How Does Capital Affect Bank Performance During Financial Crises?

Allen N. Berger University of South Carolina, Wharton Financial Institutions Center, and CentER Tilburg University Christa H.S. Bouwman Case Western Reserve University and Wharton Financial Institutions Center

March 2011 The recent financial crisis has raised important issues regarding bank capital. Various reform proposals involve requiring banks to hold more capital. But assessing these proposals requires an understanding of how capital affects bank performance. Existing theories produce conflicting predictions regarding the effect of capital on bank performance during normal times and have little to say about the effect during financial crises. This paper addresses these issues empirically by formulating and testing hypotheses regarding the effect of capital on three dimensions of bank performance survival, market share, and profitability during financial crises and normal times. We distinguish between two banking crises and three market crises that occurred in the U.S. over the past quarter century. We have two main results. First, capital helps banks of all sizes during banking crises. Higher capital helps these banks increase their probability of survival, market share, and profitability during such crises. Second, higher capital improves the performance of small banks in all three dimensions during market crises and normal times as well, but the effect on medium and large banks during these periods is less pronounced. Overall, our results suggest that capital is important for small banks at all times and is important for medium and large banks primarily during banking crises. Contact details: Moore School of Business, University of South Carolina, 1705 College Street, Columbia, SC 29208. Tel: 803-576-8440. Fax: 803-777-6876. E-mail: aberger@moore.sc.edu. Contact details: Weatherhead School of Management, Case Western Reserve University, 10900 Euclid Avenue, 362 PBL, Cleveland, OH 44106. Tel.: 216-368-3688. Fax: 216-368-6249. E-mail: christa.bouwman@case.edu. Keywords: Financial Crises, Survival, Market Share, Profitability, Liquidity Creation, and Banking. JEL Classification: G01, G28, and G21. This is a significantly expanded version of the second part of an earlier paper, Financial Crises and Bank Liquidity Creation. A previous version of this paper was entitled: Bank Capital, Survival and Performance Around Financial Crises. The authors thank Nittai Bergman, Lamont Black, Rebel Cole, Bob DeYoung, Mark Flannery, Paolo Fulghieri, Steven Ongena, Bruno Parigi, Peter Ritchken, Katherine Samolyk, Asani Sarkar, James Thomson, Greg Udell, Todd Vermilyae, and participants at presentations at the American Finance Association Meeting, the Bank for International Settlements, the University of Venice, the European School of Management and Technology in Berlin, the CREI / JFI / CEPR Conference on Financial Crises at Pompeu Fabra, the Boston Federal Reserve, the Philadelphia Federal Reserve, the San Francisco Federal Reserve, the Cleveland Federal Reserve, the International Monetary Fund, the Summer Research Conference in Finance at the ISB in Hyderabad, the Unicredit Conference on Banking and Finance, the University of Kansas Southwind Finance Conference, Erasmus University, and Tilburg University for useful comments.

1. Introduction The recent financial crisis has raised fundamental issues about the role of bank equity capital. Various proposals have been put forth which argue that banks should hold more capital (e.g., Kashyap, Rajan, and Stein 2009, Hart and Zingales 2009, Acharya, Mehran, and Thakor 2010, Basel III (2010)). 1 An underlying premise in all of these proposals is that there are externalities due to the safety net provided to banks and thus social efficiency can be improved by requiring banks to operate with more capital, especially during financial crises. Bankers, however, have typically argued that being forced to hold more capital would jeopardize their performance, especially profitability, and the argument that higher capital need not be beneficial has found some support in the academic literature as well (e.g., Calomiris and Kahn 1991). The issue of what effects capital has on bank performance, and how these effects might differ between crises and normal times, thus boils down to an empirical question, and one that we confront in this paper. In particular, the goal of this paper is to empirically examine the effects of bank capital on three dimensions of bank performance probability of survival, market share, and profitability during different types of financial crises and normal times. Theories predict that the effect of bank capital on any of these three dimensions of performance could be positive or negative. As a prelude to a more extensive discussion of this in the next section, here we briefly present the main arguments. Consider survival probability first. Holding fixed the banks asset and liability portfolios, higher capital mechanically implies a higher survival probability (also suggested by monitoringbased theories like Holmstrom and Tirole 1997), but theories on the disciplining role of demandable debt suggest that a bank with more demandable debt and less capital may make better loans that are less likely to default and cause bank failure (e.g., Calomiris and Kahn 1991). Next, consider the issue of market share. While recent banking theories suggest a positive relationship between capital and market share (e.g., Allen, Carletti, and Marquez forthcoming, Mehran and Thakor forthcoming), the literature on the interaction between a nonfinancial firms leverage and its product-market dynamics argues that the relationship is negative (e.g., Brander and Lewis 1986). Finally, consider profitability. If capital improves monitoring incentives as in Holmstrom and Tirole (1997), higher capital could enhance return on equity, whereas the literature on the disciplining role of debt suggests the opposite (Calomiris and Kahn 1991).

Hoshi and Kashyap (2010) draw lessons for the U.S. from Japans efforts to recapitalize its banking sector.

This discussion suggests that although a substantial body of theoretical research is available, these theories focus on different forces and hence produce conflicting implications, pointing strongly to the need for empirical mediation. Understanding the effects of capital is interesting in its own right, but is particularly compelling during times of stress, such as financial crises. While the papers discussed above focus on the role capital plays during normal times, we examine both crises and normal times. We therefore need to extrapolate these theoretical results to the crisis context in a manner which is still consistent with the intuition of the theories, which we do in the next section. For each of the three performance dimensions, we take our cue from the theories and formulate hypotheses that allow us to assess whether capital helps or hurts bank performance. These hypotheses are then tested using data on virtually every U.S. bank from 1984:Q1 until 2009:Q4. We test our survival hypotheses using logit regressions. We regress the log odds ratio of the

probability of survival on the banks pre-crisis capital ratio interacted with a banking crisis dummy, a market crisis dummy, and a normal times dummy, plus a set of control variables. As discussed below, banking crises are those that originated in the banking sector, and market crises are those that originated outside banking in the financial markets. The interaction terms capture the effect of capital on survival during banking crises, market crises, and normal times, respectively. Moreover, we examine small banks (gross total assets or GTA up to $1 billion), medium banks (GTA exceeding $1 billion and up to $3 billion), and large banks (GTA exceeding $3 billion) as three separate groups, since the effect of capital likely differs by bank size (e.g., Berger and Bouwman 2009). 2 Our results support the hypothesis that capital enhances the survival probability for banks of all sizes during banking crises and, in the case of small banks, also during market crises and normal times. We test our market share hypotheses by defining market share in terms of the banks share of aggregate bank liquidity creation. We prefer this measure over the banks market share of assets because liquidity creation is a better representation of bank output than assets alone, since liquidity creation takes into account all on- and off-balance sheet activities. 3 We regress the percentage change in market share during the crisis on the banks average pre-crisis capital ratio (interacted with the banking crisis, market crisis, and

Gross Total Assets or GTA equals total assets plus the allowance for loan and lease losses and the allocated transfer risk reserve (a reserve for certain foreign loans). Total assets on Call Reports deduct these two reserves, which are held to cover potential credit losses. We add these reserves back to measure the full value of the loans financed. Section 5.4 shows results based on alternative cutoffs between medium and large banks. 3 Robustness checks yield comparable results based on gross total assets market share (see Section 5.1).

normal times dummies mentioned above) and a set of control variables. Our results support the hypothesis that capital helps to increase market shares for banks of all sizes during banking crises and normal times, and for small banks also during market crises. To test our profitability hypotheses, we run regressions that are similar to the market share regressions except that we use the change in return on equity (ROE) during a crisis as the dependent variable. ROE is an appropriate profitability measure since both net income (the numerator) and equity (the denominator) reflect all of the banks on- and off-balance sheet activities.4 Our results support the hypothesis that capital improves profitability for small banks at all times, for medium banks during market crises, and for large banks during banking and market crises. We perform a variety of robustness checks. First, to check the sensitivity of the results to our definitions of various performance measures, we use alternative definitions of survival, market share, and profitability. Second, we use regulatory capital ratios instead of the equity-to-assets ratio to define capital. Third, we drop banks that are Too-Big-To-Fail from the large-bank sample to see if our large-bank conclusions are driven by the dominance of a few very large banks. Fourth, we use alternative cutoffs to separate medium and large banks to examine the sensitivity of our results to the manner in which banks are classified by size. Fifth, we measure pre-crisis capital ratios averaged over the four quarters before the crisis or one quarter before the crisis starts rather than averaging them over the eight quarters before the crisis. The purpose is to determine if the time period over which pre-crisis capital is defined affects our results. Sixth, while the theories suggest a causal relationship from capital to performance, we recognize that in practice both may be jointly determined. Although our main regression analyses use lagged capital to mitigate this potential endogeneity problem, we also address it more directly using an instrumental variable approach. While most of our analyses focus on the effect of book capital on the performance of individual banks, we also perform analyses for listed entities (listed banks and listed bank holding companies). Specifically, we examine how book and market capital ratios affect the performance of these organizations during crises and normal times. We can broadly summarize our findings as follows. First, capital is especially beneficial during

A robustness check based on return on assets (ROA), which divides net income by total assets, a measure of the banks on-balance sheet activities, shows similar results for small banks and weaker results for medium and large banks (see Section 5.1).

banking crises. It helps banks of all sizes during such crises higher pre-crisis capital increases the odds of survival and enhances market share for banks of all sizes and it increases profitability for all but medium-sized banks during such crises. Second, small banks benefit in all respects from higher capital during market crises and normal times as well. For large and medium banks, higher capital improves only profitability during market crises and only market share during normal times. While the survival and market share results are highly robust, the profitability results are less so. The remainder of this paper is organized as follows. Section 2 develops the empirical hypotheses. Section 3 explains our empirical approach, describes the financial crises and normal times, describes all the variables and the sample, and provides summary statistics. Section 4 discusses the results of our empirical tests. Section 5 includes the robustness tests. Section 6 contains the additional analysis of listed entities. Section 7 concludes.

2. Development of the empirical hypotheses In this section, we review existing theories to formulate our empirical hypotheses about the effects of bank capital on the survival probability, market share, and profitability of banks during crises and normal times.

2.1. Survival Hypothesis 1a: Capital enhances the banks survival probability during crises and normal times. Hypothesis 1b: Capital diminishes the banks survival probability during crises and normal times.

There are several reasons, grounded in existing theories, to believe that capital improves a banks survival probability. First, there is a set of theories that emphasizes the role of capital as a buffer to absorb shocks to earnings (e.g., Repullo 2004, Von Thadden 2004). If the banks portfolio, screening, and monitoring choices are held fixed, then this buffer role immediately implies that higher capital increases the probability of the banks survival. 5 We can view this as the mechanical effect of higher capital. Second, there is another set of theories that focuses on the incentive effects of capital. This set includes theories based on monitoring, screening, and asset-substitution-moral-hazard. The monitoring-based papers include Holmstrom and Tirole (1997), Allen, Carletti, and Marquez (forthcoming), and Mehran and Thakor (forthcoming). A key result in
5

We know from various theories, however, that all of these choices are influenced by the banks capital structure.

these papers is that higher bank capital induces higher levels of borrower monitoring by the bank, thereby reducing the probability of default or otherwise improving the banks survival odds indirectly by increasing the surplus generated by the bank-borrower relationship. 6 In their screening-based theory of banking with behaviorally-biased agents, Coval and Thakor (2005) show that a minimum amount of capital may be essential to the very viability of the bank. The asset-substitution-moral-hazard theories argue that government

guarantees cause shareholders to prefer low capital and excessive risk to increase the value of the deposit insurance put option (Merton 1977), and that banks with more capital will optimally choose less risky portfolios (see the overview of this literature in Freixas and Rochet 2008). A different strand of the theoretical literature suggests that banks with higher capital may experience lower survival odds. Calomiris and Kahn (1991) show that a capital structure with sufficiently high demand deposits (and by implication lower equity) leads to more effective monitoring of bank managers by informed depositors and hence a smaller likelihood of bad investment decisions. This suggests that a bank with higher capital (and consequently lower deposits) may face a higher probability of bad loans and hence loan default, which may result in a lower survival probability. This paper has spawned a sizeable literature on the market discipline role of bank leverage (see Freixas and Rochet (2008) for an overview). 7 Thus, some theories predict that higher bank capital should lead to a higher survival probability for the bank, whereas others suggests that higher capital may worsen the portfolio choices and liquidity of banks and hence lead to a lower survival likelihood.8 These papers do not focus on financial crises per se. However, since a banks likelihood of survival is lower during a crisis, it follows that the effect of capital on survival may be more positive during a crisis, particularly in light of regulatory discretion in closing banks, forcing them into assisted mergers, or otherwise

In Mehran and Thakor (forthcoming), the mechanical effect of capital also shows up higher-capital banks are more likely to survive not just because they monitor more, but also because they have a greater capital cushion to absorb losses and avoid interim closure. 7 For example, Diamond and Rajan (2001) argue that replacing demandable deposits with capital weakens the banks incentive to collect repayments from borrowers, thereby reducing loan liquidity. In fact, in an all-equity bank, with no demand deposits to discipline the bank the loan would be completely illiquid because the bank cannot credibly precommit to collect repayment on behalf of others, so the loan could not be sold to another bank, and the bank could additionally credibly threaten to withhold its repayment collection technology. 8 The empirical literature on bank failure finds that capital reduces the probability of failure (e.g., Wheelock and Wilson 2000), but this literature does not focus on crises.

resolving problem institutions based on their capital ratios.9

2.2. Market share Hypothesis 2a: Capital enhances the banks market share during crises and normal times. Hypothesis 2b: Capital diminishes the banks market share during crises and normal times.

The theories on the effect of capital on market share also produce opposing predictions. In the banking models of Holmstrom and Tirole (1997), Allen and Gale (2004), Boot and Marinc (2008), Allen, Carletti, and Marquez (forthcoming), and Mehran and Thakor (forthcoming), banks derive a competitive advantage from higher capital. These papers imply that higher-capital banks will end up with higher market shares. In contrast, there is also a literature that focuses on the relationship between leverage and market share for nonfinancial firms (e.g., Brander and Lewis 1986, Lyandres 2006). This literature shows that more highlylevered firms will be more aggressive in their product-market-expansion strategies and hence suggests that capital and market share will be negatively correlated. The literature discussed above does not focus on crises, so the predictions of these papers should be viewed as applying during normal times. However, it is plausible to argue that the competitive advantage of capital is more pronounced during crises, particularly during banking crises. There are several reasons for this. First, the banks customers may be more cognizant of the banks capital during a crisis, making it easier for better-capitalized banks to take customers away from lesser-capitalized peers. Second, banks with more capital may have greater flexibility to make certain types of loans, offer credit lines and otherwise create liquidity in various ways that may be unavailable to lower-capital banks that may feel particularly constrained, possibly by regulators, during crises. Third, banking crises are generally associated with numerous bank failures and near failures. Failing and near-failing banks tend to be bought by competitors and such

acquisitions have to be approved by bank regulators. Since regulatory approval depends in part on the acquiring banks capital, banks with higher capital ratios are better positioned to buy their troubled brethren, and hence improve their market share.

The benefits of higher capital may be weaker if regulators delay closure of troubled banks because of regulatory career concerns (Boot and Thakor 1993) or because regulators perceive a high value associated with avoiding bankruptcy as, for example, happened during the recent subprime lending crisis (Veronesi and Zingales 2010).

2.3. Profitability Hypothesis 3a: Capital enhances the profitability of the bank during crises and normal times. Hypothesis 3b: Capital diminishes the profitability of the bank during crises and normal times.

The theoretical literature also offers conflicting predictions about how capital should affect bank profitability. One strand of theory predicts that higher capital enhances profitability. As pointed out above, Holmstrom and Tirole (1997), Allen, Carletti, and Marquez (forthcoming), and Mehran and Thakor (forthcoming) show that high bank capital increases the total surplus generated in the bank-borrower relationship. Assuming that banks keep a large enough portion of the surplus, higher capital will lead to higher bank profitability. Moreover, if the ratio of the surplus generated by high- versus low-capital banks is higher during crises, it follows that highcapital banks will be able to improve their profitability during crises relative to low-capital banks. 10 In contrast, another strand of theory predicts that higher capital should lead to reduced profitability. The most obvious argument here goes back to Modigliani and Miller (1963), who show that higher capital mechanically leads to lower ROE. Moreover, the literature on the disciplining role of debt (e.g., Calomiris and Kahn 1991) suggests that higher leverage improves the banks asset choice and hence its profitability. Thus, banks with higher capital have lower-quality assets. If these assets deteriorate in value more during a crisis than do higher-quality assets, then banks with higher capital may suffer bigger declines in profitability during crises than their lower-capital counterparts.

2.4. Empirical implications of the theories It is clear from the discussions above that the theories have opposing views on the effect of capital on performance. Our empirical analyses attempt to assess the impact of capital on performance to determine which theories have the strongest empirical support. Not finding support for a particular theory does not constitute a rejection of that theory, but simply implies that the theory for which we do find support is more
A recent survey paper by Campello, Giambona, Graham, and Harvey (2009) sheds further light on why banks with high capital may improve their profitability during crises relative to banks with low capital. It shows that during the subprime lending crisis, banks renegotiated in their own favor the terms for lines of credit with borrowers, possibly by threatening to invoke the material-adverse-change clause. It may be that banks with more capital could do this more easily because their stronger reputation in the loan commitment market gives them greater bargaining power with their borrowers (see Boot, Greenbaum, and Thakor 1993), which would result in increased profitability for high-capital banks (relative to lowcapital banks) during crises. Berger (1995) finds that the relationship between capital and earnings can be positive or negative, but does not differentiate between financial crises and normal times.
10

dominant empirically. This is because what we may be picking up in the data is the net effect of opposing forces identified by different theories.

3. Methodology This section first explains our empirical approach and describes the financial crises and normal times. It then explains the performance measures. Next, it discusses the key exogenous variables and the control variables. Finally, it describes the sample and provides summary statistics. In our empirical approach, we examine the effect of capital (and other bank conditions) measured prior to a crisis on bank performance during a crisis. We measure capital before a crisis for two reasons. First, since it is not known a priori when a crisis will strike, the interesting question is whether banks that have higher capital going into a crisis benefit from these higher capital ratios during a crisis. That is, we want to know whether higher pre-crisis capital ratios result in better performance during a crisis. Second, it mitigates endogeneity concerns because lagged capital and current performance are less likely to be jointly determined.

3.1. Empirical approach and description of financial crises and normal times Our analyses focus on crises that occurred between 1984:Q1 and 2009:Q4. They include two banking crises (crises that originated in the banking sector) and three market crises (crises that originated outside banking in the financial markets). The banking crises are the credit crunch of the early 1990s (1990:Q1 1992:Q4) and the recent subprime lending crisis (2007:Q3 2009:Q4). The market crises are the 1987 stock market crash (1987:Q4); the Russian debt crisis plus Long-Term Capital Management (LTCM) bailout of 1998 (1998:Q3 1998:Q4); and the bursting of the dot.com bubble and the September 11 terrorist attacks of the early 2000s (2000:Q2 2002:Q3). Appendix I describes these crises in detail. Our hypotheses focus on the effect of a banks (pre-crisis) capital on its performance (survival, market share, and profitability) during a crisis. A key issue is how to measure pre-crisis capital. In all of our main analyses, we average each banks capital ratio over the eight quarters before the crisis to reduce the impact of outliers. In robustness checks, we alternatively define the pre-crisis period as the four quarters before a crisis or the quarter before a crisis (see Section 5.5). Our survival analyses then link this average pre-crisis capital to whether a bank survived a crisis (it was in the sample one quarter before the crisis and is still in the sample one

quarter after the crisis; see Section 3.2). Our market share analyses link pre-crisis capital to the banks percentage change in market share (see Section 3.3), defined as the banks average market share during a crisis minus its average share over the eight quarters before the crisis, normalized by its average pre-crisis market share and multiplied by 100. Similarly, our profitability analyses link pre-crisis capital to the banks change in profitability (see Section 3.4), defined as the banks average ROE during a crisis minus its ROE over the eight quarters before the crisis. We expect the effect of capital to be more positive during financial crises than during normal times. While we highlight above how we examine the effect of (average) pre-crisis capital on bank performance during a crisis, we still have to address how we measure normal times. A nave approach would be to simply view all non-crisis quarters as such. However, if so, it is not clear then how to examine the effect of capital on bank performance during normal times. To ensure that we analyze actual crises and normal times in a comparable way, we create fake crises to represent normal times. In essence, these fake crises act as a control group. To construct these fake crises, we use the two longest time periods between actual financial crises over our entire sample period. These periods are between the credit crunch and the Russian debt crisis, and between the bursting of the dot.com bubble and the subprime lending crisis. In each case, we take the entire period between the crises, designate the first eight quarters as pre-crisis and the last eight quarters as post-crisis and the remaining quarters in the middle as the fake crisis. This treatment of the first eight quarters as pre-crisis is consistent with our analysis of the banking and market crises. We thus end up with a six-quarter fake crisis period between the credit crunch and the Russian debt crisis (from 1995:Q1 to 1996:Q2) and a three-quarter fake crisis period between the dot.com bubble and the subprime lending crisis (from 2004:Q4 to 2005:Q2). 11 To analyze how capital affects banks ability to survive crisis and normal times, we pool the data of the banking crises, the market crises, and the normal times. 12 Since we have two banking crises, three market crises, and two normal time periods (the fake crises), we have up to seven observations per bank, i.e., . Using these data, we run the following logit regressions:

Results are qualitatively similar if we instead use five- or four-quarter crisis periods for the first fake crisis, and twoor one-quarter crisis periods for the second fake crisis. 12 Our regressions group the two banking crises together and the three market crises together. This approach allows us to contrast the effect of capital during banking and market crises. To check the robustness of our findings, we also run our main regressions using individual crisis dummies instead, and obtain broadly consistent results (not shown for brevity).

11

(1) where measures whether bank i survived crisis or normal time period t (see Section 3.2).

is the banks average capital ratio over the eight quarters before crisis or normal time period t (see Section 3.5). , , and are dummy variables that equal 1 if t is a is a set of control

banking crisis, market crisis, or normal time period, respectively, and 0 otherwise. variables measured over the pre-crisis period (see Section 3.6).

To examine the impact of capital on a banks market share and profitability during banking crises, market crises, and normal times, we use the following regression specifications:

(2)

(3) where is the percentage change in bank is aggregate market share (see Section 3.3) and is the change in bank is profitability (see Section 3.4). To mitigate the influence of outliers, both variables are winsorized at the 3% level.13 variables over the pre-crisis period (see Section 3.6). Since each bank enters up to seven times in these regressions, all regressions are estimated with robust standard errors, clustered by bank, to control for heteroskedasticity as well as possible correlation between observations of the same bank in different years. The regressions also include individual crisis and normal times dummies which act as time fixed effects. The literature documents differences by bank size in terms of portfolio composition (e.g., Kashyap, Rajan, and Stein 2002, Berger, Miller, Petersen, Rajan, and Stein 2005) and the effect of capital on liquidity creation which we use to construct our main market share variable (Berger and Bouwman 2009). We therefore
Unwinsorized data yielded changes in market share (profitability) that were roughly 54,600 (6,600) times the average change. Winsorization at the 1% level reduced this to 6.3 (23.6) times the average change, which still seemed too high for the change in profitability. Winsorization at the 3% level brought these numbers down to 2.0 and 7.2, respectively. Winsorization at the 1% level yields market share and profitability results that are very similar to the ones presented here, except that the market share results for medium banks and the profitability results for small banks are both weaker during banking crises.
13

is as defined above.

is a set of control

10

split the sample into small banks (gross total assets (GTA) up to $1 billion), medium banks (GTA exceeding $1 billion and up to $3 billion), and large banks (GTA exceeding $3 billion) and run all regressions separately for these three sets of banks. All dollar values are in 2009:Q4 terms. Our definition of small banks conforms to the usual notion of community banks. The $3 billion cutoff for GTA divides the remaining observations roughly in half. 14

3.2. Definition of survival To measure whether a bank survived a crisis, we require that the bank was not acquired and did not fail during the crisis. Specifically, we use SURV, a dummy that equals 1 if the bank is in the sample one quarter before such a crisis started and is still in the sample one quarter after the crisis, and 0 otherwise. 15,16

3.3. Definition of market share While most of the literature focuses on the role of capital in traditional banks that engage only in onbalance-sheet activities, several papers highlight the importance of banks off-balance sheet activities (Boot, Greenbaum, and Thakor 1993, Holmstrom and Tirole 1997, Kashyap, Rajan, and Stein 2002). We therefore measure a banks competitive position as the banks market share of overall bank liquidity creation. Liquidity creation is a superior measure of bank output since it is the only measure based on all the banks on- and offbalance sheet activities. 17 We calculate the dollar amount of liquidity created by each bank using Berger and Bouwmans (2009) preferred liquidity creation measure. The three-step procedure used to construct this measure is explained in Appendix II. A banks liquidity creation market share is the dollar amount of liquidity creation by the bank divided by the dollar amount of liquidity created by the industry. 18 To establish whether banks improve their competitive positions during banking crises, market crises,
Berger and Bouwman (2009) also use these cutoffs. As shown in Section 5.4, broadly similar results are obtained when $5 billion and $10 billion cutoffs between medium and large banks are used instead. 15 Banks that were merged within a bank holding company are not classified as non-survivors since it unclear whether these consolidations occur because the bank is troubled or not. 16 Robustness checks using a slightly longer time window yield similar results (see Section 5.1). 17 As a robustness check, we rerun our main regressions using the market share of gross total assets, and obtain similar results (see Section 5.1). 18 Market shares are not merger adjusted because acquisitions are a key way for banks to increase their market shares, particularly during banking crises. In our analysis, a banks market share rises if it acquires another bank. Mergeradjusting market shares would take out this effect.
14

11

and normal times, we define each banks percentage change in liquidity creation market share, %LCSHARE, as the banks average market share during a crisis minus its average market share over the eight quarters before the crisis, normalized by its average pre-crisis market share and multiplied by 100.

3.4. Definition of profitability We measure a banks profitability using the banks return on equity (ROE), i.e. net income divided by stockholders equity. 19 This is a comprehensive profitability measure, since banks may have substantial offbalance sheet portfolios. Banks must allocate capital against every off-balance sheet activity in which they engage. Hence, net income and equity both reflect the banks on- and off-balance sheet activities.20 To examine whether a bank improves its profitability during banking crises, market crises, and normal times, we focus on the change in ROE (ROE), defined as the banks average profitability during these crises minus the banks average ROE over the eight quarters before these crises.

3.5. Key exogenous variables The key exogenous variables are three bank capital ratio-crisis interaction terms: EQRAT * BNKCRIS, EQRAT * MKTCRIS, and EQRAT * NORMALTIME. EQRAT is the ratio of equity capital to gross total assets, averaged over the eight quarters before the crisis. 21 BNKCRIS, MKTCRIS, and NORMALTIME are as defined in Section 3.1.

3.6. Control variables The survival regressions contain X, a set of control variables which include: bank credit risk, bank size, bank holding company membership, local market power, and profitability. The market share and profitability regressions contain a slightly smaller set of controls Z, which excludes profitability from X. We discuss these variables in turn. Each control variable is averaged over the eight-quarter pre-crisis period, except when noted
If a banks capital to GTA ratio is less than one percent, we calculate ROE as net income divided by one percent of GTA. For observations for which equity is between 0% and 1% of GTA, dividing by equity would result in extraordinarily high values. For observations for which equity is negative, the conventionally-defined ROE would not make economic sense. We considered the alternative of dropping negative-equity observations, but rejected it because these are the banks that are most likely to be informative of banks ability to survive crises. 20 Section 5.1 includes a robustness check in which we use return on assets (ROA) instead. 21 Similar results are obtained in robustness checks which use a banks regulatory capital ratio (see Section 5.2) or measure capital at alternative points before the crisis starts (see Section 5.5).
19

12

otherwise. Credit risk, defined as the banks Basel I risk-weighted assets divided by gross total assets, is used as a measure of bank risk taking (e.g., Logan 2001). Risk-weighted assets, a weighted average of the banks assets and off-balance-sheet activities designed to measure credit risk, is the denominator in the Basel I risk-based capital requirements. Since these requirements only became effective in December 1990 and, hence, were only reported in Call Reports from that moment onward, we use a Federal Reserve Board program to construct risk-weighted assets from the beginning of our sample period. Banks with riskier portfolios (i.e. higher riskweighted assets relative to gross total assets) may be less likely to survive crises. They may also find it harder to improve their market shares and profitability during crises. We therefore interact the credit risk variable with the three crisis dummies, and expect the coefficients on the banking and market crisis interaction terms to be negative in all regressions. Bank size is controlled for by including lnLC, the log of liquidity creation, in all regressions. 22 In addition, we run regressions separately for small, medium, and large banks. Bank size is expected to have a positive effect on the probability of survival, since it is well-known that larger banks have higher survival odds than smaller banks. In contrast, the coefficient on bank size is expected to be negative and significant for all size classes in the market share regressions, since the law of diminishing marginal returns suggests that it will be more difficult for bigger banks (that already have larger market shares) to improve their market shares. It is unclear whether bank size will have a significant effect on banks ability to improve their profitability (measured as a banks ROE) during crises. To control for bank holding company status, we include D-BHC, a dummy variable that equals 1 if the bank was part of a bank holding company (BHC) at any time in the eight quarters preceding the crisis. BHC membership is expected to help a bank survive and strengthen its competitive position because the holding company is required to act as a source of strength to all the banks it owns, and may also inject equity voluntarily when needed. In addition, other banks in the holding company provide cross-guarantees. Houston, James, and Marcus (1997) find that bank loan growth depends on BHC membership. Thus, the coefficient on bank holding company status is expected to be positive and significant in the survival and market-share regressions. BHC membership is also expected to reduce the cost of financing and enhance profitability
Robustness checks which instead include lnGTA, the log of gross total assets, yield similar results. shown for brevity.
22

These are not

13

because it provides greater access to internal capital markets and provides potential protection against the vagaries of market financing. We control for local market power by including HHI, the bank-level Herfindahl-Hirschman index of deposit concentration for the local markets in which the bank is present.23 From 1984-2004, we define the local market as the Metropolitan Statistical Area (MSA) or non-MSA county in which the offices are located. 24 After 2004, we use the new local market definitions based on Core Based Statistical Area (CBSA) and nonCBSA county. 25 The larger is HHI, the greater is a banks market power. Since more market power should help a bank survive, the coefficient on HHI is expected to be positive in the survival regressions for banks of all sizes. More market power likely makes it harder to improve market share since regulatory approval for acquisitions will be more difficult to obtain. The coefficient on HHI is therefore likely to be negative in the market share regressions. Since higher market power increases the profitability of local loans and deposits, more market power should make it easier to improve profitability. Thus, the coefficient on HHI in the profitability regressions is expected to be positive. The survival regressions also include a measure of profitability because banks that are more profitable before the crisis may be more likely to survive crises. We use a banks return on assets, ROA, for this purpose. The reason to use ROA instead of ROE is that we want to measure the effect of capital on banks ability to survive crises and avoid using control variables that include capital.

3.7. Sample and summary statistics For every bank in the U.S., we obtain quarterly Call Report data from 1984:Q1 to 2009:Q4. We keep a bank in the sample if it: 1) has commercial real estate or commercial and industrial loans outstanding; 2) has deposits; and 3) has gross total assets or GTA exceeding $25 million. As indicated above, we split these banks into three size categories. Analyses that focus on the effect
While our focus is on the change in banks competitive positions measured in terms of their aggregate liquidity creation market shares, we control for local market competition measured as the bank-level Herfindahl-Hirschman index based on local market deposit shares. This is a standard measure of competition used in antitrust analysis in the U.S. Deposits are used for this purpose because deposits are the only bank output variable for which location is known. 24 When appropriate, we use New England County Metropolitan Areas (NECMAs) instead of MSAs, but refer to these as MSAs. 25 The term CBSA collectively refers to Metropolitan Statistical Areas and newly-created Micropolitan Statistical Areas. Areas based on these new standards were announced in June 2003. For recent years, the Summary of Deposits data needed to construct HHI is available on the FDICs website only based on the new definition. It is not possible to use the new definition for our entire sample period.
23

14

of capital on survival have 46,107 small-bank, 1,599 medium-bank, and 1,194 large-bank observations. Analyses that focus on the effect of capital on market share and profitability have 52,107 small-bank, 1,911 medium-bank, and 1,382 large-bank observations. 26 Table 1 contains summary statistics on the regression variables. The sample statistics are shown for banking crises, market crises, and normal times. All financial values are put into real 2009:Q4 dollars (using the implicit GDP price deflator) before size classes are constructed.

4. Main regression results In this section, we discuss the main empirical results.

4.1. Does capital affect the banks ability to survive during crises and normal times? Table 2 Panel A presents the survival findings for small, medium, and large banks. Two main results are apparent. First, higher capital helps banks in all size classes to improve the probability of surviving banking crises. Second, higher capital also helps small banks improve their odds of survival during market crises and normal times. These results generally support the hypothesis that capital helps banks survive. These findings have sensible economic interpretations. Capital is the main line of defense against negative shocks for small banks, since they have limited (and relatively costly) access to the financial market in the event of unanticipated needs. Hence, higher capital enhances the probability of survival for such banks at all times. This finding is consistent with the earlier-discussed theoretical papers that predict that higher capital increases a banks survival probability. Medium and large banks can rely on financial market access, and correspondent and other interbank relationships as risk-mitigation sources in addition to their on-balancesheet capital to survive negative shocks. So, capital is less pivotal for survival during normal times for these banks than it is for small banks. However, banking crises create stresses for all banks, and financial market access and interbank relationships may offer inadequate protection against negative shocks for all but the very largest banks. Hence, capital may be important for survival for medium and large banks during banking crises. The market crisis results for medium and large banks suggest that such crises do not pose much of a survival threat for these banks. Given the access that medium and large banks have to the interbank lending market
The survival regressions have fewer observations because we do not have data on the first quarter after the end of the subprime lending crisis and because we drop within-bank-holding-company consolidations from these regressions.
26

15

(e.g., federal funds) that itself is unlikely to be adversely impacted by a market crisis, capital may not be critical for these banks to survive market crises. To judge the economic significance of our findings, Table 2 Panels B, C, and D show the predicted probabilities of non-survival. The top part of each panel shows the average capital ratio and the average

capital ratio plus or minus one standard deviation of small, medium, and large banks over the eight quarters before banking crises, market crises, and normal times. The bottom part of each panel shows the predicted probability of not surviving banking crises, market crises, and normal times at these capital ratios. A small, medium, or large bank with an average capital ratio (10.00%, 8.76%, and 8.30%, respectively) had a probability of not surviving banking crises of 0.61%, 0.55%, and 0.12%, respectively. Reducing capital by one standard deviation more than doubles these probabilities of non-survival. Similarly, increasing capital by one standard deviation reduces the probabilities of non-survival by more than one half for all size classes. The corresponding probabilities for market crises and normal times are generally higher. 27 Turning to the control variables, we find that the coefficients generally have the predicted signs. As expected, being part of a bank holding company, greater market power, and higher profitability helps banks survive. Banks that held more credit risk before banking and market crises are less likely to survive

(significant for small banks; for medium banks only before banking crises). Bank size has no significant effect on medium and large banks ability to survive. Among small banks, larger banks are less likely to survive, which is surprising.

4.2. Does capital affect the market shares of banks during crises and normal times? Table 3 summarizes the results from regressing the percentage change in a banks market share (%LCSHARE) during crises on the banks pre-crisis capital ratio plus control variables. As before, results are shown separately for small, medium, and large banks. The t-statistics are based on robust standard errors clustered by bank. We find two main results. First, higher capital helps banks of all sizes improve their market shares during banking crises and normal times. Second, higher capital also helps small banks to improve their market

The survival probabilities of medium banks actually decrease slightly when capital is higher during market crises and normal times. This is consistent with the negative but insignificant coefficients on EQRAT for these banks shown in Table 2 Panel A.

27

16

shares during market crises. 28 Capital does not appear to produce such benefits during market crises for medium and large banks. Somewhat surprisingly, medium banks with higher capital ratios seem to lose market share during market crises. These results generally support the hypothesis that capital helps banks to improve their market shares. We can again interpret these results in the context of the theories discussed earlier. Note that capital is essential to survival for small banks, as discussed earlier. Moreover, these banks engage largely in

relationship lending, and long-lasting bank-borrower relationships are crucial for relationship banking to create value. 29 This means that relationship borrowers will tend to gravitate toward high-capital banks, since higher capital leads to a higher survival probability for small banks at all times (see Section 4.1). 30 Thus, it is not surprising that higher capital benefits small banks in terms of gaining market share at all times. During banking crises and normal times, capital also helps improve the market shares of medium and large banks. In contrast, during market crises, capital seems to hurt the market shares of medium banks (no obvious interpretation) and does not significantly affect large banks, possibly because it is relatively easy for these banks to turn to the discount window and the interbank market both of which may not experience stress during market crises to ensure unruptured relationships with their borrowers during such crises. This means that while capital may offer large banks a benefit during market crises, this benefit may be no greater than that before such crises. To judge the economic significance of these results, focus first on the effect of capital during banking crises. The coefficients on the EQRAT * BNKCRIS interaction terms imply that a one standard deviation increase in EQRAT would lead to a 0.373, 0.114, and 0.216 standard deviation change in for

small, medium, and large banks, respectively, during such crises (not shown for brevity). The corresponding figures for EQRAT * MKTCRIS and EQRAT * NORMALTIME are (0.310, -0.152, and -0.040) and (0.231, 0.185, and 0.250), respectively. These results seem economically significant for banks of all sizes during banking crises and for small banks also during market crises and normal times.
The positive effect of capital on large banks market shares during normal times disappears when we use a $5 billion or $10 billion cutoff between medium and large banks, suggesting that our main result is driven by smaller large banks (see Section 5.4). 29 See, e.g., Ongena and Smith (2001) Bae, Kang, and Lim (2002), and Bharath, Dahiya, Saunders, and Srinivasan (2007) for empirical evidence. 30 This argument is consistent with the intuition in Song and Thakor (2007) who show that banks that make relationship loans will prefer to finance with stable funds because this increases the likelihood that a relationship loan will not need to be terminated prematurely.
28

17

Turning to the control variables, we find that banks with higher pre-crisis credit risk are generally less likely to improve their market shares during crises. The reason may be similar to the argument given above for capital: safer banks create more surplus and therefore are more likely to increase their market shares. Among banks of all size classes, the coefficient on bank size is negative and significant, likely because it is harder for larger banks to increase their percentage market share. Being part of a bank holding company helps to improve market share (significant for small and medium banks). Among small banks, the ones with greater market power increase their market shares less, probably because these banks tend to operate locally and regulators do not approve mergers that increase their market shares significantly. This does not hold for medium and large banks, since they operate more on a national or international basis.

4.3. Does capital affect bank profitability during crises and normal times? Table 4 contains the results of regressing the change in profitability (ROE) during crises on the banks precrisis capital ratio plus control variables. The setup of the table is similar to the previous one. As before, tstatistics are based on robust standard errors clustered by bank. We again find two main results. First, high-capital banks of all sizes improve their profitability during banking crises (not significant for medium banks) and market crises. Second, capital also enhances the profitability of small banks during normal times. These results generally support the hypothesis that capital improves bank profitability. The reason for these results can be found in our earlier discussion of the central importance of capital for small banks at all times. Because small banks with higher capital have higher survival probabilities and can offer higher-valued relationships, it is intuitive that higher-capital small banks improve their profitability at all times. It may seem surprising, however, that capital helps to improve profitability for medium and large banks during market crises, since higher capital improves neither the market share nor the survival probability for these banks during such crises. One possible explanation is that these banks continue to have access to

uninsured sources of funding during market crises, but there may be a flight to quality phenomenon, whereby the cost of such funds becomes more sensitive to the amount of capital the bank has. To judge the economic significance of these results, focus first on the effect of capital during banking crises. The coefficients on the EQRAT * BNKCRIS interaction terms imply that a one standard deviation

18

increase in EQRAT would lead to a 0.041, 0.061, and 0.125 standard deviation change in

for small,

medium, and large banks, respectively, during such crises (not shown for brevity). The corresponding figures for EQRAT * MKTCRIS and EQRAT * NORMALTIME are (0.076, 0.078, and 0.184) and (0.085, 0.026, and 0.011), respectively. These results seem less economically significant than the survival and market share results. The control variables generally have the expected signs. Banks that operate with higher credit risk before banking and market crises find it harder to improve their profitability during such crises. Bank size has no significant impact on banks ability to improve their profitability. Bank holding company status increases profitability for small banks only. Higher HHI increases profitability not just for small banks but also for medium banks.

5. Robustness checks This section presents a number of checks to establish the robustness of our results. First, we use alternative specifications of survival, market share, and profitability. Second, we use regulatory capital ratios instead of the equity-to-assets ratio. Third, we drop Too-Big-To-Fail banks from the large-bank sample. Fourth, we use alternative cutoffs separating medium and large banks. Fifth, we measure pre-crisis capital ratios alternatively one quarter before the crisis starts or averaged over the four quarters before the crisis. Sixth, we deal with the potential endogeneity issues related to capital by using an instrumental variable approach. Our survival and market share findings are generally qualitatively similar to the main results. The profitability results are less robust. To preserve space, we only present the coefficients on the variables of interest, the pre-crisis capital ratio interacted with the crisis dummies. The same control variables as used in the main regression

specifications are included with some exceptions noted below.

5.1. Use alternative specifications of survival, market share, and profitability Our main survival results are based on SURV, survival until one quarter after a crisis. As a robustness check, we now use a slightly longer time window. Specifically, we use SURV_alt, a dummy that equals 1 if the bank is still in the sample four quarters after the crisis.

19

Table 5 Panel A1 shows the results. As can be seen, the coefficients and significance levels are similar to those shown in Table 2, except that based on this alternative definition, the effect of capital on large banks ability to survive banking crises is no longer significant. Thus, our survival results do not seem to be driven appreciably by the choice of time window. Our main competitive position results are based on %MKTSHARE, a banks liquidity creation market share. As a robustness check, we now use %MKTSHARE_alt, a banks market share of aggregate gross total assets (GTA). GTA is a traditional measure of size that focuses on the banks on-balance sheet activities. GTA market share is calculated as the banks gross total assets divided by the industrys gross total assets. The main shortcoming of this measure is that it ignores off-balance sheet activities and treats all assets identically, i.e., it neglects the qualitative asset transformation nature of the banks activities (e.g., Bhattacharya and Thakor 1993, Kashyap, Rajan, and Stein 2002). For consistency, we also use lnGTA, the log of GTA, instead of lnLC as a size control in these regressions. Table 5 Panel A2 shows that based on GTA market share, small banks are able to improve their market shares at all times, while medium and large banks improve their market shares only during banking crises. The difficult-to-explain finding that medium banks with higher capital ratios lose market share during market crises is still present, but the effect of capital on market share during normal times for medium and large banks has disappeared. Thus, the results confirm the small-bank results and the banking crisis (not market crisis) results for medium and large banks shown in Table 3. Our main profitability results focus on PROF, the banks change in ROE. As a robustness check, we now use PROF_alt, the change in ROA. This measure divides net income, generated based on all the banks on- and off-balance sheet activities, by GTA, a measure of the banks on-balance sheet activities. The change in ROA is meaningful for small banks since most of their activities are on the balance sheet. It is less appropriate for medium and large banks, since these banks engage in more off-balance sheet activities. Table 5 Panel A3 shows the results. As before, capital helps small banks improve their profitability at all times. Not surprisingly, the medium and large bank results are generally not significant. Thus, the results confirm the main profitability results shown in Table 4 for small banks, and are understandably weaker for medium and large banks.

20

5.2. Use regulatory capital ratios The main results highlight the benefits of higher capital ratios. One may wonder to what extent these results are specific to our choice of the definition of capital, EQRAT, the ratio of equity capital to assets. To examine this issue, we rerun our regressions using regulatory capital ratios. Basel I introduced two risk-based capital ratios, the Tier 1 and Total risk-based capital ratios. Since these ratios only became fully effective as of December 1990, we use each banks ratio of equity capital to GTA before this date and its Tier 1 risk-based capital ratio from that moment onward. We obtain very similar results when we use the Total risk-based capital ratio instead of the Tier 1 capital ratio. Prior to 1996, all banks with assets over $1 billion had to report this information, but small banks only had to report their riskbased capital positions if they believed that their capital was less than eight percent of adjusted total assets. In all these cases, we use a Federal Reserve Bank program to reconstruct what those banks risk-based capital ratios are based on (publicly-available) Call Report data. Table 5 Panel B shows the results. 31 Clearly, for small and large banks, we obtain significance in exactly the same cases as before (see Tables 2 4). For medium banks, the survival and profitability results are similar, but the market share results are less significant (capital does not help these banks improve their market shares during banking crises). We also ran all the robustness checks presented in the rest of this section using regulatory capital and obtain results that are very similar to the main survival, market share, and profitability results. Nonetheless, we believe that our choice of using EQRAT in the main regressions is appropriate. The regulatory ratios mix capital with credit risk which is already controlled for in our main regressions. We prefer to focus on the effect of capital per se.

5.3. Exclude Too-Big-To-Fail banks Very large banks are often considered Too-Big-To-Fail (TBTF). For these banks, capital may not provide the benefits we highlight in this paper since such banks know they will be bailed out and supported if needed. To make sure that our large-bank results are not overly influenced by TBTF banks, we rerun our main analyses
While most of the control variables used before should clearly be included in these regressions, the risk interaction terms may not belong. The reason is that the numerator of the risk variable (Basel I risk-weighted assets) is identical to the denominator of the regulatory capital ratio. We present results here excluding the risk interaction terms, but results are comparable if we instead include them (not shown for brevity).
31

21

for large banks while excluding the TBTF banks. Since no official definition of TBTF exists, we use two alternative definitions. First, in every quarter, we deem all banks with GTA exceeding $50 billion to be TBTF. Second, we classify the 19 largest banks in each period as TBTF. This definition is inspired by the

governments disclosure in April 2009 that the 19 largest banks had to undergo stress tests, and would be assisted with capital injections if they could not raise capital on their own. This effectively made them TBTF. Table 5 Panels C1 and C2 contain the results for large banks excluding the TBTF banks based on the two alternative definitions. The coefficients are bigger than those presented in the main tests in all but one case and significance is found in similar cases. Both sets of results suggest that the presence of TBTF banks weakened our main results to some extent, but leave our main conclusion unchanged.

5.4. Use alternative size cutoffs In our main analyses, small banks have GTA up to $1 billion; medium banks have GTA between $1 and $3 billion; large banks have GTA exceeding $3 billion. As explained above, our small-bank definition captures community banks, while the remaining observations were split roughly in half by choosing a $3 billion cutoff (see also Berger and Bouwman 2009). One may wonder to what extent the medium and large bank results are driven by our choice of the $3 billion cutoff. We reran our analyses using $5 billion and $10 billion cutoffs between medium and large banks, which reclassify some large banks as medium banks. Table 5 Panels D1 and D2 show the results for medium and large banks using these alternative cutoffs. The survival results indicate that, as before, capital helps medium and large banks survive banking crises only. As before, capital helps medium and large banks expand their market shares during banking crises, but now, the effect is also significant for large banks during market crises (based on a $10 billion cutoff only). The reclassification of some large banks as medium banks makes the effect of capital on market share stronger for medium banks and not significant for large banks during normal times. Finally, while our main results suggest that capital helps large banks improve their profitability during banking crises, the reclassification of some large banks as medium banks causes this effect to now only be significant for medium banks. This suggests that the positive effect of capital on profitability occurs mainly for banks that are smaller than $5 billion.

22

5.5. Measure capital ratios at alternative times before the crisis Our main results are based on regression specifications which include every banks capital ratio averaged over the eight quarters before a crisis. The advantage of using eight-quarter averages is that such averages are not very sensitive to the effects of outliers. As robustness checks, we rerun our regressions while measuring bank capital averaged over the four quarters before a crisis or measuring it the quarter before a crisis. Table 5 Panels E1 and E2 contain the results. Clearly, based on these two alternative capital measures, the survival and market share results are similar to the main results. The profitability results, however, are somewhat weaker. When capital is averaged over the four quarters before the crisis, capital helps to improve small banks profitability during banking crises, but the effect is no longer significant (t-statistic 1.52). When capital is measured at the end of the quarter before the crisis, high-capital small banks are not able to improve their profitability relative to low-capital banks at any time, and high-capital large banks do not improve their profitability during banking crises.32

5.6. Use an instrumental variable analysis The analyses presented so far suggest that capital helps banks of all sizes during banking crises, and improves the performance of small banks during market crises and normal times as well. However, a potential

endogeneity issue clouds the interpretation of our results. The theories suggest a causal link from capital to performance. But we know that in practice, capital is an endogenous choice variable for a bank, so the banks market share, profitability, and even its likelihood of survival may impact the banks capital choice. This issue is addressed partly in our analysis because capital is lagged relative to the dependent variables. However, this may not be enough because of intertemporal rigidities in some of these variables. It may therefore be difficult to know if we have detected causality or mere correlation. instrumental variable approach. We have three potentially endogenous variables, capital interacted with the three crisis dummies (EQRAT * BNKCRIS, EQRAT * MKTCRIS, and EQRAT * NORMALTIME), so we need three To address this concern, we now use an

The reason for the weaker profitability results, especially for the small banks, may be that the beneficial effects of higher capital that are manifested through greater monitoring may be reflected in the banks profitability only with a significant lag. When pre-crisis capital is measured relatively close to the crisis, we may not fully capture the effect of higher capital.

32

23

instruments. 33 We employ different instruments for banks of different sizes for reasons explained below. Specifically, we use SENIORS, the fraction of seniors (people aged 65 and over) in the market in which a bank is active, interacted with the three crisis dummies as instruments for medium and small banks; and EFFTAX, the effective state income tax rate a bank has to pay, interacted with the three crisis dummies as instruments for large banks. This instrumentation strategy assumes that SENIORS and EFF-TAX are correlated with the amount of capital, but do not directly affect performance. Both variables, used by Berger and Bouwman (2009) as instruments for capital, seem to meet these conditions. To see why, consider SENIORS first. Seniors own larger equity portfolios than the average family. Furthermore, using U.S. data, Coval and Moskowitz (1999) document that investors have a strong preference for investing close to home. They find that this preference is greater for firms that are smaller, more highly levered, and those that produce goods that are not traded internationally. In combination, this evidence suggests that banks particularly small and medium banks that operate in markets with more seniors have easier access to equity financing and hence, will operate with higher capital ratios (see Berger and Bouwman 2009 for empirical evidence). We calculate the fraction of seniors using county- and MSA-level population data from the 1990 and 2000 decennial Census. While SENIORS can be used for small and medium banks, it is not as attractive for large banks because these banks are unlikely to be locally-based when it comes to raising equity. For this reason, we use EFF-TAX for large banks. Since interest on debt is tax-deductible while dividend payments are not, banks that operate in states with higher income tax rates are expected to have lower equity ratios, keeping all else equal (see Ashcraft 2008 and Berger and Bouwman 2009 for empirical evidence). Similar to Ashcraft (2008), we define EFF-TAX as the effective income tax rate to be paid on $1 million in pretax income. If a bank operates in multiple states, we use the banks weighted-average state income tax rate, calculated using the share of deposits in each state (relative to the banks total deposits) as weights. Since ordinary least squares is preferred to instrumental variable regressions if we do not have an endogeneity problem, we perform a Hausman test for endogeneity. As shown in Table 5 Panel F1, we do not reject the null that the three potentially endogenous variables (the EQRAT interaction terms) are exogenous

It is not correct to view EQRAT as the endogenous right-hand-side variable, create a predicted value of EQRAT in the first stage and then interact it with the three crisis dummies. Wooldridge (2002, p. 236) calls this the forbidden regression.

33

24

for medium and large banks, suggesting that our original analyses were appropriate for these banks. However, we do find evidence of endogeneity for small banks and therefore perform instrumental variable regressions only for those banks. Since we have three endogenous variables, we run three first-stage regressions. In each, we regress EQRAT interacted with one of the three crisis dummies on all the exogenous variables used before plus the three instruments (SENIORS interacted with the three crisis dummies). Importantly, when EQRAT *

BNKCRIS is the endogenous variable, the coefficient on the corresponding instrument (SENIORS * BNKCRIS) is positive and highly significant, and we obtain similar results for the other endogenous variables (not shown for brevity). In the second-stage regressions, we regress bank performance on all the exogenous variables and the predicted values from the first stage. Table 5 Panel F2 shows the second-stage instrumental variable

regressions for small banks. Using an instrument for small banks capital, we find that capital helps these banks survive banking crises only, improve their market shares at all times (not significant during normal times: t-statistic 1.59), and improve their profitability during market crises and normal times (surprisingly, a negative effect during banking crises). Thus, most of our earlier results hold up in our instrumental variable analysis, and the analysis broadly confirms our main results.

6. Comparing the effects of book and market capital during crises and normal times The analyses above focus on the effect of book capital on the performance of individual banks. A subset of these banks is independently listed or part of a listed BHC. This allows us to compare the effects of book and market capital on performance for this set of banking organizations (jointly referred to here as listed entities for ease of exposition). This is interesting because regulators focus on book capital ratios, which merely measure the amount of capital at a particular point in time, while market capital ratios are forward-looking and capture market participants beliefs about future performance. 34 We expect the book-capital listed-entity results to be similar to our large-bank results (higher capital

There are well-known difficulties with book equity because many assets, particularly loans, are not marked-to-market. While the loan and lease loss reserves are subtracted from book equity to account for credit losses on loans and leases, these may be imprecise and they leave out gains and losses from interest rate risks on fixed-rate loans. In addition, the book value of equity may be significantly overstated for failing institutions, given that the FDIC usually takes considerable losses when selling off the assets of these institutions.

34

25

helps especially during banking crises), except that the relationship between book capital and listed entity survival is likely weaker since many of the listed entities may be Too-Big-To-Fail. The listed entity results are likely stronger based on market capital (than on book capital) because of their forward-looking nature. For this analysis, we include listed banks and listed one-bank-holding companies, and we aggregate the data of all the banks in a listed multi-bank-holding company. Book capital for the listed entity,

EQRAT_listed, is calculated as before.35 Each listed entitys pre-crisis market capital ratio, MKTRAT_listed, is calculated as the listed entitys market value of equity (i.e., the number of shares outstanding multiplied by the share price) divided by its assets in market value terms (i.e., the book value of liabilities plus the market value of equity), averaged over the eight quarters before the crisis. We rerun regressions (1) (3) using listedentity variables (except that we leave out the BHC dummy because almost all of these entities are BCHs). Table 6 shows the results based on book and market capital. As expected, our listed-entity results support our large-bank findings in that capital helps listed entities in particular during banking crises. Specifically, we find that those with higher capital ratios are more likely to survive banking crises (based on MKTRAT_listed only; as expected, this result is not significant based on EQRAT_listed); increase their market shares during banking crises (based on both capital measures; also significant during market crises based on EQRAT_listed); and increase their profitability during banking crises (based on MKTRAT_listed only) and market crises (based on both capital measures). Also as expected, the results are strongest based on market capital ratios. Nonetheless, we present the book value analysis as the main approach in our paper for several reasons. First, most of the policy proposals to raise capital in the wake of the recent financial crisis aim to directly strengthen the book value of capital. While this may also increase the market value, a book value analysis may be more relevant for policy purposes. Second, market values are only available for the relatively small subset of publicly-traded banks / BHCs, so we would have an analysis that is not representative of small and medium banks. Thus, only a book value analysis allows us to investigate the effects of capital on all sizes of banks in the industry. Finally, market values of equity are largely based on the markets expectation of future performance. To the extent that the market is prescient, then regressing current performance on lagged market values of capital may create

For example, each multi-BHCs pre-crisis book capital ratio, EQRAT_listed,, is calculated as the ratio of equity capital (summed over all banks in the BHC) to gross total assets (summed over all banks in the BHC), averaged over the eight quarters before the crisis.

35

26

a significant bias in favor of finding favorable effects of capital. By presenting both book-value and marketvalue analyses, we can provide greater confidence in the robustness of our results.

7. Conclusion This paper aims to help a recent debate on whether banks should hold more capital. Existing theories produce conflicting predictions regarding the effect of capital on bank performance during normal times and have little to say about the effect during financial crises. We formulate and test hypotheses regarding the effect of bank capital on bank performance (survival probability, market share, and profitability) during financial crises and normal times. Our two main results are as follows. First, capital enhances the performance of all sizes of banks during banking crises. Second, during normal times and market crises, capital helps only small banks unambiguously in all performance dimensions; it helps medium and large banks improve only profitability during market crises and only market share during normal times. Our survival and market share findings are generally robust, and our profitability results are less so. 36

36

Given our findings, it may seem surprising that banks often resist calls for higher capital (e.g., Mishkin 2000), although most banks are found to hold capital well in excess of regulatory minimums (e.g., Berger, DeYoung, Flannery, Lee, and Oztekin 2008). While outside the scope of our paper, there may be several reasons for resistance: managerial hubris (e.g. Roll 1986) which may include underestimating the probability of banking crises, especially when times are good; private benefits related to the government safety net; and forced departure from privately-optimal capital structure choices (e.g., Mehran and Thakor forthcoming).

27

Appendix I: Description of the financial crises This Appendix describes the two banking crises and the three market crises that occurred in the U.S between 1984:Q1 and 2008:Q4.

Two banking crises Credit crunch (1990:Q1 1992:Q4): During the first three years of the 1990s, bank commercial and industrial lending declined in real terms, particularly for small banks and for small loans. The ascribed causes of the credit crunch include a fall in bank capital from the loan loss experiences of the late 1980s (e.g., Peek and Rosengren 1995), the increases in bank leverage requirements and implementation of Basel I risk-based capital standards during this time period (e.g., Hancock, Laing, and Wilcox 1995, Thakor 1996), an increase in supervisory toughness evidenced in worse examination ratings for a given bank condition, and reduced loan demand because of macroeconomic and regional recessions (e.g., Bernanke and Lown 1991). The existing research provides some support for each of these hypotheses. Subprime lending crisis (2007:Q3 2009:Q4): The subprime lending crisis has been characterized by turmoil in financial markets as banks have experienced difficulty in selling loans in the syndicated loan market and in securitizing loans. The supply of liquidity by banks dried up, as did the provision of liquidity in the interbank market. Many banks experienced substantial losses in capital. Massive loan losses at Countrywide resulted in a takeover by Bank of America. Bear Stearns suffered a fatal loss of confidence among its financiers and was sold at a fire-sale price to J.P. Morgan Chase, with the Federal Reserve guaranteeing $29 billion in potential losses. Washington Mutual, the sixth-largest bank, became the biggest bank failure in the U.S. financial history. J.P. Morgan Chase purchased the banking business while the rest of the organization filed for bankruptcy. IndyMac Bank was seized by the FDIC after it suffered substantial losses and depositors had started to run on the bank. The FDIC sold all deposits and most of the assets to OneWest Bank, FSB, and incurred an estimated loss of about $10 billion. The Federal Reserve also intervened in some unprecedented ways in the market. It extended its safety-net privileges to investment banks and one insurance company (AIG), intervened in the commercial paper market, and began holding mortgage-backed securities and lending directly to investment banks. The Treasury initially set aside $250 billion out of its $700 billion bailout package (TARP program) to enhance capital ratios of selected banks. This included $125 billion in $10 billion and $25 billion increments to each of nine large banking organizations. Some of these banks used these funds

28

to acquire lesser-capitalized peers. For example, PNC Bank used TARP funds to acquire National City Bank, and Bank of America bought Merrill Lynch. In all, the Treasury invested over $300 billion in almost 700 financial institutions (as well as over $80 billion in the automobile industry). During 2009, 140 U.S. banks failed, and the FDIC Bank Insurance Fund fell into a deficit position. By the first quarter of 2010, much of the TARP funds invested in financial institutions had been repaid, and order had been restored to most of the financial markets, although small banks continued to fail at a high rate.

Three market crises Stock market crash (1987:Q4): On Monday, October 19, 1987, the stock market crashed, with the S&P500 index falling about 20%. During the years before the crash, the level of the stock market had increased dramatically, causing some concern that the market had become overvalued. 37 A few days before the crash, two events occurred that may have helped precipitate the crash: legislation was enacted to eliminate certain tax benefits associated with financing mergers; and information was released that the trade deficit was above expectations. Both events seemed to have added to the selling pressure and a record trading volume on Oct. 19, in part caused by program trading, overwhelmed many systems. Russian debt crisis / LTCM bailout (1998:Q3 1998:Q4): Since its inception in March 1994, hedge fund Long-Term Capital Management (LTCM) followed an arbitrage strategy that was avowedly market neutral, designed to make money regardless of whether prices were rising or falling. When Russia defaulted on its sovereign debt on August 17, 1998, investors fled from other government paper to the safe haven of U.S. treasuries. This flight to liquidity caused an unexpected widening of spreads on supposedly low-risk portfolios. By the end of August 1998, LTCMs capital had dropped to $2.3 billion, less than 50% of its December 1997 value, with assets standing at $126 billion. In the first three weeks of September, LTCMs capital dropped further to $600 million without shrinking the portfolio. Banks began to doubt its ability to meet margin calls. To prevent a potential systemic meltdown triggered by the collapse of the worlds largest hedge fund, the Federal Reserve Bank of New York organized a $3.5 billion bail-out by LTCMs major creditors on September 23, 1998. In 1998:Q4, several large banks had to take substantial write-offs as a result of losses on their investments.

37

E.g., Raging bull, stock markets surge is puzzling investors: When will it end? on page 1 of the Wall Street Journal, Jan. 19, 1987.

29

Bursting of the dot.com bubble and Sept. 11 terrorist attack (2000:Q2 2002:Q3): The dot.com bubble was a speculative stock price bubble that was built up during the mid- to late-1990s. During this period, many internet-based companies, commonly referred to as dot.coms, were founded. Rapidly increasing stock prices and widely available venture capital created an environment in which many of these companies seemed to focus largely on increasing market share. At the height of the boom, many dot.coms were able to go public and raise substantial amounts of money even if they had never earned any profits, and in some cases had not even earned any revenues. On March 10, 2000, the Nasdaq composite index peaked at more than double its value just a year before. After the bursting of the bubble, many dot.coms ran out of capital and were acquired or filed for bankruptcy (examples of the latter include WorldCom and Pets.com). The U.S. economy started to slow down and business investments began falling. The September 11, 2001 terrorist attacks may have exacerbated the stock market downturn by adversely affecting investor sentiment. By 2002:Q3, the Nasdaq index had fallen by 78%, wiping out $5 trillion in market value of mostly technology firms.

30

Appendix II: Construction of bank liquidity creation (Berger and Bouwman 2009) We calculate a banks dollar amount of liquidity creation using a three-step procedure, which is discussed below and illustrated in Table A-1. Step 1: All bank activities (assets, liabilities, equity, and off-balance sheet activities) are classified as liquid, semi-liquid, or illiquid. For assets, this is done based on the ease, cost, and time for banks to dispose of their obligations in order to meet liquidity demands. For liabilities and equity, this is done based on the ease, cost, and time for customers to obtain liquid funds from the bank. We follow a similar approach for off-balance sheet activities, classifying them based on functionally similar on-balance sheet activities. For all activities other than loans, this classification process uses information on both product category and maturity. Due to data restrictions, loans are classified entirely by category. Step 2: Weights are assigned to all the bank activities classified in Step 1. The weights are consistent with liquidity creation theory, which argues that banks create liquidity on the balance sheet when they transform illiquid assets into liquid liabilities. Positive weights are therefore applied to illiquid assets and liquid liabilities. Following similar logic, negative weights are applied to liquid assets and illiquid liabilities and equity, since banks destroy liquidity when they use illiquid liabilities to finance liquid assets. Weights of and - are used because liquidity creation is only half attributable to the source or use of funds alone.38 An intermediate weight of 0 is applied to semi-liquid assets and liabilities. Weights for off-balance sheet activities are assigned using the same principles. Step 3: The activities as classified in Step 1 and as weighted in Step 2 are combined to construct Berger and Bouwmans (2009) preferred liquidity creation measure. This measure classifies loans by category, while all activities other than loans are classified using information on product category and maturity, and includes offbalance sheet activities.39 To obtain the dollar amount of liquidity creation at a particular bank, we multiply the weights of , -, or 0, respectively, times the dollar amounts of the corresponding bank activities and add the weighted dollar amounts.

The following examples illustrate this principle. Maximum liquidity is created when $1 of liquid liabilities is used to finance $1 in illiquid assets: * $1 + * $1 = $1. When $1 of liquid liabilities is used to finance $1 in liquid assets, no liquidity is created ( * $1 + - * $1 = $0) because the bank holds items of approximately the same liquidity as those it gives to the nonbank public. Maximum liquidity is destroyed when $1 of illiquid liabilities or equity is used to finance $1 of liquid assets: * $1 + - * $1 = -$1. 39 Berger and Bouwman (2009) construct four liquidity creation measures by alternatively classifying loans by category or maturity, and by alternatively including or excluding off-balance sheet activities. However, they argue that the measure we use here is the preferred measure since for liquidity creation, banks ability to securitize or sell loans is more important than loan maturity, and banks do create liquidity both on the balance sheet and off the balance sheet.

38

31

Table A-1: Liquidity classification of bank activities and construction of the liquidity creation measure
This table explains the Berger and Bouwman (2009) methodology to construct their preferred liquidity creation measure that classifies loans by category and includes off-balance sheet activities in three steps. Step 1: Classify all bank activities as liquid, semi-liquid, or illiquid. For activities other than loans, information on product category and maturity are combined. Due to data limitations, loans are classified entirely by product category. Step 2: Assign weights to the activities classified in Step 1. ASSETS:
Illiquid assets (weight = ) Commercial real estate loans (CRE) Loans to finance agricultural production Commercial and industrial loans (C&I) Other loans and lease financing receivables Other real estate owned (OREO) Investment in unconsolidated subsidiaries Intangible assets Premises Other assets Semi-liquid assets (weight = 0) Residential real estate loans (RRE) Consumer loans Loans to depository institutions Loans to state and local governments Loans to foreign governments Liquid assets (weight = - ) Cash and due from other institutions All securities (regardless of maturity) Trading assets Fed funds sold

LIABILITIES PLUS EQUITY:

Liquid liabilities (weight = ) Transactions deposits Savings deposits Overnight federal funds purchased Trading liabilities Semi-liquid liabilities (weight = 0) Time deposits Other borrowed money Illiquid liabilities plus equity (weight = - ) Subordinated debt Other liabilities Equity

OFF-BALANCE SHEET GUARANTEES (notional values):

Illiquid guarantees (weight = ) Unused commitments Net standby letters of credit Commercial and similar letters of credit All other off-balance sheet liabilities Semi-liquid guarantees (weight = 0) Net credit derivatives Net securities lent Liquid guarantees (weight = - ) Net participations acquired

OFF-BALANCE SHEET DERIVATIVES (gross fair values):

Liquid derivatives (weight = -) Interest rate derivatives Foreign exchange derivatives Equity and commodity derivatives

Step 3: Combine bank activities as classified in Step 1 and as weighted in Step 2 to construct the liquidity creation (LC) measure.
LC = + * illiquid assets + * liquid liabilities + * illiquid guarantees + 0 * semi-liquid assets + 0 * semi-liquid liabilities + 0 * semi-liquid guarantees * liquid assets * illiquid liabilities * equity * liquid guarantees * liquid derivatives

32

References Ashcraft, Adam B., 2008, Does the market discipline banks? New evidence from the regulatory capital mix, Journal of Financial Intermediation 17: 787-821. Acharya, Viral V., Hamid Mehran, and Anjan V. Thakor, 2010, Caught between Scylla and Charybdis? Regulating bank leverage when there is rent seeking and risk shifting, working paper. Allen, Franklin, Elena Carletti, and Robert Marquez, forthcoming, Credit market competition and capital regulation, Review of Financial Studies. Allen, Franklin, and Douglas Gale, 2004, Competition and financial stability, Journal of Money, Credit and Banking 36: 453-480. Bae, Kee-Hong, Jun-Koo Kang, and Chan-Woo Lim, 2002, The value of durable bank relationships: Evidence from Korean banking shocks, Journal of Financial Economics 64: 181-214. Basel III BIS Press Release, 2010, Group of Governors and Heads of Supervision announces higher global minimum capital standards (September 12). Berger, Allen N., 1995, The relationship between capital and earnings in banking, Journal of Money, Credit and Banking 27: 432-456. Berger, Allen N., and Christa H.S. Bouwman, 2009, Bank liquidity creation, Review of Financial Studies 22: 37793837. Berger, Allen N., Robert DeYoung, Mark J. Flannery, David Lee, and Ozde Oztekin, 2008, How do large banking organizations manage their capital ratios, Journal of Financial Services Research 34: 123-149. Berger, Allen N., Nathan H. Miller, Mitchell A. Petersen, Raghuram G. Rajan, and Jeremy C. Stein, 2005, Does function follow organizational form? Evidence from the lending practices of large and small banks, Journal of Financial Economics 76: 237-269. Bernanke, Ben S., and Cara S. Lown, 1991, The credit crunch, Brookings Papers on Economic Activity (2:1991): 205-48. Bharath, Sreedhar, Sandeep Dahiya, Anthony Saunders, and Anand Srinivasan, 2007, So what do I get? The banks view of lending relationships, Journal of Financial Economics 85: 368-419. Bhattacharya, Sudipto, and Anjan V. Thakor, 1993, Contemporary banking theory, Journal of Financial Intermediation, 3: 2-50. Boot, Arnoud W.A., Stuart Greenbaum, and Anjan V. Thakor, 1993, Reputation and discretion in financial contracting, American Economic Review 83: 1165-1183. Boot, Arnoud W.A., and Matej Marinc, 2008, Competition and entry in banking: Implications for capital regulation, working paper. Boot, Arnoud, W.A., and Anjan V. Thakor, 1993, Self-interested bank regulation, American Economic Review 83: 206-212. Brander, James A., and Tracy R. Lewis, 1986, Oligopoly and financial structure: The limited liability effect, American Economic Review 76: 956-970. Calomiris, Charles W., and Charles M. Kahn, 1991, The role of demandable debt in structuring optimal banking arrangements, American Economic Review 81: 497-513. Campello, Murillo, Erasmo Giambona, John R. Graham, and Campbell R. Harvey, 2009, Liquidity management and corporate investment during a financial crisis, working paper. Coval, Joshua D., and Tobias J. Moskowitz, 1999, Home bias at home: Local equity preferences in domestic portfolios, Journal of Finance 54: 2045-2073.

33

Coval, Joshua D., and Anjan V. Thakor, 2005, Financial intermediation as a beliefs-bridge between optimists and pessimists, Journal of Financial Economics 75: 535-569. Diamond, Douglas W., and Raghuram G. Rajan, 2001, Liquidity risk, liquidity creation, and financial fragility: a theory of banking, Journal of Political Economy 109: 287-327. Freixas, Xavier, and Jean-Charles Rochet, 2008, Microeconomics of banking (2nd edition), MIT Press. Hancock, Diana, Andrew J. Laing, and James A. Wilcox, 1995, Bank balance sheet shocks and aggregate shocks: their dynamic effects on bank capital and lending, Journal of Banking and Finance 19: 661-77. Hart, Oliver, and Luigi Zingales, 2009, A new capital regulation for large financial institutions, working paper. Holmstrom, Bengt, and Jean Tirole, 1997, Financial intermediation, loanable funds, and the real sector, Quarterly Journal of Economics 112: 663-691. Hoshi, Takeo, and Anil K. Kashyap, 2010, Will the U.S. bank recapitalization succeed? Eight lessons from Japan, Journal of Financial Economics 97: 398-417. Houston, Joel F., Christopher James, and David Marcus, 1997, Capital market frictions and the role of internal capital markets in banking, Journal of Financial Economics 46: 135-164. Kashyap, Anil K., Raghuram G. Rajan, and Jeremy C. Stein, 2002, Banks as liquidity providers: an explanation for the coexistence of lending and deposit-taking, Journal of Finance 57: 33-73. Kashyap, Anil K., Raghuram G. Rajan, and Jeremy C. Stein, 2009, Rethinking capital regulation, working paper. Lyandres, Evgeny, 2006, Capital structure and interaction among firms in output markets: Theory and evidence, Journal of Business 79: 2381-2421. Mehran, Hamid, and Anjan V. Thakor, forthcoming, Bank capital and value in the cross-section, Review of Financial Studies. Merton, Robert C., 1977, An analytic derivation of the cost of the deposit insurance and loan guarantees: An application of modern option pricing theory, Journal of Banking and Finance 1: 3-11. Mishkin, Frederick, 2000, The economics of money, banking and financial markets (6th edition), Addison Wesley, New York. Modigliani, Franco, and Merton Miller, 1963, Corporate income taxes and the cost of capital: a correction, American Economic Review 53: 433 443. Ongena, Steven, and David C. Smith, 2001, The duration of bank relationships, Journal of Financial Economics 61: 449-475. Peek, Joe, and Eric S. Rosengren. 1995, The capital crunch: neither a borrower nor a lender be, Journal of Money, Credit and Banking 27: 625-38. Repullo, Rafael, 2004, Capital requirements, market power, and risk taking in banking, Journal of Financial Intermediation, 13: 156-182 Roll, Richard, 1986, The hubris hypothesis of corporate takeovers, Journal of Business 59: 197-216. Song, Fenghua, and Anjan V. Thakor, 2007, Relationship banking, fragility and the asset-liability matching prolem, Review of Financial Studies 200: 2129 2177. Thakor, Anjan V., 1996, Capital requirements, monetary policy, and aggregate bank lending: theory and empirical evidence, Journal of Finance 51: 279-324. Veronesi, Pietro, and Luigi Zingales, 2010, Paulsons gift, Journal of Financial Economics 97: 339-368. Von Thadden, Ernst-Ludwig, 2004, Bank capital adequacy regulation under the new Basel Accord, Journal of Financial Intermediation 13: 90-95. Wheelock, David C., and Paul W. Wilson, 2000, Why do Banks Disappear? The Determinants of U.S. Bank Failures and Acquisitions, Review of Economics and Statistics 82: 127-138.

34

Table 1: Summary statistics on the regression variables

This table contains means and standard deviations (in parentheses) on all the regression variables used to examine the effect of pre-crisis capital ratios on banks ability to survive crises, and improve their competitive positions and profitability during such crises. We distinguish between banking crises (the credit crunch of the early 1990s and the recent subprime lending crisis), market crises (the 1987 stock market crash, the Russian debt crisis plus LTCM bailout in 1998, and the bursting of the dot.com bubble plus September 11), and normal times (see Section 3). SURV, survival, is a dummy that equals 1 if the bank is in the sample one quarter before such a crisis started and is still in the sample one quarter after the crisis, and 0 otherwise. %MKTSHARE, the change in market share, is measured as the banks average market share during a crisis minus its average market share over the eight quarters before the crisis, normalized by its pre-crisis market share and multiplied by 100. Market share is the banks liquidity creation (LC) as a fraction of total LC. PROF, the change in profitability, is measured as the banks average profitability during a crisis minus its average profitability over the eight quarters before the crisis. Profitability is ROE, net income divided by equity capital. All independent variables are measured as averages over the eight quarters prior to a crisis (except as noted). EQRAT is the equity capital ratio, calculated as equity capital as a proportion of GTA. GTA equals total assets plus the allowance for loan and the lease losses and the allocated transfer risk reserve (a reserve for certain foreign loans). CREDRISK, credit risk, is defined as the banks Basel I risk-weighted assets divided by GTA. lnLC is the log of liquidity creation. D-BHC is a dummy variable that equals 1 if the bank has been part of a bank holding company over the eight quarters before the crisis. HHI is a bank-level Herfindahl index based on bank and thrift deposits (the only variable for which geographic location is publicly available). We first establish the Herfindahl index of the local markets in which the bank has deposits and then weight these market indices by the proportion of the banks deposits in each of these markets. ROA is net income divided by total assets. All dollar values are expressed in real 2009:Q4 dollars using the implicit GDP price deflator. Small banks Dependent variables: SURV %MKTSHARE PROF Independent variables: EQRAT CREDRISK lnLC D-BHC HHI ROA Obs 0.941 (0.236) 0.255 (0.760) -0.024 (0.080) 0.100 (0.051) 0.633 (0.135) 9.755 (1.544) 0.709 (0.447) 0.235 (0.171) 0.008 (0.010) 16206 Banking crises Medium Large banks banks 0.866 (0.342) 0.183 (0.554) -0.071 (0.094) 0.088 (0.045) 0.745 (0.144) 13.088 (0.801) 0.854 (0.348) 0.174 (0.113) 0.010 (0.010) 600 0.909 (0.288) 0.192 (0.605) -0.077 (0.101) 0.083 (0.044) 0.795 (0.178) 14.960 (1.418) 0.927 (0.255) 0.164 (0.089) 0.010 (0.008) 422 Small banks 0.976 (0.154) 0.213 (0.666) -0.014 (0.078) 0.097 (0.038) 0.607 (0.118) 9.544 (1.419) 0.706 (0.443) 0.216 (0.130) 0.009 (0.011) 25873 Market crises Medium banks 0.956 (0.206) 0.212 (0.531) -0.015 (0.070) 0.083 (0.040) 0.684 (0.136) 12.814 (0.871) 0.883 (0.309) 0.169 (0.070) 0.011 (0.010) 788 Large banks 0.976 (0.155) 0.166 (0.471) -0.020 (0.080) 0.076 (0.028) 0.768 (0.178) 15.019 (1.420) 0.945 (0.223) 0.164 (0.078) 0.011 (0.011) 573 Small banks 0.972 (0.165) 0.183 (0.603) 0.001 (0.048) 0.101 (0.035) 0.614 (0.131) 9.861 (1.411) 0.725 (0.440) 0.210 (0.135) 0.011 (0.008) 15522 Normal times Medium banks 0.958 (0.200) 0.136 (0.434) -0.005 (0.050) 0.093 (0.044) 0.680 (0.163) 12.946 (0.824) 0.845 (0.357) 0.163 (0.080) 0.012 (0.008) 578 Large banks 0.965 (0.185) 0.161 (0.487) -0.006 (0.057) 0.089 (0.036) 0.704 (0.168) 14.969 (1.494) 0.935 (0.240) 0.162 (0.080) 0.013 (0.010) 426

35

Table 2: The effect of the banks pre-crisis capital ratio on its ability to survive crises and normal times
Panel A shows the results of logit regressions which examine how pre-crisis capital ratios affect banks ability to survive banking crises (BNKCRIS: the credit crunch of the early 1990s and the recent subprime lending crisis), market crises (MKTCRIS: the 1987 stock market crash, the Russian debt crisis plus LTCM bailout in 1998, and the bursting of the dot.com bubble plus September 11), and normal times (NORMALTIME) (see Section 3). In Panel A, the dependent variable is , where SURV is a dummy that equals 1 if the bank is in the sample

one quarter before the crisis started and is still in the sample one quarter after the crisis, and 0 otherwise. Panels B D contain the predicted probability of surviving banking crises, market crises, and normal times, respectively, at different capital ratios. Results are shown for small banks (GTA up to $1 billion), medium banks (GTA exceeding $1 billion and up to $3 billion), and large banks (GTA exceeding $3 billion). GTA equals total assets plus the allowance for loan and the lease losses and the allocated transfer risk reserve (a reserve for certain foreign loans). The key exogenous variables (EQRAT * BNKCRIS, EQRAT * MKTCRIS, and EQRAT * NORMALTIME) and control variables are averaged over the eight quarters before a crisis (except as noted). EQRAT is the equity capital ratio, calculated as equity capital as a proportion of GTA. GTA equals total assets plus the allowance for loan and the lease losses and the allocated transfer risk reserve (a reserve for certain foreign loans). CREDRISK, credit risk, is defined as the banks Basel I risk-weighted assets divided by GTA. lnLC is the log of liquidity creation. D-BHC is a dummy variable that equals 1 if the bank has been part of a bank holding company over the eight quarters before the crisis. HHI is a bank-level Herfindahl index based on bank and thrift deposits (the only variable for which geographic location is publicly available). We first establish the Herfindahl index of the local markets in which the bank has deposits and then weight these market indices by the proportion of the banks deposits in each of these markets. ROA is net income divided by total assets. All dollar values are expressed in real 2009:Q4 dollars using the implicit GDP price deflator. t-statistics are in parentheses. *, **, and *** denote significance at the 10%, 5%, and 1% level, respectively. Panel A: The effect of the banks pre-crisis capital ratio on its ability to survive Small banks 15.839 (8.70)*** 12.864 (8.28)*** 9.264 (4.79)*** -4.658 (-9.16)*** -1.401 (-3.05)*** 0.259 (0.52) -0.047 (-1.86)* 0.136 (2.25)** 1.634 (6.04)*** (50.26) (19.87)*** 1.704 (4.31)*** 46017 SURV Medium banks 23.257 (2.22)** -0.623 (-0.10) -3.947 (-0.89) -2.849 (-2.20)** 2.266 (1.21) -1.146 (-0.62) 0.147 (0.87) 0.688 (2.32)** 5.042 (2.74)*** (77.20) (4.97)*** 0.064 (0.030) 1599 Large banks 38.825 (2.11)** 10.898 (0.76) 8.627 (0.62) -0.554 (-0.32) -1.163 (-0.53) 0.001 (0.00) 0.109 (0.73) 1.064 (2.07)** 7.577 (2.63)*** (28.38) (1.66)* -1.738 (-0.780) 1194

EQRAT * BNKCRIS EQRAT * MKTCRIS EQRAT * NORMALTIME CREDRISK * BNKCRIS CREDRISK * MKTCRIS CREDRISK * NORMALTIME lnLC D-BHC HHI ROA Constant

Obs

36

Panel B: Predicted probability of surviving banking crises at different capital ratios Capital ratio: Average minus 1 standard deviation Average Average plus 1 standard deviation Small banks 4.9% 10.0% 15.1% Medium banks 4.3% 8.8% 13.3% Large banks 3.9% 8.3% 12.7%

Predicted survival probability when capital is at:

Average minus 1 standard deviation Average Average plus 1 standard deviation

92.8% 96.7% 98.5%

82.5% 93.0% 97.4%

88.1% 97.2% 99.4%

Panel C: Predicted probability of surviving market crises at different capital ratios Small banks Average minus 1 standard deviation 5.9% Capital ratio: Average 9.7% Average plus 1 standard deviation 13.5% Predicted survival probability when capital is at:

Medium banks 4.3% 8.3% 12.4%

Large banks 4.7% 7.6% 10.4%

Average minus 1 standard deviation Average Average plus 1 standard deviation

97.2% 98.1% 98.8%

98.3% 98.2% 98.1%

99.0% 99.3% 99.5%

Panel D: Predicted probability of surviving normal times at different capital ratios Small banks Average minus 1 standard deviation 6.5% Capital ratio: Average 10.1% Average plus 1 standard deviation 13.6% Predicted survival probability when capital is at:

Medium banks 4.9% 9.3% 13.6%

Large banks 5.2% 8.9% 12.5%

Average minus 1 standard deviation Average Average plus 1 standard deviation

97.2% 97.8% 98.4%

97.3% 96.9% 96.4%

97.3% 97.6% 97.8%

37

Table 3: The effect of the banks pre-crisis capital ratio on its market share during crises and normal times
This table shows how pre-crisis bank capital ratios affect banks competitive positions during banking crises (BNKCRIS: the credit crunch of the early 1990s and the recent subprime lending crisis), market crises (MKTCRIS: the 1987 stock market crash, the Russian debt crisis plus LTCM bailout in 1998, and the bursting of the dot.com bubble plus September 11), and normal times (NORMALTIME) (see Section 3). Results are shown for small banks (GTA up to $1 billion), medium banks (GTA exceeding $1 billion and up to $3 billion), and large banks (GTA exceeding $3 billion). GTA equals total assets plus the allowance for loan and the lease losses and the allocated transfer risk reserve (a reserve for certain foreign loans). The dependent variable is %MKTSHARE, the percentage change in the banks liquidity creation market share measured as the banks average market share during a crisis minus its average market share over the eight quarters before the crisis, divided by its pre-crisis market share and multiplied by 100. The key exogenous variables (EQRAT * BNKCRIS, EQRAT * MKTCRIS, and EQRAT * NORMALTIME) and control variables are averaged over the eight quarters before a crisis. EQRAT is the equity capital ratio, calculated as equity capital as a proportion of GTA. GTA equals total assets plus the allowance for loan and the lease losses and the allocated transfer risk reserve (a reserve for certain foreign loans). CREDRISK, credit risk, is defined as the banks Basel I risk-weighted assets divided by GTA. lnLC is the log of liquidity creation. D-BHC is a dummy variable that equals 1 if the bank has been part of a bank holding company over the eight quarters before the crisis. HHI is a bank-level Herfindahl index based on bank and thrift deposits (the only variable for which geographic location is publicly available). We first establish the Herfindahl index of the local markets in which the bank has deposits and then weight these market indices by the proportion of the banks deposits in each of these markets. All dollar values are expressed in real 2009:Q4 dollars using the implicit GDP price deflator. t-statistics based on robust standard errors are in parentheses. *, **, and *** denote significance at the 10%, 5%, and 1% level, respectively. %MKTSHARE Medium banks 1.305 (2.24)** -1.640 (-2.79)*** 1.988 (1.74)* -0.213 (-1.00) 0.517 (2.77)*** 0.554 (2.83)*** -0.218 (-7.31)*** 0.222 (5.72)*** 0.226 (1.600) 2.526 (7.33)*** 1911 0.14

EQRAT * BNKCRIS EQRAT * MKTCRIS EQRAT * NORMALTIME CREDRISK * BNKCRIS CREDRISK * MKTCRIS CREDRISK * NORMALTIME lnLC D-BHC HHI Constant

Small banks 5.124 (28.29)*** 4.072 (20.52)*** 3.382 (14.27)*** -0.266 (-3.98)*** -0.572 (-10.50)*** -0.734 (-12.49)*** -0.072 (-21.77)*** 0.013 (1.79)* -0.216 (-10.04)*** 0.885 (23.82)*** 52107 0.13

Large banks 2.448 (3.34)*** -0.492 (-0.43) 2.841 (2.43)** -0.519 (-2.37)** -0.055 (-0.30) -0.035 (-0.17) -0.071 (-4.56)*** 0.005 (0.05) -0.017 (-0.120) 1.258 (5.19)*** 1382 0.09

Obs Adjusted R2

38

Table 4: The effect of the banks pre-crisis capital ratio on its profitability during crises and normal times
This table shows how pre-crisis bank capital ratios affect banks profitability during banking crises (BNKCRIS: the credit crunch of the early 1990s and the recent subprime lending crisis), market crises (MKTCRIS: the 1987 stock market crash, the Russian debt crisis plus LTCM bailout in 1998, and the bursting of the dot.com bubble plus September 11), and normal times (NORMALTIME) (see Section 3). Results are shown for small banks (GTA up to $1 billion), medium banks (GTA exceeding $1 billion and up to $3 billion), and large banks (GTA exceeding $3 billion). GTA equals total assets plus the allowance for loan and the lease losses and the allocated transfer risk reserve (a reserve for certain foreign loans). The dependent variable is PROF, the change in profitability measured as the banks average ROE (net income divided by GTA) during a crisis minus its average ROE over the eight quarters before the crisis. The key exogenous variables (EQRAT * BNKCRIS, EQRAT * MKTCRIS, and EQRAT * NORMALTIME) and control variables are averaged over the eight quarters before a crisis. EQRAT is the equity capital ratio, calculated as equity capital as a proportion of GTA. GTA equals total assets plus the allowance for loan and the lease losses and the allocated transfer risk reserve (a reserve for certain foreign loans). CREDRISK, credit risk, is defined as the banks Basel I risk-weighted assets divided by GTA. lnLC is the log of liquidity creation. D-BHC is a dummy variable that equals 1 if the bank has been part of a bank holding company over the eight quarters before the crisis. HHI is a bank-level Herfindahl index based on bank and thrift deposits (the only variable for which geographic location is publicly available). We first establish the Herfindahl index of the local markets in which the bank has deposits and then weight these market indices by the proportion of the banks deposits in each of these markets. All dollar values are expressed in real 2009:Q4 dollars using the implicit GDP price deflator. t-statistics based on robust standard errors are in parentheses. *, **, and *** denote significance at the 10%, 5%, and 1% level, respectively. PROF Medium banks 0.109 (0.70) 0.132 (2.00)** 0.044 (0.64) -0.224 (-4.76)*** -0.066 (-2.65)*** -0.008 (-0.37) 0.004 (1.48) 0.001 (0.12) 0.036 (1.91)* -0.025 (-0.850) 1911 0.22

EQRAT * BNKCRIS EQRAT * MKTCRIS EQRAT * NORMALTIME CREDRISK * BNKCRIS CREDRISK * MKTCRIS CREDRISK * NORMALTIME lnLC D-BHC HHI Constant

Small banks 0.061 (3.41)*** 0.108 (6.39)*** 0.135 (7.88)*** -0.201 (-30.18)*** -0.049 (-8.24)*** 0.029 (6.12)*** 0.000 (0.44) 0.004 (5.20)*** 0.009 (4.52)*** 0.017 (6.08)*** 52107 0.09

Large banks 0.239 (2.71)*** 0.388 (2.33)** 0.020 (0.20) -0.151 (-5.24)*** -0.030 (-1.24) -0.041 (-1.70)* -0.001 (-0.81) -0.003 (-0.40) 0.031 (1.240) 0.022 (0.960) 1382 0.18

Obs Adjusted R2

39

Table 5: Robustness
This table presents the results of six checks to establish the robustness of our results. Panel A uses alternative specifications of survival, competitive position, and profitability. Panel B uses regulatory capital ratios instead of the equity-to-assets ratio. Panel C drops Too-Big-To-Fail banks from the large-bank sample. Panel D uses alternative cutoffs separating medium and large banks. Panel E measures pre-crisis capital ratios alternatively one quarter before the crisis starts or averaged over the four quarters before the crisis. Panel F deals with the potential endogeneity issues related to capital by using instrumental variable regressions. The crises include banking crises (BNKCRIS: the credit crunch of the early 1990s and the recent subprime lending crisis), market crises (MKTCRIS: the 1987 stock market crash, the Russian debt crisis plus LTCM bailout in 1998, and the bursting of the dot.com bubble plus September 11), and normal times (NORMALTIME) (see Section 3). Results are shown for small, medium, and large banks the cutoffs are GTA of $1 billion and $3 billion, respectively, unless otherwise noted. GTA equals total assets plus the allowance for loan and the lease losses and the allocated transfer risk reserve (a reserve for certain foreign loans). SURV, survival, is a dummy that equals 1 if the bank is in the sample one quarter before such a crisis started and is still in the sample one quarter after the crisis, and 0 otherwise. %MKTSHARE, the change in market share, is measured as the banks average market share during a crisis minus its average market share over the eight quarters before the crisis, normalized by its pre-crisis market share and multiplied by 100. Market share is the banks liquidity creation (LC) as a fraction of total LC. PROF, the change in profitability, is measured as the banks average profitability during a crisis minus its average profitability over the eight quarters before the crisis. Profitability is ROE, net income divided by equity capital. To preserve space, we only present the coefficients on the key exogenous variables although all the control variables (see Tables 2 4) are generally included in the regressions. Exception: the risk interaction terms are not included in Panel B because the numerator of the risk variable (Basel I risk-weighted assets) is identical to the denominator of the regulatory capital ratio. t-statistics based on robust standard errors are in parentheses. *, **, and *** denote significance at the 10%, 5%, and 1% level, respectively.

Panel A: Robustness use alternative definitions of survival, market share, and profitability
A1: SURV_alt Until Q4 (instead of Q1) after crisis Small Medium Large banks banks banks 10.648 22.777 9.170 (7.40)*** (2.53)** (0.71) 12.414 -1.814 4.681 (10.91)*** (-0.57) (0.58) 7.796 -1.303 -3.656 (5.85)*** (-0.33) (-0.70) 46017 1599 1194 A2: %MKTSHARE_alt Based on GTA (instead of LC) Small Medium Large banks banks banks 1.593 0.653 1.158 (31.27)*** (1.97)** (4.20)*** 0.939 -0.280 0.502 (18.06)*** (-2.15)** (1.40) 0.618 0.117 0.653 (8.59)*** (0.31) (1.61) 52107 0.14 1911 0.06 1382 0.04 A3: PROF_alt ROA (instead of ROE) Small Medium banks banks 0.013 -0.014 (6.93)*** (-0.67) 0.004 -0.004 (2.09)** (-0.68) 0.016 -0.015 (7.94)*** (-1.73)* 52107 0.08 1911 0.24

EQRAT * BNKCRIS EQRAT * MKTCRIS EQRAT * NORMALTIME

Large banks -0.004 (-0.57) -0.003 (-0.16) -0.008 (-0.71) 1382 0.19

Obs Adj R2

40

Panel B: Robustness use regulatory capital ratios

SURV Medium banks 24.316 (2.27)** -1.282 (-0.80) -0.853 (-0.44) 1599 %MKTSHARE Medium Large banks banks 1.015 3.487 (1.43) (3.36)*** -0.384 -0.314 (-0.93) (-0.48) 0.868 1.727 (1.99)** (2.10)** 1911 0.12 1382 0.09 PROF Medium banks 0.282 (2.51)** 0.049 (2.18)** 0.038 (1.23) 1911 0.18

TIER1RAT * BNKCRIS TIER1RAT * MKTCRIS TIER1RAT * NORMALTIME

Small banks 16.707 (8.91)*** 5.324 (6.66)*** 2.942 (3.52)*** 46017

Large banks 40.914 (2.23)** 4.954 (0.74) 3.383 (0.53) 1194

Small banks 2.761 (20.92)*** 2.575 (22.23)*** 2.213 (18.52)*** 52107 0.13

Small banks 0.098 (8.32)*** 0.047 (5.89)*** 0.000 (0.00) 52107 0.07

Large banks 0.445 (3.69)*** 0.143 (2.03)** 0.071 (1.26) 1382 0.16

Obs Adj R2

Panel C: Robustness exclude Too-Big-To-Fail banks

C1: Drop banks with GTA > $50 billion %MKT SURV SHARE PROF 42.959 2.707 0.267 (2.15)** (3.17)*** (2.68)*** 9.947 -0.510 0.417 (0.69) (-0.42) (2.44)** 6.068 3.025 0.006 (0.43) (2.55)** (0.05) 1106 1273 0.10 1273 0.19 C2: Drop 19 largest banks each quarter %MKT SURV SHARE PROF 47.774 2.700 0.285 (2.32)** (3.14)*** (2.83)*** 10.689 -0.548 0.324 (0.73) (-0.43) (2.02)** 6.816 3.040 0.012 (0.46) (2.58)** (0.12) 1088 1260 0.10 1260 0.19

EQRAT * BNKCRIS EQRAT * MKTCRIS EQRAT * NORMALTIME Obs Adj R2

41

Panel D: Robustness use alternative cutoffs between medium and large banks of $5 billion and $10 billion
D1: $5 billion cutoff %MKTSHARE Medium Large banks banks 1.756 2.502 (3.05)*** (2.11)** -1.392 -0.586 (-2.37)** (-0.34) 2.304 1.886 (2.45)** (1.11) 2312 0.12 981 0.09 PROF Medium Large banks banks 0.206 0.103 (2.24)** (0.72) 0.156 0.521 (2.19)** (1.94)* 0.028 0.011 (0.48) (0.06) 2312 0.21 981 0.16

EQRAT * BNKCRIS EQRAT * MKTCRIS EQRAT * NORMALTIME Obs Adj R2

SURV Medium banks 29.779 (2.97)*** 3.309 (0.48) -3.751 (-0.86) 1942

Large banks 48.853 (1.86)* 8.839 (0.42) 11.747 (0.72) 851

EQRAT * BNKCRIS EQRAT * MKTCRIS EQRAT * NORMALTIME Obs Adj R2

SURV Medium banks 25.939 (2.68)*** 3.448 (0.51) -1.078 (-0.22) 2315

Large banks 68.809 (1.89)* 47.787 (1.00) 5.419 (0.21) 239

D2: $10 billion cutoff %MKTSHARE Medium Large banks banks 1.694 2.746 (2.96)*** (1.75)* -1.370 4.032 (-2.56)** (1.95)* 2.464 -0.251 (2.71)*** (-0.16) 2739 0.10 554 0.10

PROF Medium Large banks banks 0.211 0.113 (2.41)** (0.72) 0.164 0.498 (2.07)** (1.73)* 0.043 -0.166 (0.75) (-0.84) 2739 0.21 554 0.16

42

Panel E: Robustness measure pre-crisis capital ratios at different points before a crisis
E1: Use the average capital ratio over 4 (instead of 8) quarters before the crisis SURV %MKTSHARE PROF Medium Large Small Medium Large Small Medium banks banks banks banks banks banks banks 19.646 29.526 4.787 1.326 2.450 0.029 0.070 (2.07)** (1.71)* (24.07)*** (2.22)** (3.18)*** (1.52) (0.42) -1.030 9.398 3.448 -1.991 -0.404 0.070 0.182 (-0.19) (0.63) (17.29)*** (-4.34)*** (-0.32) (3.99)*** (2.21)** -4.197 3.762 2.704 1.493 2.569 0.069 -0.005 (-0.96) (0.32) (10.92)*** (1.39) (2.20)** (4.03)*** (-0.08) 1599 1194 52029 0.12 1911 0.14 1382 0.08 52029 0.09 1911 0.22

EQRATave4Q * BNKCRIS EQRATave4Q * MKTCRIS EQRATave4Q * NORMALTIME

Small banks 19.164 (10.35)*** 12.863 (8.36)*** 8.482 (4.41)*** 45947

Large banks 0.167 (1.80)* 0.533 (2.72)*** -0.040 (-0.42) 1382 0.18

Obs Adj R2

EQRATQ1 * BNKCRIS EQRATQ1 * MKTCRIS EQRATQ1 * NORMALTIME

E2: Measure capital 1 quarter before the crisis (instead of the average capital ratio over 8 quarters before the crisis) SURV %MKTSHARE PROF Small Medium Large Small Medium Large Small Medium Large banks banks banks banks banks banks banks banks banks 24.208 23.558 39.415 4.225 1.512 2.288 0.004 0.091 0.136 (12.94)*** (2.46)** (2.47)** (16.25)*** (2.29)** (2.72)*** (0.17) (0.58) (1.26) 13.144 0.506 7.574 2.299 -1.821 0.039 0.009 0.168 0.389 (8.29)*** (0.09) (0.53) (11.18)*** (-3.18)*** (0.03) (0.45) (2.15)** (1.91)* 6.930 -4.077 -2.044 1.647 0.273 2.298 -0.008 -0.125 -0.117 (3.53)*** (-0.95) (-0.22) (6.40)*** (0.49) (2.11)** (-0.44) (-2.53)** (-1.40) 45751 1599 1194 51809 0.10 1911 0.13 1382 0.08 51809 0.09 1911 0.22 1382 0.17

Obs Adj R2

43

Panel F: Robustness instrumental variable analysis

SURV Medium banks 0.98 F1: Hausman test for endogeneity %MKTSHARE Small Medium Large banks banks banks 0.00*** 0.16 0.18 PROF Medium banks 0.95

Hausman endogeneity test (p-value)

Small banks 0.27

Large banks 0.33

Small banks 0.00***

Large banks 0.68

EQRAT * BNKCRIS EQRAT * MKTCRIS EQRAT * NORMALTIME

F2: Second-stage IV regressions for small banks %MKT SURV SHARE PROF 2.531 8.640 -1.267 (1.79)* (2.43)** (-2.06)** 1.071 24.341 2.956 (1.10) (5.71)*** (4.87)*** -2.471 19.933 5.213 (-0.58) (1.59) (1.93)* 23575 26671 26671

Obs

44

Table 6: Comparing the effects of book and market capital ratios during crises and normal times
This table examines how book and market capital ratios affect the performance of listed entities during crises and normal times. For this analysis, we include listed banks and listed one-bank-holding companies, and we aggregate the data of all the banks in a listed multi-bank-holding company. The crises include banking crises (BNKCRIS: the credit crunch of the early 1990s and the recent subprime lending crisis), market crises (MKTCRIS: the 1987 stock market crash, the Russian debt crisis plus LTCM bailout in 1998, and the bursting of the dot.com bubble plus September 11), and normal times (NORMALTIME) (see Section 3). Results are shown for small, medium, and large banks the cutoffs are GTA of $1 billion and $3 billion, respectively, unless otherwise noted. GTA equals total assets plus the allowance for loan and the lease losses and the allocated transfer risk reserve (a reserve for certain foreign loans). SURV, survival, is a dummy that equals 1 if the bank is in the sample one quarter before such a crisis started and is still in the sample one quarter after the crisis, and 0 otherwise. %MKTSHARE, the change in market share, is measured as the banks average market share during a crisis minus its average market share over the eight quarters before the crisis, normalized by its pre-crisis market share and multiplied by 100. Market share is the banks liquidity creation (LC) as a fraction of total LC. PROF, the change in profitability, is measured as the banks average profitability during a crisis minus its average profitability over the eight quarters before the crisis. Profitability is ROE, net income divided by equity capital. To preserve space, we only present the coefficients on the key exogenous variables although all the control variables (see Tables 2 4) except for the BHC dummy are included in the regressions. EQRAT_listed, a listed entitys pre-crisis book capital ratio, is the listed entitys book equity capital as a proportion of its GTA. MKTRAT_listed, a listed entitys pre-crisis market capital ratio, is the listed entitys market value of equity (i.e., the number of shares outstanding multiplied by the share price) divided by its assets in market value terms (i.e., the book value of liabilities plus the market value of equity), averaged over the eight quarters before the crisis. t-statistics based on robust standard errors are in parentheses. *, **, and *** denote significance at the 10%, 5%, and 1% level, respectively. Listed entity book capital ratios %MKT SURV SHARE PROF -17.363 1.642 -0.349 (-0.80) (2.06)** (-1.20) -3.020 1.370 0.242 (-0.45) (1.86)* (1.89)* 12.003 0.676 -0.042 (1.17) (1.08) (-0.37) Listed entity market capital ratios %MKT SURV SHARE PROF

EQRAT_listed * BNKCRIS EQRAT_listed * MKTCRIS EQRAT_listed * NORMALTIME MKTRAT_listed * BNKCRIS MKTRAT_listed * MKTCRIS MKTRAT_listed * NORMALTIME

22.173 (1.65)* 4.528 (1.56) -0.135 (-0.06) 1619 1889 0.08 1889 0.31 1619

0.579 (4.04)*** 0.282 (1.10) 0.253 (1.37) 1889 0.08

0.339 (1.93)* 0.158 (2.70)*** 0.052 (1.56) 1889 0.24

Obs Adj R2

45