are taken from the Quarterly Banking Profile of the FDIC and we include in column (C) the bank s coverage ratio which is defined as the ratio of loan loss allowance to non-current loans 18. We expect a negative sign since a higher ratio implies a higher risk buffer. Consistently, we find that the coefficient is -12 bps and significant at the 5% level. In the next column we additionally use the ratio of loan 17 It should be noticed that we are dealing with a so-called level-log model where the dependent variable is in level form and some of the independent variables are in logarithmic form. Then formally the β 1 interpretation is given by: y = (β 1 /100)% x. 18 Of course, we calculate all portfolio risk measures on the individual bank level. 12

13 loss provision to loan (net) charge-offs. Again, the coefficient is negative and significant at the 1% level. Lastly, we add in column (E) the ratio of a bank s total loans to its total assets and find that the coefficient is negative and significant at the 10% level. The total capital ratio coefficient stays nearly unaltered with being significant at the 1% level across all specifications. With respect to borrower characteristics we obtain signs in line with the literature (e.g. Santos and Winton, 2008; Bharath et al., 2009), i.e. negative for market-to-book, tangibility, coverage, profitability and the current ratio. All these borrower characteristics are always significant at least at the 5% level except for tangibility. The coefficients of market-to-book, coverage (in logs), profitability and current ratio are -5 bps, -16 bps, -57 bps and -3 bps, respectively, across all specifications. The coefficient of tangibility (-5 bps) is not significant at a meaningful level of confidence. A positive coefficient is present for leverage (+32 bps) which is significant at the 5% level. For maturity (in logs) we obtain, comparable to Hubbard et al. (2002) and Santos and Winton (2008), a negative coefficient which is significant at the 5% level. The various loan purposes, included as dummy variables in our analysis, show higher spreads for recapitalization, acquisition and leverage buy-outs at the 1% significance level. However, the increase in spread ranges from 12 bps (Recapitalization) to 111 bps (LBO). Bridge loans (+60 bps) and revolvers with a maturity of less than one year (+17 bps) also show positive coefficients. Our findings for loan purpose and loan type are are comparable to the literature (e.g. Carey and Nini, 2007; Bharath et al., 2009). In order to address potential concerns that syndicate structure and loan characteristics are jointly determined, potentially inducing a simultaneity bias in our estimates (Wooldridge, 2002), we estimate the set-up of column (E) in reduced form (i.e., without all loan characteristics) in column (F) 19. We obtain coefficients closely resembling the ones found in columns (A) to (E). Interestingly, the largest increase is shown by the total capital ratio (+55 bps) 19 This approach is commonly used to deal with the identification issue which loan term is endogenous or exogenous (e.g. Sufi, 2007). 13

14 which further supports our hypothesis that well-capitalized banks charge higher spreads. 4.3 The impact of regulation on loan spreads The most striking observation in Table 3 is the positive correlation between the All-in-Spread- Drawn of a syndicated loan and the total capital of a bank. Although clearly visible and significant, our results may even understate the effect for two reasons. First, we use only publicly listed companies and thus have a sample which is likely to be biased towards large companies. Thus, additionally including smaller companies might probably enhance the capitalization effect. Secondly, although we control for loan, borrower, and bank characteristics, other loan terms such as covenants could also notably influence the loan spreads (Nini et al., 2009). Interestingly, we identify a positive impact of a bank s capitalization on loan spreads which is in contrast to Hubbard et al. (2002) who identify a negative correlation between the AISD and capital ratios in their sample. One reason might be the two different measures of capitalization used in these studies. While Hubbard et al. (2002) use equity capital to total assets (equity capital ratio), we use total capital (which consists of tier-1 and tier-2 capital) to risk-weighted assets (total capital ratio). The total capital ratio as defined in the Basel Accord by the Bank for International Settlements (BIS) was adopted by U.S. banks at the beginning of The total capital ratio explicitly accounts for different credit risk categories of on-balance and off-balance sheet items. However, to better control for the different capital ratio definitions, we rerun our regression set-up with the equity capital ratio (in logs) in Table 4. [Table 4] Since we now cover the time period from 1987 to 2007 we incorporate an interaction term between equity capital ratio and loan originated after 1992, i.e. after the end of Hubbard et al. (2002) s observation period ( ). Consistent with Hubbard et al. (2002), we find a negative correlation between a bank s equity capital ratio and AISD (significant at least 14

15 at the 5% level) for the years 1987 to However, our interaction term is positive and significant at least at the 5% level. Our results suggest that first, banks with lower capital charge higher spreads but since the beginning of the 1990s this causality has reversed. Our findings are consistent with changes in the banking sector at the beginning of the 1990s 20. A study prepared by the FDIC s Division of Research and Statistics (Federal Deposit Insurance Corporation, 1997) points out that the 1980s and early 1990s were undoubtedly a period of greater stress and turmoil for U.S. financial institutions than any other since the Great Depression. Over this period more than 1,600 commercial and savings banks insured by the FDIC were closed or received FDIC financial assistance. As a consequence, the bank regulatory system came under intense scrutiny, and fundamental questions were raised about its effectiveness in anticipating and limiting the number of bank failures and losses to the deposit insurance fund. Therefore, many responding actions were undertaken by regulators at the beginning of the 1990s, most prominently the Federal Deposit Insurance Corporation Improvement Act (FDICIA) of 1991 which among others dictate prompt corrective actions to be taken as a bank s capital ratio declines to certain levels. The passage of the FDICIA is seen as effective in raising capital ratios and (at least partially) in reducing risk (e.g. Mishkin, 1997). Therefore the capital growth of banks at the beginning of the 1990s can be seen as a rational response to changes in regulatory and market environments. It is not just the result of high profitability in the banking sector since the 1990s (Flannery and Rangan, 2008). Further, as argued by Chen (2001), Van den Heuvel (2009) and Fonseca and González (2010), there are costs to adjusting regulatory capital that impede complete adjustment to a banks target capital at any particular time. Therefore banks hold more capital than required not to violate the de-jure capital standard and thereby avoiding the supervisory penalties (Flannery and Rangan, 2008; Fonseca and González, 2010). So even if not absolutely reliable, capital ratios are probably the most credible measure of 20 Berger et al. (1995) show that the average equity capital ratio rose from 6.21% in 1989 to 8.01% in 1993, representing an increase of nearly 30% in four years. 15

16 risk and quality (Berger et al., 2008) and well-capitalized banks being aware of this fact can so easily motivate a certification bonus by charging higher loan spreads to their borrowers. In combination with Allen et al. (2009) argument that a strong capitalization of a bank is a credible commitment to dutifully monitor a borrower, the aforementioned reasoning should influence a bank s reputation as a credible lender. So our result that well-capitalized banks can charge higher spreads is consistent with Klein and Leffler (1981), Shapiro (1983) and Allen (1984), all of whom argue that higher reputation has a positive impact on the level of prices and the quality of issues. It is also in line with Booth and Smith (1986), Chemmanur and Fulghieri (1994) and Fang (2005) that reputable banks can charge higher fees for their services. In the next section, we introduce a high total capital ratio indicator variable as a robustness to address concerns that our results are driven by a few banks with extraordinary high risk-based capital ratios and explicitly compare banks with high capital ratios to banks with low capital ratios. 4.4 Robustness: Bank capital ratio dummies Table 5 extends our analysis started in Table 3 by replacing the total capital ratio level variable with a dummy variable indicating a high total capital ratio. For columns (A) to (C) the dummy variable indicating a high total capital ratio is one if a bank s total capital ratio is above the median meaning that the total capital ratio dummy cut-off points is 11.1%. [Table 5] The set-up of columns (A) to (C) is comparable to the set-ups of columns (A), (D) and (E) in Table 3 with the exception that the aforementioned capital ratio dummy replace the total capital ratio level variable. The coefficient of the above-median total capital ratio dummy is around 7 basis points (bps) and significant at the 1% level across all three columns. Other included bank control variables have coefficients and signs closely comparable to the ones of Table 3. Therefore, even after controlling for additional loan portfolio risk measures the 16

17 capital ratio dummy is positive, suggesting that borrowers seem to pay a certification bonus to those banks with higher total capital ratios. Further, the magnitude and the significance of the borrower and loan characteristics are nearly unchanged in comparison to Table In columns (D) to (F) we compare the spread of loans which were underwritten by banks with total capital ratios in the fourth quartile with banks with those in the first quartile. In other words, we compare the spreads of loans where lead arrangers had either relatively high or relatively low total capital ratios which means that we have a larger gap between the total capital ratio values of these two groups. In line with our expectations we find that the coefficients of the high total capital ratio are larger than the ones found in column (A) to (C). So banks with total capital ratios in the fourth quartile can ceteris paribus charge between 10 and 13 bps higher spreads for a loan in comparison to banks with total capital ratios in the first quartile. The coefficients of the total capital ratio dummy are significant at the 1% level across the three columns (D) to (F). The obtained coefficients of bank, loan and borrower are comparable to the ones in columns (A) to (C). Overall, we clearly find evidence for a positive correlation between banks capital ratios and pricing of syndicated loans. However, there exists the possibility that well-capitalized banks might use their information monopoly power to prevent borrowers from switching to other lenders. To further explore this circumstance we extend our analysis by including information proxies in the following analysis. 4.5 Bank capital ratios and information asymmetry (H2) An interesting question is if well-capitalized banks exploit their monopoly power to extract higher prices and if borrowers are priced differently with respect to this additional loan spread premium. In other words, if higher loan costs induced by high capital ratios are simply an artifact of hold-up, this higher spread can be competed away at least for informationally transparent firms. To test this we proxy switching costs as it is commonly done with measures 21 We will not further discuss the coefficients of loan, borrower and bank characteristics in great detail if they do not change a lot in the following analyses. 17

18 of borrower s information opacity (e.g. Sufi, 2007; Bharath et al., 2009; Saunders and Steffen, 2010). Our switching cost proxies for public firms are: (i) no S&P senior unsecured debt rating (column A), (ii) prime dependency of the borrower (B), (iii) low asset size (C), (iv) low market capitalization (D) 22. Further, we look at if the firm is young, i.e. IPO is less than three years ago (E), and if the firm s is not listed on the New York Stock Exchange (F). Then each of the six information opacity proxies is interacted with the high total capital ratio dummy ( 11.1%). The results are reported in Table 6. The control variables are the same as in the regression set-up of column (C) in Table 5 but are not repeated here for brevity. Column (A) of Table 6 exhibits a positive high total capital ratio dummy (+9 bps) as well as a positive not-rated dummy (+54 bps). Both dummies are significant at the 1% level. However, the interaction term is not significant at a meaningful level. If the interaction term between high total capital ratio and not rated would be positive and significant, this would suggest that well-capitalized banks use their monopoly power particularly over borrowers without a credit rating. In other words, a lock-in effect would be present 23. By looking at the prime dependency dummy in column (B), we obtain also a positive and significant information asymmetry dummy and a non-significant interaction term, supporting the findings of column (A). [Table 6] However, since nearly 40% of all firms included in our sample have no credit rating and only 3% of the borrowers are prime dependent, we extend our analysis by looking at financial and capital market information and transform each information type in a dummy. Therefore, column (C) includes a low assets dummy which is one if the asset size is in the bottom tercile. The high total capital ratio dummy and the low assets dummy are both positive (+8 bps and +22 bps, respectively) and significant at the 1% level. The interaction term between these two 22 Total Assets and Market Cap are defined as low if in the bottom tercile (in constant 2000 USD). 23 If only the interaction term would be positive and significant, this would suggest a hold-up of informational opaque borrowers by well-capitalized banks. 18

19 variables is not significant with a value of -3 bps, offering no support for the above mentioned hold-up scenario. The results of column (D) (i.e., low market capitalization) confirm the finding of column (C). The high total capital ratio dummy is positive and significant at the 1% level which is also true for the low market capitalization dummy. Once again, the interaction term is not significant. If a firm is young this increases the spread by 8 bps, but the interaction term between high total capital ratio dummy and young firm is not significant at a meaningful level. The non-significant interaction term is also observable for a firm not listed on the NYSE (+ 27 bps). In both columns (E) and (F), the high total capital ratio dummy is positive (+6 bps) and significant at the 1% level. All coefficients of other loan, bank and borrower characteristics of columns (A) to (F) are comparable to the ones of our base set-up in column (C) of Table 5 and are thus, for better visibility not depicted in Table 6. Overall, our results clearly show that the loan spread premium is not simply an artifact of hold-up. 4.6 Bank capital ratios and competition (H3) Following the argument of Petersen and Rajan (1994), a testable hypothesis emerge whether competition mitigates bank s market power by lowering spreads. We approximate competition with concentration as it is commonly done in the financial intermediation literature 24. In the first four columns we use four different concentration measures calculated on a single metropolitan statistical areas (MSA) 25 level, namely the Herfindahl-Hirschman Index (HHI) (column A), Top-3 lender market share (B), Top-5 lender market share (C) and number of banks present (D) 26. All four proxies measure competition but in different ways. The Top-3 24 For example, Bikker and Haaf (2002) conclude that their findings provides support for the conventional view that concentration impairs competitiveness. 25 Metropolitan Statistical Areas are defined by the U.S. Office of Management and Budget (OMB) and are geographic entities with at least one urbanized area of 50,000 or more inhabitants. Currently there are 362 metropolitan statistical areas with the most populous metro areas being New York, Los Angeles and Chicago. 26 Technically, we separate the borrowers according to their MSAs and then calculate each lender s market share of every single MSA for three different five year periods to capture dynamic changes in the lending 19

20 and Top-5 proxies measure the dominance of a few banks within an MSA whereas the HHI measures the concentration of a MSA. The number of total banks acting in a MSA reflects the attractiveness of a MSA. The HHI dummy is one if the HHI of the respective MSA is in the upper tercile of all MSA s HHIs. The Top-3 (Top-5) market share dummy is one, if the market share of the Top-3 (Top-5) banks in a MSA is also in the upper tercile of all MSAs. Low number of total banks is one if the number of total banks acting in a MSA is in the bottom tercile. We also incorporate, broadly comparable to Carey and Nini (2007), a dummy for the activity of foreign banks since foreign banks acting as lead arranger are likely to increase competition. So column (D) includes a dummy that is one if no foreign bank is among the lead arrangers in a loan syndicate. Then each competition proxy is interacted with the high total capital ratio dummy ( 11.1%). The competition dummies in columns (A) to (D) of Table 7 show the expected positive coefficients and are at least significant at the 5% level. The positive coefficients are consistent with findings of Boyd and De Nicolo (2005) and De Graeve et al. (2007), namely that a bank s market share has a positive effect on interest rates charged to firms. It is also in line with the theoretical predictions of Allen et al. (2009) that in market structures that are somewhat between perfect competition and monopoly, a surplus is split between banks and borrowers, with each obtaining a positive expected return. [Table 7] Nevertheless, the high total capital ratio dummies remain positive and significant at the 1% level. Comparable to the case of information opacity, all four interaction terms are not significant at a meaningful level. In other words, the non-significant interaction terms suggest that well-capitalized banks do not additionally lock-in borrowers if they have a dominating position in a MSA. The coefficients of other loan, bank and borrower characteristics of columns (A) to (D) are comparable to the ones of our base set-up in column (C) in Table 5 environment. 20

21 and are therefore not depicted in Table 7. The total capital ratio dummy in column (E) remains positive (+11 bps) significant at the 5% level, but both the no-foreign lead dummy and the interaction term are not significant at a meaningful level. All in all, the results of Table 7 show that the positive correlation between capitalization and spreads is not driven by less competition among banks and thereby amplifying the market power of well-capitalized banks. 4.7 Further robustness tests We perform several robustness tests to support our main finding that well-capitalized banks charge higher spreads 27. Instead of clustering at the bank level only, we simultaneously cluster at the bank and borrower level (Petersen, 2009). The coefficient of the total capital ratio varies, comparable to Table 3, from 38 to 42 basis points (bps) and stays significant at the 1% level across all specifications. If we re-run the set-up of Table 3 for unrated firms only, we obtain coefficients for the total capital ratio between 60 and 69 bps (significant at the 1% level). In a regression for firms with a credit rating the total capital ratio coefficient is around 30 bps and across all set-ups significant at the 1% level. In further tests, we include firm (bank) fixed effects to account for time invariant (even not observed) borrower (bank) characteristics. Using the bank fixed effects estimation, we have an average number of 142 observations per group and an overall R 2 of 0.58 with the total capital ratio being positive at around 22 to 25 basis points and significant at least at the 5% level. For the borrower fixed effect model, the average observation per group is 7 and the overall R 2 is around 0.40 across all specifications. The total capital ratio stays positive and significant at the 1% level and floats between 33 and 42 basis points. Additionally, we modify Table 5 columns (D) to (F) in the sense that we set the high total capital ratio dummy being 1 if the total capital ratio is in the 95 th percentile (P95) and 0 if the ratio is in the 5 th percentile (P5). The coefficient of the total capital ratio dummy is as expected higher compared to the base case and ranges 27 We do not report these tests for brevity but all tables are available upon request. 21

22 between 15 and 19 bps (in all cases significant at the 1% level). Overall, all results of the robustness tests are in line with the ones of the base tests. 5 Conclusion We explore a unique dataset of U.S. syndicated loans from 1993 to 2007 and analyze whether a bank s capitalization influences the cost of corporate borrowing. We provide evidence that well-capitalized banks charge significantly higher spreads. We further ask whether wellcapitalized banks lock-in their borrowers. Using various measures for information asymmetry, we cannot find evidence that opaque firm ceteris paribus pay higher spreads to wellcapitalized banks than transparent firms. Additionally, the loan spread premium charged by well-capitalized banks cannot be explained by less competition among banks. By recalling that capital ratios in the U.S. are higher than in Europe our findings might therefore offer a dimension that is relevant to price but that has received little attention (Carey and Nini, 2007) so far and thereby could at least partially explain why loan costs are higher in the U.S. than in Europe. Further, with respect to the current debate about increasing the capital requirements for banks, our findings suggest that an increase in capital requirements will lead to an increase in borrowing costs for all firms. References Allen, F. (1984). Reputation and product quality. Rand Journal of Economics 15 (3), Allen, F. (1990). The market for information and the origin of financial intermediation. Journal of Financial Intermediation 1 (1), Allen, F., E. Carletti, and R. S. Marquez (2009). Credit market competition and capital regulation. Review of Financial Studies, forthcoming. 22

23 Berger, A. N., R. DeYoung, M. J. Flannery, D. Lee, and O. Oztekin (2008). How do large banking organizations manage their capital ratios? Journal of Financial Services Research 34 (2-3), Berger, A. N., R. J. Herring, and G. P. Szegö (1995). The role of capital in financial institutions. Journal of Banking and Finance 19 (3-4), Bharath, S., S. Dahiya, A. Saunders, and A. Srinivasan (2009). Lending relationships and loan contract terms. Review of Financial Studies, forthcoming. Bikker, J. and P. Metzemakers (2007). Is bank capital pro-cyclical? a cross-country analysis. Kredit und Kapital 40 (2), Bikker, J. A. and K. Haaf (2002). Competition, concentration and their relationship: An empirical analysis of the banking industry. Journal of Banking and Finance 26 (11), Billett, M. T., M. J. Flannery, and J. A. Garfinkel (1995). The effect of lender identity on a borrowing firm s equity return. Journal of Finance 50 (2), Booth, J. R. (1992). Contract costs, bank loans, and cross-monitoring hypothesis. Journal of Financial Economics 31 (1), Booth, J. R. and R. L. Smith (1986). Capital raising, underwriting, and the certification hypothesis. Journal of Financial Economics 15 (1-2), Boyd, J. H. and G. De Nicolo (2005). The theory of bank risk-taking and competition revisited. Journal of Finance 60 (3), Brewer III, E., G. G. Kaufman, and L. D. Wall (2008). Bank capital ratios across countries: Why do they vary? Journal of Financial Services Research 34 (2), Calomiris, C. W. and T. Pornrojnangkool (2009). Relationship banking and the pricing of financial services. Journal of Financial Services Research 35 (3),

24 Carey, M. and G. Nini (2007). Is the corporate loan market globally integrated? a pricing puzzle. Journal of Finance 62 (6), Chemmanur, T. J. and P. Fulghieri (1994). Investment bank reputation, information production, and financial intermediation. Journal of Finance 49 (1), Chen, N.-K. (2001). Bank net worth, asset prices and economic activity. Journal of Monetary Economics 48 (2), Coleman, A. D. F., N. Esho, and I. G. Sharpe (2006). Does bank monitoring influences loan contract terms. Journal of Financial Services Research 30 (2), De Graeve, F., O. De Jonghe, and R. V. Vennet (2007). Competition, transmission and bank pricing policies: Evidence from belgian loan and deposit markets. Journal of Banking and Finance 31 (1), Diamond, D. W. (1984). Financial intermediation and delegated monitoring. Review of Economic Studies 51 (3), Fang, L. H. (2005). Investment bank reputation and the price and quality of underwriting services. Journal of Finance 60 (6), Federal Deposit Insurance Corporation (1997). History of the Eighties - Lessons for the Future. Federal Deposit Insurance Corporation. Flannery, M. J. and K. P. Rangan (2008). What caused the capital build-up of the 1990s? Review of Finance 12 (2), Fonseca, A. R. and F. González (2010). How bank capital buffers vary across countries: The influence of cost of deposits, market power and bank regulation. Journal of Banking and Finance 34 (4), Freixas, X. and J.-C. Rochet (2008). Microeconomics of banking. MIT Press. 24