Does bank competition affect bank liquidity creation? Evidence from a natural experiment

Transcription

1 Does bank competition affect bank liquidity creation? Evidence from a natural experiment Huyen Nguyen 1 ABSTRACT I leverage the staggered nature of the interstate branching reforms following the Interstate Banking and Branching Efficiency Act (IBBEA) of 1994 as a natural experiment to provide the empirical evidence on the impact of banking competition on bank liquidity creation. I further examine the impact of bank competition on bank risk to answer the question if the changes in liquidity creation ex-post the deregulation coincide with bank stability. Using a difference-indifference estimator applied to panel data with 61,135 bank-year observations from 1994 to 2003 in 50 U.S states, I document that increased bank competition is associated with higher bank liquidity creation and lower bank risk taking. The result is mainly driven by small and medium banks which is of interest for policy makers as those banks are essential to maintain a competitive and robust banking market. Keywords: Liquidity Creation, Risk, Competition, Bank Deregulation, Natural Experiment 1 University of Nottingham, huyen.nguyen@nottingham.ac.uk I am grateful to Professor Klaus Schaeck, Dr. Danny McGowan and Dr. Saverio Lopes for helpful comments and suggestions. 1

2 I. Introduction Throughout history, the U.S. banking industry has been hit by more shocks than any other sector of the economy (DeYoung, 2015). Arguably, the most prominent shock banks have experienced were regulatory changes that deregulated the industry (Strahan 2003, Rice and Strahan 2010). Historically, U.S. banks were restricted to operating in a specific state. As a result, a bank s relevant market was defined by state borders and competition was on a local level. Beginning in the 1994, U.S. states introduced legislation that deregulated the industry by removing geographical restrictions on where banks could operate. 2 The removal of entry barriers allowed banks to enter new out-of-state markets, and provoked a tremendous change in the competitive landscape (Barth et al. 2004, Beck et al. 2010, Lewis and Pescetto 1996, DeYoung 2015). Owing to their competitive effects, these regulatory changes have attracted a great deal of attention among researchers interested in the finance-economic growth nexus, and the relationship between competition and bank lending (Strahan 2003, Jayaratne and Strahan 1997, Huang 2008, Rice and Strahan 2010). However, previous empirical studies have not yet focused on the relationship between banking competition and bank liquidity creation an essential bank output. The intuitive quiet life theory of Hicks (1935) and subsequently elaborated on in detail by Berger and Hannan (1998) suggests that competition is positively associated with higher bank output. I test these predictions by examining how bank liquidity creation responds to changes in the level of competition banks face using a natural experiment: The staggered deregulation of interstate bank branching regulations in the U.S. following the passing of the IBBEA My study bridges the gap between literatures on the impact of bank branching deregulations and the determinants of bank liquidity creation. Jayaratne and Strahan (1996) offer one of the early contributions to the bank branching deregulation literature. They use a generalized differencein-difference method to test the impact of intrastate branching deregulation on economic growth through the bank lending quality channel. Later, the same authors show in Jayaratne and Strahan (1998) that the U.S. banking industry became more efficient once restrictions on branching are lifted. Strahan (2003) uses bank deregulation as an empirical laboratory to test the impact of financial development on bank market structures and the real economy. More recently, Rice and Strahan (2010) use interstate branching deregulation as an instrument for credit competition to examine the effect on small firm finance. Beck et al. (2010) exploit cross-time and cross-sectional variation in bank deregulation to assess the effects on income distribution. Goetz et al. (2013), on 2 This process concluded with the Interstate Banking and Branching Efficiency Act of 1994 (IBBEA). 3 IBBEA still contains four provisions which enable states to restrict out-of-state branching by distorting their means of entry 2

3 the other hand, point out issues of the interstate branching deregulation. They find that exogenous increases in geographic diversity following the IBBEA 1994 reduce BHC valuations and increase insider lending. They also claim that this geographic diversity reduces loan quality and implies the agency problems. The literature about liquidity creation and bank output starts with the modern theory of financial intermediation which emphasizes that beside risk transformation, liquidity creation is the key role of banks (Bhattacharya and Thakor 1993, Greenbaum and Thakor 2007 and Diamond and Rajan 2001). This research is also motivated by the literature on the role of policy in affecting credit availability and bank liquidity. Loutskina and Strahan (2009), Paravisini (2008), Berger et al. (2014) and Kashyap and Stein (2000) suggest the change of banking-related policies have a significant impact on bank output by increasing market competition. Taken together these two distinct literatures yield unambiguous predictions for the effect of bank branching deregulation on liquidity creation. I depart from existing research in three respects. First, I focus on the recent intrastate branching reforms introduced between 1994 and 2003 following the IBBEA Apart from Rice and Strahan (2010), who examine the link between credit competition and small-firm finance, this episode of deregulation has not received much attention. Second, I use the bank liquidity creation measure developed by Berger and Bouwman (2009) to measure bank output. A key advantage of this variable over bank lending is that it simultaneously captures both bank assets and bank liabilities. Third, this paper is the first study to investigate the open question of whether the higher bank competition leads to greater bank liquidity creation and how this effect differs across different type of banks. Since liquidity creation is an important function of financial intermediation, this research about how deregulation affects bank liquidity creation is of interest to both practitioners, researchers, and regulators. Using a difference-in-difference estimator applied to bank-level panel data with 61,135 observations for 9,804 banks operating in 50 states in the U.S. from 1994 to 2003, I find that higher banking competition causes banks to increase liquidity creation. Removing all restrictions on branch expansions triggers an increase in abnormal liquidity creation as a percentage of gross total assets by percentage points. The result is mainly driven by community banks that play a central role in the U.S. banking market. 4 My result is supported by Kerr and Nanda (2009), Rice and Strahan (2010) and Kashyap and Stein (2000) who find higher competition correlates 4 Critchfield et al. (2004) and Hanc (2004) argue that maintaining the central role of community banks (banks with gross total assets less than 1 billion USD) reflects long-standing public policies to alleviate the concern about the concentration of economic power. 3

4 with increased provision of small firm finance which is funded mainly by small banks. For large banks this effect is only significant among banks that have previously been involved in M&A activity. The heterogeneous treatment effect accords with the arguments made by Strahan (2003) that bank deregulation leads to more M&A activity. When splitting banks by capitalization level, I find that only poorly capitalized banks whose equity capital ratio lies below the median equity capital ratio are significantly affected. This result is consistent with the findings of Berger et al. (2014) and supported by the financial fragility crowding out hypothesis by Gorton and Winton (2000). My results are robust when I run sensitivity tests using an alternative measure of bank liquidity creation, an alternative index of bank branching deregulation and splitting banks by holding company status. I also provide evidence that the impact of the difference in interstate branching restriction after passing the IBBEA 1994 is persistent in the long-run. To supplement the main research question about bank liquidity creation, I run additional tests to investigate the impact of bank branching deregulation on bank risk taking using the Z-score as the dependent variable. The motivation for this extension is to answer a natural question that once the nexus between branching deregulation and liquidity creation is found, whether the nexus between branching deregulation and bank risk taking - another key role of banks exists?? The answer from this study is that higher banking competition leads to lower bank risk taking which is in line with findings of Schaeck et al (2009) and Caminal and Matutes (2002). The most important assumption of difference-in-difference estimation is the existence of parallel trends. This implies that in the absence of treatment, the behavior of banks in deregulated states (the treatment group) and banks in non-deregulated states (the control group) is statistically similar. I verify this assumption by conducting t-tests following the approach suggested by Lemmon and Roberts (2010). Furthermore, establishing causality relies on interstate bank branch deregulation being exogenous. To alleviate this concerns, I use a logit model to verify that bank liquidity creation cannot predict the occurrence of interstate branching deregulation which support the exogeneity assumption. My study is based on a comprehensive set of bank-year data from different sources between 1994 and Call reports of U.S. commercial banks and details about U.S. bank location are collected from Federal Deposit Insurance Corporation (FDIC) and SNL. Information about the year of statelevel deregulation of restrictions on geographical expansion after the passage of IBBEA is collected from Rice and Strahan (2010). Regarding bank liquidity creation data, since using the same approach for calculating bank liquidity creation for U.S. commercial banks in the same period of study, I exploit the data set of Berger and Bouwman (2009). Other macro-economic variables are collected from the U.S. Census Bureau. 4

5 The rest of the paper is organized as follows. Section 2 analyses the institutional background of the interstate branching deregulation. I describe data in Section 3. Section 4 presents the methodology. Section 5 reports empirical result and Section 6 concludes. 2. Institutional Background 2.1. The U.S. Bank Branching Deregulation History of U.S. Bank Branching Deregulation Unlike most other deregulation at the national level such as telecommunications, railroads, trucking, and securities, bank branching deregulation occurred on a state-by-state basis (Kroszner and Strahan, 1999). For that reason, branching deregulation provides a greater source of time variation than law changes at the nation-wide level. Strahan (2003), Rice and Strahan (2010), Jayaratne and Strahan (1996) and Kroszner and Strahan (1999) provide an extensive summary bank branching deregulation in the US. (1) From 1927 to 1970: The period of restrictions on banks geographic expansion Before 1970, banks were prohibited from operating across state boundaries. This period is the so-called century of restriction on bank geographic expansion. White (1983) argues that the McFadden Act of 1927 imposed geographic restrictions on bank expansion to protect local banks from competition during the 20 th century. Through the Act the Supreme Court gave states authority over the regulation of local national bank branching activities 5. However, subsequent innovations in technology, financial services and improvements to the legal environment decreased the monopoly power of local banks and prevented them from taking advantages from geographic limitations (Strahan, 2003; Berger and Mester, 1999). During the 1930s, some deregulation of branching restrictions was permitted (e.g. Glass Steagall Act of 1933 permitted nationally chartered banks to open branches in the same areas as state chartered banks), although most states continued to enforce geographic restrictions until 1970s 6. Under these restrictions, banks formed multi-bank Holding Companies (MHBC) to expand geographic scope (Kroszner and Strahan 1999). However, the Douglas Amendment to the 1956 Bank Holding Company Act put an end to these practices by limiting MBHCs to acquiring out-of-state banks under the authorization of state regulators where the target bank was operating. (2) From 1970s to 1990: Intrastate bank branching deregulation 5 National banks mean banks with national charters. 6 Operating in other states either by opening new branches or M&A with other banks was prohibited throughout the US in that period. Even for intrastate branching, many states still restricted or banned it (Strahan, 2003). 5

6 From the 1970s through the 1990s, most U.S. states removed restrictions on intrastate branching through three types of refoms (Kroszner and Strahan, 1999): (1) The establishment of bank holding companies (BHCs); (2) The permission for BHCs to convert their subsidiaries into branches; (3) The acceptance of de novo branching. In most cases, branching by M&A occurred first and then not so long thereafter, unrestricted branching deregulation occurred. However even under new regulation of de novo branching deregulation, banks tend to enter new markets by buying existing banks or branches rather than opening new ones (Strahan, 2003). Only 12 states permitted full intrastate branching before However, by 1994, 38 states deregulated their restrictions on branching (Strahan, 2003). Figure 1 illustrates the year of intrastate bank branching deregulation states. (3) From 1994 to 2003: The Interstate Banking and Branching Efficiency Act of 1994 In 1994, the passage of the Riegle-Neal Interstate Banking and Branching Efficiency Act (IBBEA) removed all remaining federal restrictions on interstate expansion and permitted banks and bank holding companies to expand across state lines. After 3 years, in 1997, IBBEA completed this process by extending the legislation to all U.S. states. Despite the IBBEA sets out the conditions for branching at nationwide level, IBBEA still granted states the rights to set roadblocks which are the regulations to restrict interstate branching by four types of particular provisions including: (1) The minimum age of the target institution - How long a bank must have been in existence prior to its interstate acquisition or merger. This requirement cannot be set to be more than 5 years - IBBEA Section 102(a), codified at 12 USC 1831u(a)(5)(A). (2) De novo interstate branching The opening of new out-of-state branches only applied when states opted-in to this provision. (3) The acquisition of individual branches - An interstate merger transaction may involve the acquisition of a branch/a number of branches without the acquisition of the whole bank in the state permitting this kind of purchase - IBBEA, Section 102(a), codified at 12 USC 1831u(a)(2) - only applied when states opted-in to this provision. (4) A statewide deposit cap A limit on state deposit imposed on banks to avoid monopoly power in financial market. IBBEA, Section 102(a), codified at 12 USC Under these requirements, the cost of doing out-of-state banking increases and as a result, older and established in-state banks are protected (Johnson and Rice, 2008). However, the adoption of IBBEA differs between states and from time to time which implies the staggered nature of this deregulation. In 2003, Massachusetts and Ohio have the most open environment toward out-ofstate entry as they remove all four interstate branching restriction barriers and Arkansas, 6

7 Colorado and Montana seem to have the most restrictive market as they still use all four provisions to protect their intrastate bank and branches. [INSER TABLE 1] Real impacts of the interstate branching deregulation Strahan (2003) finds that the average growth in real per capita state income accelerated following branching and banking reforms and the growth effects are permanent. Investigating the impact of intrastate bank branch deregulation, Jayaratne and Strahan (1996) provide evidence that the relaxation of bank branching restriction from 1970s to 1990s lead to significant increase in the rates of real per capita growth in income and output of the economy. More recently, Huang (2008), Cetorelli and Gambera (2001), Freeman (2002), Claessens and Laeven (2005) also find the positive impact of bank deregulation on economic growth. Jayaratne and Strahan (1998) and Hubbard and Palia (1995) show that relaxing restrictions on the entry of banks from other states and the branching of banks within states led to higher efficiency, lower cost of productivity and more effective credit allocation. Since banks are an important source of capital for firms, a literature review on the impact of bank deregulation on firm finance also helps us to see the channel of deregulation on bank output. Using differencein-difference specification, Beck et al. (2010) suggest that deregulation materially tightened the distribution of income since it boosts the income of poor entities (lower part of income distribution) but has very little impact on incomes of rich entities (above the median part of income distribution). Strahan (2003) also finds that the level and growth of entrepreneurial activity increased following banking deregulation. The channel in the Beck et al. (2004) and Strahan (2003) papers is that once deregulation intensified bank competition, creditconstrained firms gain access to credit more easily. Cetorelli and Strahan (2006) and Ogura (2012) also suggest a positive association between competition and credit availability. In contrast, Petersen and Rajan (1995) reinforce the fragility channel view that indicates higher competition leads to lower loan supply. Other papers reject both two views and find either no or a very weak relationship between any credit quantity measures and bank competition but significant evidence for loan rates, credit quality and bank competition. Testing such predictions in the US market, Rice and Strahan (2010) found that bank branching deregulation leads to lower loan rates; Jayaratne and Strahan (1997) observed improvements in loan quality after branch reform. Greenwood and Jovanovic (1990), and Banerjee and Newman (1993) argue that if the banking industry is unregulated, monopolistic banks will 7

8 gain an unfair advantage and hurt the poor. The stories of the winner and loser from bank branching deregulations are also shown by Beck et al. (2010). Regarding the impact on bank structure, Strahan (2003) suggests that while the acquisitions in U.S. banking market increased sharply following interstate banking deregulation, there was no significant increase in bank acquisition following branching deregulation. Banks seem to prefer to expand by purchasing branches of existing banks rather than by acquiring all the branches of whole banks. Calem (1994) argues that a number of community banks are acquired as branches into large BHC after branching deregulation. Examining the period , Berger and Mester (2003) and Berger, Demsetz and Strahan (1999) find the increase in profit productivity and the decline in cost productivity, especially for banks engaging in mergers many of which were allowed by nationwide branching. Combining these two stories about the impact of nationwide branching on M&A bank groups helps me form the subsequent regressions to investigate how merger & acquisition status affects the ability of banks to create liquidity under the impact of interstate bank branching deregulation Bank Liquidity Creation Banks create liquidity on the balance sheet by financing relatively illiquid assets such as business loans and long term credit with relatively liquid liabilities such as customer deposits (e.g., Bryant 1980, Diamond and Dybvig 1983, Boyd and Prescott 1986), and off the balance sheet through loan commitments and other liquidity claims (e.g., Boot et al., 1993; Holmstrom and Tirole, 1998; Kashyap et al., 2002). Bhattacharya and Thakor (1993) document that liquidity transformation plays an important role in the economy in the scope of Qualitative asset transformation function (QAT). More recently, Greenbaum and Thakor (2007), in the respect of financial intermediation theory, explain banks as liquidity creators when they provide loans from deposits of public and equity of banks shareholders. Strahan (2008) claims that the channel of liquidity creation from the asset side and liability side. On the liability side, the most importance item is deposits. Transactions deposits allow account holders to withdraw cash on demand from financial institutions and the main characters of deposits are low in risk and high in quality. However, at the same time, since not all customers tend to withdraw their deposits, banks can therefore use the average balance of all deposit accounts to invest in loans to businesses and households with higher yield. This business model is also called asset transformation (Strahan, 2008). On the asset 8

9 side, banks also create liquidity by issuing credit lines such as loan commitments, which enables customers to receive cash by drawing down the line with pre-arranged rate of interest. Although the measure for bank liquidity creation is not a universal term until now, there are several models suggested by researchers. Formal models of banks as liquidity creators were developed by Bryant (1980) and Diamond and Dybvig (1983). In those models, depositors can suffer interim liquidity shocks, so being able to hold liquid (demand) deposit claims improves welfare. To best of my knowledge, there are three models suggested by Berger and Bouwman (2009), Deep and Schaefer (2004) and Brunnermeier et al. (2011) and Bai et al. (2013) that can give us deeper understanding of how banks create liquidity in different perspectives. Based on maturity transformation functions in on-balance sheet activities, Deep and Schaefer (2004) construct a method to quantify liquidity creation, which is the so-called Liquidity Transformation Gap (LT Gap). The LT Gap calculates the degree to which a bank transforms liabilities into assets with higher maturity. In my view, this methodology would be very helpful for research studies considering interest rate risk and liquidity risk. The next prominent research providing methodology to quantify liquidity creation is the the study of Berger and Bouwman (2009) where the authors develop four measures of liquidity creation based on either categories or maturity of assets, liabilities, equities and off-balance sheet issues. This method analyzes the ease, cost and time it takes to transfer each balance sheet item into cash from the owners perspectives of the assets and liabilities. More recently, Brunnermeier et al. (2011) initiate and Bai et al. (2013) develop a liquidity mismatch index (LMI), which is conceptually similar to the Berger and Bouwman (2009) method and is suggested to be useful for macroprudential purposes. In this study, I deploy the measurement suggested by Berger and Bouwman (2009) because it not only enables researchers take into account the impact of all items in balance sheet on bank liquidity creation but it also has advantages on data availability and data accuracy Deregulation, Bank competition and Bank liquidity creation One may argue that to test the impact of bank competition on bank liquidity creation, we can directly use the measure for competition such as Panzar-Rosse H-statistic 7, Lerner Index 8, 7 See: Molyneux et al. (1994), Schaeck et al. (2009) and Claessens and Laeven ( 2004) 8 See: Carbo-Valverde et al. (2009); and Fungacova and Weill (2013) 9

10 Boone indicator 9 or market concentration index 10. However, the key issue of using competition and concentration indexes is that these indexes are potentially endogenous to bank characteristics (Gonzales, 2005). Hence, using the bank branching deregulation which is an exogenous shock with respect to bank liquidity creation can remove the endogeneity problems. Theory suggests different mechanisms in which competition can affect liquidity creation. On the one hand, the efficiency channel implies that the competition tends to force financial institutions to invest in financial innovation and improve efficiency and thus, produce more liquidity (Boot and Thakor, 2000; Black and Strahan, 2002; Laeven, Levine, and Michalopoulos, 2015). Furthermore, the pricing channel suggests that when states remove entry restrictions, competition increase, which will influence the pricing policy of a bank. To attract more customers, the banks have to decrease offer rates on the asset side (see Carbo- Valverde et al. 2009, Rice and Strahan 2010) and increase rates on the liabilities side. As a result, the volume of both assets and liabilities will increase which indicates a positive relationship between bank branch deregulation and liquidity creation. On the other hand, according to the fragility crowding out effect which is reinforced by Petersen and Rajan (2005), bank competition correlates with higher fragility in bank asset liability structure by reducing bank profits and stretching out capital buffer. As a result, banks reduce liquidity creation by issuing less loans and accepting less deposit to avoid bank run threats. Empirical studies documents heterogeneous relationship between bank competition and bank outputs depending on bank sizes. Kashyap and Stein (2000) argue that only small banks lending is impacted by policy changing in non-crisis periods. Moreover, on the demand side, Rice and Strahan (2010); Kerr and Nanda (2009); Black and Strahan (2002) and Cetorelli and Strahan (2006) suggest that banking and branching deregulation enables credit-constrained firms access to finance, enabling them to enter to the market. Berger and Udell (2002); Cole et al. (2004) and Diamond and Rajan, (2000, 2001) provide consistent evidence that community banks are the main source for small firm finance. In this study, I empirically test if the interstate branching deregulation improves bank liquidity creation and provide a clear explanation for which transmission the effect comes from. 9 See: Boone, Griffith and Harrison (2005) 10 See: Black and Strahan (2002) and Beck et al., (2006) 10

11 3. Data and research methodology 3.1. Identification strategy At the heart of the identification strategy lies the question how to identify the treatment group and the control group. In this paper, I define the treatment group to be banks which experience a deregulation event and the control groups are banks which do not experience a deregulation event. To run the difference-in-differences estimation, I construct the baseline regression using ordinary least squares as follows: Y i,k,t = α + β.bdi k,t + γ.x i,k,t-1 + FeB + FeT+ ε k,t (2) where i refers to bank i, k indexes state j and t refers to year t. Y i,k,t includes two dependent variables Bank liquidity creation and bank risk taking: LCR i,k,t - Liquidity creation ratio is equal to the catfat liquidity creation value calculated by Berger and Bouwman (2009) method divided by gross total asset. I use the liquidity creation ratio instead of the notional value because it makes the dependent variable comparable across banks. For reason of brevity, I relegate the calculation of bank liquidity creation in the Appendix 1. Z-score i,k,t is calcuated as follows: ROA + E/TA SD(ROA) where ROA and E/TA are ratios of mean return on asset and mean equity-to-total asset of each bank during the ten years of the study; SD(ROA) is the standard deviation of return on assets calculated for the period of this study to replicate volatility in earnings. I use logarithm of the Z-score to remove outliers and avoid the skewness and non-normality of this variable. BDI k,t - Branching deregulation indicator is an interaction between treated banks and a dummy indicating the post deregulation year. BDI equals to 1 indicates that banks operate in the deregulated state after the deregulation year and 0 otherwise. X i,k,t-1 - Vector of control variables for bank characteristics contains bank-specific variables: X i,k,t-1 = (SIZE j,k,t-1, ECAP j,k,t-1, RISK j,k,t-1, ROE j,k,t-1, BANK HOLDING COMPANY STATUS j,k,t-1) SIZE is the logarithm of total assets; ECAP is the Equity Capital ratio; RISK refers to bank risks which contains two indicators Z-SCORE (logarithm) and Equity Volatility; ROE is return on equity which controls for profitability; BANK HOLDING COMPANY STATUS is dummy variable which equals 1 if the banks is a bank holding company;. All control variables are lagged one period to mitigate any possible endogeneity problem. 11

12 To avoid the multi-collinearity between Z-SCORE (logarithm), the Equity Volatility and the Capital Ratio, following Berger and Bouwman (2009), the two risk indicators are orthogonalized. First, I estimate the residuals from the two regressions: The first one is between logarithm of Z-core, equity capital ratio and all other control variables. The second one is between equity volatility, equity capital ratio and all other control variables. Second, the error terms are included in all main regressions to represent the variation of two risk indicators not captured by the capital ratio. FeB - Bank fixed effect is used to control for unobserved time-invariant heterogeneity at the bank level. Second, FeT Year fixed effect is used to control for time-varying forces that might drive changes in the dependent variable. In the spirit of Bertrand et al. (2004), I cluster the estimated standard errors ε k,t at the state-year level as the implementation of the IBBEA affects all banks located in the same state at the same year Data I collect yearly data for commercial banks in the US market from 1994 to 2003 (the year of passaging IBBEA). Data for liquidity creation is from Berger and Bouwman (2009). For the interstate branching deregulation indicator, the effective date in each state is collected from Rice and Strahan (2010). As the deregulation variable varies on the state-level, I extract the bank location data from Federal Deposit Insurance Corporation and then, match the bank branch deregulation dummy variable to the bank location data to obtain the panel data breakdown at bank-level. For the bank characteristics, the data sources are from SNL, Bank Call Reports - Federal Deposit Insurance Corporation and Bank Holding Company Y-9 reports - Federal Reserve will be used. It should be noted that this study only investigates the single-state banks in the U.S. market because of two main reasons. First, bank liquidity creation data is only available at the bank level. Second, the main objective of this project is investigating how does the relaxation of geographic restrictiveness of bank expansion impacts banks output in this state. Hence, using the single-state banks does not lead to any biased results. My data sample includes all single-state banks in all 51 U.S. states in the 10 years from 1994 to To achieve a clean data set and ensure that my data set only contains viable commercial banks, following Berger and Bowman (2009), I exclude a bank if it has (1) no deposits; (2) zero or negative equity capital in the current or lagged year; (3) gross total assets below $25 million. All the data selection and cleaning process result in a sample of 72,169 bank-year observations: 1,218 for large banks; 2,198 for medium banks; and 68,752 for small banks. However, as I lag each of the control variables by one period, the number of observations in the regressions is 61,135 in total, with 985 for large banks, 1,900 for medium banks and 58,250 for small banks. 12

13 [INSERT TABLE 2] As can be seen from Table 3, the two risk indicators and capital ratio are highly correlated (ρ>83%). In order to remove the multi-collinearity concerns, the Z-SCORE (logarithm) and Equity Volatility are orthogonalized 11. [INSERT TABLE 3] 4. Diagnostic tests and the validity of the natural experiment 4.1. Exogeneity of the regulatory barriers with respect to liquidity creation Kroszner and Strahan (1999) claim that the timing of deregulation is potentially endogenous with respect to bank output when they investigate the stories of the winners (small firms and large banks) and losers (insurance firms and small banks) from bank deregulation during the 1970s and 1980s. The groups that can benefit from deregulation will lobby to accelerate the reform, whereas groups that can be harmed by the deregulation tend to lobby to delay it. Hence, if the dependent variable is statewide output, the intrastate deregulation could be an endogenous regressor and the deregulation indicator will need to be instrumented by another exogenous variable. I argue that the deregulation is exogenous with respect to the dependent variable in this study. As bank liquidity creation is not an overall economy-wide output and the deregulation is interstate in nature, it is implausible that the decision of state-level authorities is affected by out-of-state interests. Furthermore, there is no evidence showing that the key motivation of interstate bank branching deregulation is the health of the banking systems. Other contemporary studies have also used bank branching deregulation as an exogenous shock. For example, Beck et al. (2009) exploit the exogenous cross-state, cross-year variation in timing of branch deregulation to establish causality between bank deregulation and income distribution. Jayaratne and Strahan (1996) use difference-in-difference estimations to examine the impact of bank branching deregulation on economic growth. Rice and Strahan (2010) employ interstate bank branching deregulation as an instrument to test the impact of credit competition on small-firm finance. To empirically test this exogenity assumption, I employ a logit model to show that bank liquidity creation is not driving the changes in regulatory barrier level to interstate branching. The loglikelihood function for the model is: 11 First, I estimate the residuals from the two regressions: The first one is between the logarithm of the Z- core, the equity capital ratio and all other control variables. The second one is between equity volatility, equity capital ratio and all other control variables. Second, the error terms are included in all main regression to represent the variation of two risk indicators not captured by the capital ratio. 13

14 LnL = t=1 T i=1 n{p(i, t) ln[f(β X(i, t))] + (1 P(i, t)) ln[1 F(β X(i, t))]} (1) where: P(i,t) is a dummy variable that takes on the value one when state i deregulated at the time t, and zero otherwise. β is the vector of coefficients, the explanatory variables are denoted by X(i,t), and F(β X(I,t)) is the cumulative distribution function evaluated at β X(i,t). The explanatory variable is either bank liquidity creation ratio or Ln Z-score. I cluster standard errors at the stateyear level to correct for possible time-varying correlation in unobservable factors affecting banks in the same state each year. (Bertrand et al., 2004). The results in Table 2 substantiate my argument. I find evidence that neither liquidity creation nor bank risk can systematically predict whether interstate branching deregulation is in force. [INSERT TABLE 4] 4.2. Parallel Trends The parallel trends assumption in the difference-in-difference method implies that in the absence of treatment, the behavior of banks in deregulated states and banks in non-deregulated states is statistically similar. Following Lemmon and Roberts (2010), I conduct t-tests to verify that liquidity creation within the treatment and control groups behave similarly in the period t-1, t-2 and t-3 and report the results in Table 5. Figure 1 visualizes the growth rates of liquidity creation ratio and bank risk and confirms the validity of parallel assumption. [INSERT TABLE 5] [INSERT FIGURE 1] 5. Results and discussion 5.1. Bank Competition and liquidity creation Table 6 reports the OLS estimates of my baseline difference-in-difference regressions. In column 1, the magnitude of the coefficient of the BRI is 0.11, suggests that the average treatment effect is approximately 0.11 percentage points implying banks in deregulated states produce 0.11 percentage points LC ratio higher than banks that locate in the states using 4 methods of branching restrictions. This effect is statistically significant at the 1% level. Next, I include additional covariates into the regression to capture other potential determinant of liquidity creation. The literature suggests that bank output is distinctive among bank sizes (see, for example: Berger and Bouwman, 2009; Berger et al. 2014, Jayratne and Wolken, 1999). The estimates reported in Column 2 to 6 suggest positive association between bank size and bank 14

15 liquidity creation. A 1 percentage point increase in gross total asset is associated with 0.02 percentage points increase in liquidity creation ratio. Regarding the capital ratio, I find consistent results with the study of Berger and Bouwnman (2009) on small banks: higher capital leads to higher liquidity creation. This leads me to consider subsample tests later on to see if the results are mainly driven by small community banks. Bank risk taking is included in the model from column 4 onwards. These two indicators, Z-score and equity volatility, consistently suggest that the more stable banks create higher liquidity. In Column 5, I further control for bank profitability. More profitable banks can create approximately 4% higher liquidity creation relative to gross total asset. Finally, as contemporary literature suggests that multibank holding companies are the one that can involve in branching expansion the most, I include a dummy variable for MBHCs and document that MBHCs are associated with higher liquidity creation. [INSERT TABLE 6] Table 7 reports the heterogeneous effect of the deregulation on liquidity creation across different subsamples. Panel A on bank sizes shows that only small banks and medium banks are statistically significantly affected by the relaxation of branching expansion. To verify that the magnitudes of these coefficients are statistically different from each other across the subsamples, I conduct Chow tests. The result of the Chow test rejects the equality of coefficients across my subsamples. One could ask why only small and medium banks are affected by a higher level of market competition. Kashyap and Stein (2000) find that small banks are more sensitive to the policy shocks and that only small banks lending is impacted by policy changing in non-crisis periods. I suspect that, from the demand side, banking and branching deregulations enable creditconstrained firms access to finance and enter the market. A number of empirical studies suggests that small local banks may more often function as community banks which have more soft information about small local firm will take the advantages rather than large banks which prefer transparent firms (Berger and Udell, 2002; Cole et al., 2004; Diamond and Rajan, 2000, 2001), resulting in abnormal liquidity creation in the states relax their restrictions on branching expansion. This argument is in line with Rice and Strahan (2010); Kerr and Nanda (2009); Black and Strahan (2002), and Cetorelli and Strahan (2006). Panel B of Table 7 examines the effect of bank merger on bank liquidity creation in the period of interstate branching deregulation. The significantly positive nexus between the interaction term Merger x BRI implies that banks that are involved in M&A during the deregulation period create higher bank liquidity creation. The subsample analysis once again confirms that my findings are 15

16 driven by small and medium banks. When examining the characteristics of banks that create more liquidity, Berger and Bouwman (2009) also find that large banks which are engaged in M&A experienced the most noticeable increase in liquidity creation in the period of The result that higher impact found in M&A banks is also in line with the finding of Strahan (2003) which suggests that bank deregulation leads to higher M&A activities. Berger and Mester (2003) and Berger, Demsetz and Strahan (1999) argue that M&A leads to higher bank output as it increases profit and decreases cost productivity. Panel C reports the results when I split the sample by capitalization level. Poorer capitalized banks include banks with capital ratio lie below median capital ratio of the whole sample. The results suggest that better capitalized banks create higher liquidity in competitive markets. The negative coefficient between branching deregulation and liquidity creation found in poor capitalized banks for both large and small and medium banks suggest that more fragile banks are under greater pressure to attract customers under more competitive markets. [INSERT TABLE 7] 5.2. Bank competition and risk taking The evidence of better capitalized banks create higher liquidity under more competitive market also sheds the light on the effect of competition on bank stability under non-crisis period. When examining the benefits of branching deregulation, Jayaratne and Strahan (1997) wrote:...banks' loan losses and operating costs fell sharply following the state initiatives, and that the cost declines were largely passed along to bank borrowers in the form of lower loan rates. The authors argue that these efficiency gains arose because better performing banks were able to expand their market share once geographic restraints were erased... As the main findings about bank liquidity creation imply that banks create higher output following the branching deregulation, one could wonder if that higher output is the tradeoff for higher risk taking. Employing the Z-score as a dependent variable, I document that deregulation benefits financial stability (Table 8). This result is in line with empirical findings of Schaeck et al. (2009) and theoretical predictions of Caminal and Matutes (2002) but contradicts with Dick and Lehnert (2010) who argue that U.S. bank deregulation was associated with a rise in personal bankruptcy rate. In Jayratne and Strahan (1997), they find that higher competition enforced banks to do more intensively screening, thus, enhanced financial stability. [INSERT TABLE 8] 16

17 5.3. Transmission Mechanism Next, I explore whether the observed changes in liquidity creation and risk taking are indeed driven by higher competition. I adopt a two-stage least squares approach to tie together financial market competition and bank outputs. In the first stage, I estimate the equation: Competition k,t = α + β.bdi k,t + γ.x i,k,t-1 + FeB + FeS+ ε k,t (3) where all variables are as in equation (2) except Competition kt which is logarithm of local deposit Herfindahl-Hirschman index. In the second stage, I estimate LCR k,t = α + β.competition k,t + γ.x i,k,t-1 + FeB + FeS+ ε k,t (4a) Risk k,t = α + β. Competition k,t + γ.x i,k,t-1 + FeB + FeS+ ε k,t (4b) using the bank-level data. As the deregulation shocks affect all banks in state in the same year, I again cluster error terms at state*year level.the essence of this instrumental variable strategy is to exploit variation in bank competition caused by the deregulation shock and establish the direct estimation for the effect of such competition on bank outputs. The first-stage results of this test are reported in column 1 of Table 9. I find clear evidence that the deregulation caused significant decrease in market concentration, or, to put it differently, increase in bank competition. Getting results from equation (1), I regress liquidity creation and risk taking (measured by Ln Z-score) on the predicted logarithm of HHI indexes and find that higher competition indeed results in higher bank liquidity creation and lower bank risk taking. The diagnostic tests confirm the validity of the instrument in explaining bank competition and the Kleibergen-Paap F-statistic which comfortably exceeds the critial value of 10 also confirms the instrument in the first stage is relevant. [INSERT TABLE 9] 5.4. Long-run effects I have so far focused on the short-run effect of interstate branching deregulation on liquidity creation and bank risk taking. To investigate whether the effects persist for more than one year, I now examine the association of the change in restriction of interstate branching deregulation and bank liquidity creation in the long run by using the lead of dependent variables by 2, 3, 4 and 5 years. Table 10 shows that the effect of interstate branching deregulation on bank liquidity creation is still observed in subsequent years but the magnitude of the coefficients decreased in comparison with that of short-run. This finding is intuitive as the branching deregulation enhances banking 17

18 competition permanently, and therefore increases bank output. However, when it comes to the long run effects in bank risk taking, Panel B suggests that competition policies cannot maintain the positive effect on bank stability in the long run. Interestingly, the sign of the magnitude even suggests that banks could take higher risks in the long run in more competitive markets. [INSERT TABLE 10] 6. Robustness tests 6.1. Excluding off balance sheet items One of the threats to the findings is that the effects mainly come from off balance sheet items which do not reflect the traditional banking activities. To rule out this concern, I use the narrow liquidity creation ratio measure which excludes off balance sheet items. Relying on the liquidity creation measure from only on balance sheet activities in my research period makes sense because banks tend to expand their off balance sheet activities normally in unstable economic periods (Thakor, 2005). The results in Table 11 imply that even excluding off balance sheet activities, small and medium banks located in more competitive market still generate more liquidity compared to the counterfactual. [INSERT TABLE 11] 6.2. Using different indicator for bank competition In the main tests, I set up the branching deregulation index as an interaction term between deregulated state and dummies of post deregulation year. In this section, I employ the branching restrictiveness index as in Rice and Strahan (2010) which range from 0 to 4 with 0 for the most competitive market where no branching constrain is established and 4 for the least competitive one where all four restriction methods are used. The results in Table 12 are still robust for both liquidity creation and bank risk taking. For all banks, for every provision for interstate branching restriction a state imposed, liquidity creation as percentage of gross total assets of bank operating in that state decreases by 0.02 percentage points. Therefore, banks in states least restrictive (states whose restrictiveness index equals to 0) have on average 0.02*4 = 0.08 percentage points higher liquidity creation ratio than banks in states most restrictive (states whose the restriction index equals 4). The magnitude of the effect is nearly equal to what I found in the main results which is 8-10%. [INSERT TABLE 12] 18

19 6.3. Demand conditions Another threat to my identification is that the higher liquidity creation and lower bank risk observed in deregulated states come from better economic conditions in deregulated states compared to non-deregulated. Although using year and bank fixed effect in the main regressions makes it unlikely that regional macroeconomic and demand conditions can drive the key findings, another test directly examines these effect is nessecary. Collecting data from the County Business Pattern website, I further control for the business dynamics in firm establishment, establishment exit rate and job creation to alleviate the concerns regarding difference in economic conditions. The interaction term between deregulation dummies and three demand variables illustrate the differential in demand driving forces between treated and control banks. Across all tests showed in Table 13, regardless of the demand effects that I take into account, the main coefficients of interest remain robust and consistent with the main tests. Together, these tests indicate that there were not differential cross-time shocks to demand between treated and control group caused the abnormal changes in liquidity creation and bank risk. [INSERT TABLE 13] 6.4. Placebo tests Valid identification rests upon the exogeneity of the deregulation shock, and that banks did not change their behavior in anticipation of the increase in competition. This assumption is fortunately testable. I create a dummy variable, Placebo kt which take the value 1 for 2 years before the official deregulation year in deregulated states and 0 otherwise. If anticipation effects exist, I could be able to document a similar economic magnitude and statistical significant on the variable Placebo kt. However, this is not the case as the Placebo enters the regression insignificantly while BRI still remains significantly positive with the magnitude similar with what we observed from main findings. The second falsification test provides insights into whether my findings are driven by spurious post-treatment trends. I exploit the fact that in my dataset, there are 8 out of 50 states that use all four restrictions in branching expansion during the research period. These states are Arkansas, Colorado, Iowa, Kansas, Mississippi, Montana, and Oklahoma. In these states, Arkansas and Mississippi are two states that locate relatively close to a big state, Texas, where branching deregulation is in implementation from I construct a sample containing only observations from these 8 states and define a placebo treatment group TG_fal i which equals 1 19

20 if a bank is located in Arkansas and Mississipi. Next, I include a fal_year t that equals 1 for the years from 1999 onwards and 0 otherwise. Panel B confirms that the effects of bank competition on bank liquidity creation and bank risk are indeed only observable in the deregulated states. [INSERT TABLE 14] 7. Conclusion This paper leverages the staggered nature of the interstate branching reforms following the IBBEA 1994 as a natural experiment to examine the impact of banking competition on bank liquidity creation. The main research questions are motivated by bridging the literatures on the impact of bank branching deregulations (e.g. Strahan 2003, Jayaratne and Strahan 1996, 1998, Beck et al and Rice and Strahan 2010), U.S. banking market structure (Critchfield et al. 2004, Hanc 2004 and Jones and Critchfield 2005) and bank liquidity creation (Berger and Bouwman 2009, 2012, Berger et al. 2014, Loutskina and Strahan 2009 and Kashyap and Stein 2000). A key contribution of my study is therefore to provide the novel evidence that the recent interstate branching deregulation positively affects bank liquidity creation. Using the difference-in-difference estimation strategy, I document the key findings as follows: Firstly, the interstate branching deregulation after the passage of IBBEA 1994 results in greater bank liquidity creation. Overall, the treatment groups liquidity creation ratio increases by percentage points. Secondly, this effect is mainly driven by small and medium banks which are the main source for financing for small firms in the economy. Thirdly, consistent with the development of the U.S. banking market structure during the decade of interstate branching deregulation, my subsample analyses suggest that the effect is stronger for banks which have M&A history and banks which have a better capital structure. I also provide evidence that the impact of the difference in interstate branching restrictions after passing the IBBEA 1994 is permanent and can be observed over longer periods of time. My results remain intact across a variety of robustness tests. In addition, I extend the topic by looking at the impact of branching deregulation on bank risk taking given that aside from liquidity creation, risk transformation is key function of a bank. The result suggest that the interstate branching deregulation reduces bank risk-taking, adding more evidence on the current studies about banking competition, bank risk taking and financial stability (e.g. Jayaratne and Strahan 1997, Schaeck et al and Caminal and Matutes 2002). From a policy perspective, my findings suggest the possible justification for policy makers to introduce the bank branching deregulation in the U.S. During normal times, given that liquidity 20

21 creation is an important bank output which contribute to the economic growth (Bhattacharya and Thakor 1993), the introduction of competition regulatory laws can boost economic productivity and output. However, I temper my summary by pointing out some inevitable limitations as follows: First, although liquidity creation is the key role of banks and has great influence on economic growth, until now there is no universal measure for it. Although the use of liquidity creation measurement suggested by Berger and Bouwman (2009) seems to have advantages, there are still some reservations when using it regarding the simple weights assigned for each asset and liability items based on either categories or maturities. This method also has disadvantages since it cannot capture some important features such as the remaining maturity and the quality of assets and liabilities. Second, the result suggests a positive impact of competition on liquidity creation. A natural question could arise is that, how can we evaluate whether this impact is good? Under the non-crisis period, we can see this is a good sign as it increases bank output, which, in turn, could contribute to the economic growth. Also, the effect on Z-score suggests that competition policies reinforce banks to perform better and take lower risk. However, we cannot have the conclusion for the crisis periods from this study. Based on these limitations above, I suggest some recommendations for future research. First, to dig deeper to better understand the driving forces behind the increases in liquidity creation, we can extend this study by testing the impact of interstate branching deregulation separately for the asset-side, liability-side and off-balance sheet components. This test is also proposed in Berger et al. (2016). Furthermore, to examine more deeply about the quality of bank liquidity creation, once the data on liquidity risk ratios suggested by Basel III such as Net Stable Funding Ratio or Liquidity Coverage Ratio are available for longer period, on could test the impact of branching deregulation on these ratios. Secondly, regarding the impact of higher bank competition on bank liquidity creation in other periods, especially for the crisis time, we can expand the period of this study and employ the other competition indexes such as H-statistic, Lerner Index and Boone indicator, and use deregulations as the instrumental variables. Another suggestion from the second limitation is that examining the finance-growth nexus together with competition bank output in periods that include both crisis and non-crisis time and try to find out whether the effect is changed in different business cycles. 21

22 REFERENCE LIST Acharya, V.V., and Naqvi, H. (2012). The seeds of a crisis: A theory of bank liquidity and risk taking over the business cycle, Journal of Financial Economics, Vol. 106, No.2, pp Aggarwal, R., and Jacques, K. T. (2001). The impact of FDICIA and prompt corrective action on bank capital and risk: Estimates using a simultaneous equations model, Journal of Banking and Finance, Vol. 25,pp Allen, F., and Gale, D. (2004) Competition and Financial Stability, Journal of Money, Credit and Banking, Vol. 36, pp Allen, F., and Santomero, A. M. (1998) The theory of financial intermediation, Journal of Banking and Finance, Vol. 21, pp Angelini, P. and Cetorelli, N. (2003). The Effects of Regulatory Reform on Competition in the Banking Industry, Journal of Money, Credit and Banking, Vol. 35, No. 5, pp Avraham, D., Selvaggi P., Vickery, J. (2012). A Structural View of Bank Holding Companies Federal Reserve Bank of New York Economic Policy Review, Vol.18, No. 2, pp Bai, J., Krishnamurthy, A, and Weymuller, C. H. (2013). Measuring liquidity mismatch in the banking sector, Working Paper, Northwestern University. Banerjee, A. V. and Newman, A. F. (1993). Occupational choice and the process of development, Journal of Political Economy, Vol.101, pp Barth, J. R., Caprio, G. Jr. and Levine, R. (2004). Bank regulation and supervision: What works best? Journal of Financial Intermediation, Vol. 13, No. 2, pp Basel Committee on Banking Supervision (BCBS) (2010). Basel III: International framework for liquidity risk measurement, standards and monitoring, Bank for International Settlements, December. Beck, T., Demirgüc-Kunt, A., and Levine R. (2006). Bank Concentration, Competition, and Crises: First Results. Journal of Banking and Finance, Vol. 30, pp Beck, T. and Levine, R. and Levkov, A. (2010). Big Bad Banks? The Winners and Losers from Bank Deregulation in the United States, Journal of Finance, Vol. 65, No.5, pp Berger, A. N. and Bouwman, C. H. S. (2009). Bank Liquidity Creation Review of Financial Studies, Vol. 22, pp Berger, A. N. and Bouwman, C. H. S. (2012). Bank Liquidity Creation, Monetary Policy, and Financial Crises. Working Paper, Wharton Financial Institutions Center. 22

23 Berger, A. N. and Bouwman, C. H. S., Kick, T., and Schaeck, K. (2014). Bank risk taking and liquidity creation following regulatory interventions and capital support, Available at SSRN: Accessed 2 th August Berger, A.N, Demsetz, R.S., Strahan, P.E., (1999). The consolidation of the financial services industry: causes, consequences, and implications for the future, Journal of Banking Finance, Vol. 23, pp Berger, A.N., Hannan, T.H., (1998). The efficiency cost of market power in the banking industry; a test of the quiet life and related hypotheses. Review of Economics and Statistics, Vol. 80, pp Berger A. N., Mester, L. J. (2003). Explaining the dramatic changes in performance of U.S. banks: technical change, deregulation, and dynamic changes in competition, Journal of Financial Intermediation, Vol. 12, pp Berger, A. N., Miller, N., Petersen, M., Rajan, R. and Stein, J. (2005), Does function follow organizational form? Evidence from the lending practices of large and small banks, Journal of Financial Economics, Vol. 76, pp Berger, A. N., and Udell, G.F. (2002) Small business credit availability and relationship lending: The importance of bank organisational structure, The Economic Journal, Vol. 112, No. 477, pp Bernanke, B. S., (1983), Nonmonetary effects of the financial crisis in propagation of the Great Depression, American Economic Review, Vol. 73, pp Bernanke, B. S., and Blinder, A. (1988), Credit, money and aggregate demand, American Economic Review, Vol. 82, pp Bertrand, M., Duflo, E., & Mullainathan, S. (2004). How Much Should We Trust Difference in Differences Estimates? Quarterly Journal of Economics 119, pp Bhattacharya, S., and Thakor, A. V. (1993). Contemporary Banking Theory, Journal of Financial Intermediation, Vol.3, pp Black, S., and Strahan, P. E. (2002), Entrepreneurship and bank credit availability, Journal of Finance, Vol 57, pp Bikker, J.A., Spierdijk L., Finnie P., (2006), The Impact of Bank Size on Market Power, Working Paper, No. 120, De Nederlandsche Bank. Boone, J., Griffith, R. and Harrison, R. (2005). Measuring competition. Working Paper Series, AIM Research. 23

24 Boot, W. A., Greenbaum, S. I. (1992). Bank regulation, reputation and rents: theory and policy implications. In: Mayer, Colin, Vives, Xavier Eds., Capital Markets and Financial Intermediation. Cambridge University Press, Cambridge, UK, pp Boot, W. A., Greenbaum, S. I. and Thakor, A. V. (1993). Reputation and Discretion in Financial Contracting, American Economic Review, Vol. 83, pp Boustanifar, H. (2014). Finance as a Barrier to Employment: Evidence from U.S. Banking Deregulation. Journal of Banking and Finance 46, pp Boyd, J. H., Graham, S. L. and Hewitt, S. R. (1993). Bank Holding Company Mergers with Nonbank Financial Firms on the Risk of Failure. Journal of Banking and Finance, Vol. 17, pp Boyd, J. H. and Prescott, E. (1986). Financial Intermediary Coalitions, Journal of Economic Theory, Vol. 38, pp Bryant, J. (1980). A Model of Reserves, Bank Runs, and Deposit Insurance, Journal of Banking and Finance, Vol. 4, pp Brunnermeier, M. K., and Pedersen, L.H. (2009) "Market liquidity and funding liquidity." Review of Financial studies, Vol. 22, No.6, pp Caminal, R., and Matutes, C. (2002). Market Power and Bank Failures. International Journal of Industrial Organisation, Vol. 20, pp Calderon, C. and Schaeck, K. (2015). The effects of government interventions in the financial sector on banking competition and the evolution of zombie banks, Journal of Financial and Quantitative Analysis, forthcoming. Calem, P., (1994). The impact of geographic deregulation on small banks, Federal Reserve Bank of Philadelphia Business Review, Philadelphia, PA, pp Carbo-Valverde, S., Humphrey, D., Maudos, J., and Molyneux, P. (2009). Cross-country comparisons of competition and pricing power in European banking, Journal of International Money and Finance, Vol. 28, No. 1, pp Cetorelli, N., and Gambera, M. (2001), Bank structure, financial dependence and growth: International evidence from industrial data, Journal of Finance, Vol. 56, No.2, pp Cetorelli, N., Strahan, P.E. (2006), Finance as a barrier to entry: Bank competition and industry structure in U.S. local markets, Journal of Finance, Vol. 61, pp

25 Claessens, Stijn, and Laeven L. (2004). What Drives Bank Competition? Some International Evidence. Journal of Money, Credit and Banking, Vol. 36, pp Cole, R. A., Goldberg, L. G. and White, L. J. (2004), Cookie cutter vs. character: The micro structure of small business lending by large and small banks, Journal of Financial and Quantitative Analysis, Vol. 39, pp Cooper, K. and Fraser, D.R., Banking Deregulation and the New Competition in Financial Services, Cambridge, Mass Coval, J. D., and Thakor, A. V. (2005). Financial Intermediation as a Beliefs-Bridge between Optimists and Pessimists. Journal of Financial Economics, Vol. 75, pp Critchfield, T., Davis, T., Davison, L., Gratton, H., Hanc, G., and Samolyk, K. (2004). Community Banks: Their Recent Past, Current Performance, and Future Prospects, FDIC Banking Review, Vol. 16, No , pp De Nicol o, G., Bartholomew, P. Zaman, J. and Zephirin, M. (2004). Bank Consolidation, Internationalization, and Conglomerization: Trends and Implications for Financial Risk. Financial Markets, Institutions and Instruments, Vol. 13, pp Deep, A. and Schaefer, G. K. (2004). Are Banks Liquidity Transformers? Working Paper, No. RWP John F. Kennedy School of Government Faculty Research. Available at SSRN: Accessed at 16 th August Dell Ariccia, G., Detragiache, E., and Rajan, R. (2008). The real effects of banking crises, Journal of Financial Intermediation, Vol. 17, pp Demirgüç-Kunt, A. and Detragiache, E. (2002), Does deposit insurance increase banking system stability? An empirical investigation, Journal of Monetary Economics, Vol. 49, No. 7, pp Demirgüç-Kunt, A., Feyen, E. and Levine, R. (2012). The Evolving Importance of Banks and Securities Markets, World Bank Economic Review, World Bank, Washington, DC. DeYoung. R. (2015). Banking in the United States. In: Berger, A. N., Molyneux, P., Wilson, J. O. S. eds. The Oxford Handbook of Banking, 2 nd edition, Oxford: Oxford University Press. pp Diamond, D. W., and Dybvig, P. H. (1983). Bank Runs, Deposit Insurance, and Liquidity, Journal of Political Economy, Vol. 91, pp Diamond, D. W., and R. G. Rajan. (2000). A Theory of Bank Capital, Journal of Finance, Vol 55, pp

26 Diamond, D. W., and Rajan, R. G. (2001). Liquidity Risk, Liquidity Creation, and Financial Fragility: A Theory of Banking. Journal of Political Economy, Vol. 109, pp Diamond, D. W. and Rajan, R.G., (2005). "Liquidity Shortages and Banking Crises," Journal of Finance, American Finance Association, Vol. 60, No. 02, pp Dick, A. A. (2006). Nationwide Branching and Its Impact on Market Structure, Quality, and Bank Performance. Journal of Business, Vol. 79, pp Dick, A. A., and Lehnert, A., (2010). Personal Bankruptcy and Credit Market Competition, Journal of Finance, Vol 65, No. 2, pp Eljelly, A. (2004). Liquidity-Profitability Tradeoff: An empirical Investigation in an Emerging Market, International Journal of Commerce & Management, Vol. 14, No. 2 pp Freeman, D, G. (2002), Did State Bank Branching Deregulation Produce Large Growth Effects? Economics Letters, Vol. 75, No.3, pp Fu, X., Lin, Y. and Molyneux, P., (2015). Bank Competition, Efficiency and Liquidity Creation in Asia Pacific, Palgrave Macmillan: London. Fungáèová, Z., Solanko, L., and Weill, L. (2014). Does competition influence the bank lending channel in the euro area? Journal of Banking and Finance, Vol. 49, pp Gual, J. (1999), Deregulation, Integration and Market Structure in European Banking, Journal of Japanese and International Economies, 12, Goddard, J., Molyneux, P., and Wilson, J.O.S. (2004). The profitability of European banks: A cross-sectional and dynamic panel data analysis, The Manchester School, Vol. 72, No. 3, pp Gonzalez, F. (2005). Bank regulation and risk-taking incentives: An international comparison of bank risk, Journal of Banking and Finance, Vol. 29, No. 5, pp Gorton, G., and Winton, A. (2000). Liquidity Provision, Bank Capital, and the Macroeconomy, Working Paper, University of Minnesota Greenwood, J., and Jovanovic, B. (1990), "Financial Development, Growth and the Distribution of Income," Journal of Political Economy, Vol. 98, No. 47, pp Greenbaum, S. I., and Thakor, A. V. (2007). Contemporary Financial Intermediation, 2 nd edition, Amsterdam: North Holland, Elsevier / Academic Press. Hanc, G. (2004). The Future of Banking in America, FDIC Banking Review, Vol 17, No. 4, pp

27 Hicks, J. (1935). Annual Survey of Economic Theory: The Theory of Monopoly, Econometrica. Holland, D., Inscoe, D., Waldrop, R. and Kuta, W. (1999). Interstate Banking The Past, Present and Future, FDIC Banking Review, Vol 9, No. 1, pp Holmstrom, B., and Tirole, J. (1998). Public and Private Supply of Liquidity, Journal of Political Economy, Vol. 106, pp Huang R.R. (2008). Evaluating the real effect of bank branching deregulation - comparing contiguous counties across U.S. state Borders, Journal of Financial Economics, Vol. 87, pp Hubbard, R. G., and Palia, D. (1995)."Executive Pay and Performance: Evidence from the U. S. Banking Industry," Journal of Financial Economics, Vol. 39, pp Jayaratne, J. and Strahan, E.P. (1996). The Finance-Growth Nexus: Evidence from Bank Branch Deregulation Quarterly Journal of Economics, Vol. 111, No. 3, pp Jayaratne, J., Strahan, P.E. (1997). The benefits of branching deregulation, FRBNY Economic Policy Review, December Issue, pp Jayaratne, J., & Strahan, P. E. (1998). Entry Restrictions, Industry Evolution, and Dynamic Efficiency: Evidence from Commercial Banking. Journal of Law and Economics 41, pp Jayaratne, J., and Wolken, J. (1999). How important are small banks to small business lending? New evidence from a survey of small firms, Journal of Banking and Finance, Vol. 23, pp Johnson, C.A., and Rice, T. (2008), Assessing a decade of interstate bank branching, The Washington and Lee Law Review, Vol. 65, pp Jones, K. D. Critchfield, T. (2005). Consolidation in the US banking industry: Is the long strange trip about to end FDIC Banking Review, Vol 17, No. 4, pp Joskow, Paul L., and Roger G. Noll. (1981) Regulation in theory and practice: An overview. In Fromm, G. ed., Studies in public regulation, The MIT Press, pp Kashyap, A. K., and Stein, J. (2000). What do a million observations on banks say about the transmission of monetary policy? American Economic Review, Vol. 90, pp Kashyap, A. K, Stein, J. and Wilcox, D. (1993). "Monetary Policy and Credit Conditions: Evidence from the Composition of External Finance," American Economic Review, Vol. 83, pp

28 Kashyap, A. K., Rajan, R. and Stein, J. (2000). Banks as liquidity providers: An explanation for the coexistence of lending and deposit-taking Journal of Finance, Vol. 57, No.1, pp Keeton, W. R. (1995). Multi-Office Bank Lending to Small Businesses: Some New Evidence. Federal Reserve Bank of Kansas City Economic Review, Vol. 80, No. 2, pp Kennedy, P. E., (2008). A guide to econometrics, 6 th ed., Malden, MA: Blackwell. Kerr, W., Nanda, R. (2009), Democratizing entry: Banking deregulation, financing constraints, and entrepreneurship, Journal of Financial Economics, Vol. 94, pp Keys, B. J., Mukherjee, T., Seru, A. and Vig, V. (2010). Did securitization lead to lax screening? Evidence from subprime loans, Quarterly Journal of Economics, Vol. 125, pp Kroszner, R. S. and Strahan, P. E. (1999). What Drives Deregulation? Economics and Politics of the Relaxation of Bank Branching Restrictions. Quarterly Journal of Economics, Vol. 114, No. 4, pp Kroszner, R. S. and Strahan, P. E. (2006). Regulation and deregulation of the U.S. banking industry: causes, consequences, and implications for the future, Working Paper, National Bureau of Economic Research. Laeven, L., and Levine, R. (2009). Corporate Governance, Regulation, and Bank Risk Taking Journal of Financial Economics, Vol. 93, No.2, pp Lewis, A. and Pescetto, G. (1996). EU and US Banking in the 1990s, London: Academic Press. Loutskina, E., and Strahan P.E., (2009), Securitization and the declining impact of bank finance on loan supply: Evidence from mortgage originations, Journal of Finance Vol 64, pp Mattana, E. and Panetti, E. (2014). Bank Liquidity, Stock Market Participation, and Economic Growth, Journal of Banking and Finance, Vol. 48, pp Marquez, R., (2002). Competition, adverse selection and information dispersion in the banking industry, Review of Financial Studies, Vol. 15, pp Matz, L., Neu, P. (2007). Liquidity Risk: Measurement and Management, Wiley: Hoboken. McLaughlin, S. (1995). "The impact of interstate banking and branching reform: evidence from the states.", Current Issues in Economics and Finance, Vol. 1., No. 2, Federal Reserve Bank of New York. 28

29 Mishkin, F. S. (1999). Financial Consolidation: Dangers and Opportunities. Journal of Banking and Finance, Vol. 23, pp Molyneux, P., Lloyd-Williams, D.M. and Thornton, J. (1994) Competitive Conditions in European Banking. Journal of Banking and Finance, Vol. 18, pp Molynuex, P., and Thornton, J. (1992). Determinants of European Bank Profitability: A Note, Journal of Banking and Finance, Vol. 16, No. 6, pp Ogura, Y. (2012). Lending competition and credit availability for new firms: Empirical study with the price cost margin in regional loan markets, Journal of Banking and Finance, Vol. 36, No. 6, pp Panzar, J.C and Rosse, J.N. (1987). Testing for monopoly s equilibrium, The Journal of Industrial Economics, Vol. 35, pp Paravisini, D., (2008). Local bank financial constraints and firm access to external finance, Journal of Finance Vol.63, pp Petersen, M. and Rajan R. (1995). The Effect of Credit Market Competition on Lending Relations, Quarterly Journal of Economics, Vol. 110, No. 2, pp Pigou, A. C. (1938). Money wages in relation to unemployment, The Economic Journal, Vol. 48, No. 189, pp Repullo, Rafael. (2004) Capital Requirements, Market Power, and Risk-Taking in Banking, Journal of Financial Intermediation, Vol. 13, pp Rice, T. and Strahan, P.E. (2010). Does Credit Supply Affect Small-Firm Finance?, Journal of Finance, Vol. 65, No. 3, pp Roberts, M., Whited, T., Endogeneity in Empirical Corporate Finance. Unpublished Working Paper, University of Pennsylvania, University of Rochester. Schaeck, K., Cihak, M. and Wolfe, S. (2009): Are Competitive Banking Systems More Stable?, Journal of Money, Credit and Banking, Vol. 41, No. 4, pp Schaeck, K., and Cihak, M. (2012). Banking competition and capital ratios, European Financial Management, Vol. 18, No. 5, pp Smith, B. (1984) Private Information, Deposit Interest Rates, and the Stability of the Banking System. Journal of Monetary Economics, Vol. 14,pp Spieker, R. (2004) Bank branch growth has been steady: Will it continue? FDIC Draft FOB. 29

30 Strahan, P.E. (2003). The real effects of US banking deregulation, The Federal Reserve Bank of St. Louis Review, Vol. 85, pp Strahan, P. (2008). Liquidity production in 21st century banking, Working Paper, No , National Bureau of Economic Research. Stigler, G. J. (1971). The theory of economic regulation, The Bell journal of economics and management science, Vol. 2, No. 1, pp Thakor, A. V. (2005). Do Loan Commitments Cause Overlending? Journal of Money, Credit and Banking, Vol. 37, pp White, E., (1983). The Regulation and Reform of the American Banking System, , Princeton: Princeton University Press. 30

31 Table 1: Interstate branching deregulation indicators from State Interstate Branching Laws from 1994 to 2003 State Restrictivness Index Effective Year Minimum Age 31 De novo Interstate Branching Acquisition of Single Branch Deposit Cap Alabama years No No 30% Alaska years No Yes 50% Arizona years No Yes 30% Arizona years No No 30% Arkansas years No No 25% California years No No 30% Colorado years No No 25% Connecticut years Yes Yes 30% Delaware years No No 30% DC No Yes Yes 30% Florida years No No 30% Georgia years No No 30% Georgia years No No 30% Hawaii No Yes Yes 30% Hawaii years No No 30% Idaho years No No None Illinois years No No 30% Indiana years Yes Yes 30% Indiana No Yes Yes 30% Iowa years No No 15% Kansas years No No 15% Kentucky No No No 15% Kentucky years No No 15% Louisiana years No No 30% Maine No Yes Yes 30% Maryland No Yes Yes 30% Massachusetts years Yes Yes 30% Michigan No Yes Yes None Minnesota years No No 30% Mississippi years No No 25% Missouri years No No 13% Montana years No No 22% Nevada years Limited Limited 30% New Hampshire No Yes Yes 30% New Hampshire years Yes Yes 30% New Hampshire years No No 20% New Jersey No No Yes 30% New Mexico years No No 40% New York years No Yes 30% North Carolina No Yes Yes 30% North Dakota No Yes Yes 25% North Dakota No No No 25% Ohio No Yes Yes 30% Oklahoma No Yes Yes 20% Oklahoma years No No 15% Oregon years No No 30% Pennsylvania No Yes Yes 30% Rhode Island No Yes Yes 30%

32 South Carolina years No No 30% South Dakota years No No 30% Tennessee years Yes Yes 30% Tennessee years Yes Yes 30% Tennessee years No Yes 30% Tennessee years No No 30% Texas No Yes Yes 20% Utah years Yes Yes 30% Utah years No Yes 30% Vermont No Yes Yes 30% Vermont years No Yes 30% Virginia No Yes Yes 30% Washington years No No 30% West Virginia No Yes Yes 25% Wisconsin years No No 30% Wyoming years No No 30% Source: Rice and Strahan (2010) 32

33 Table 2: Summary Statistics Variable Obs Mean Std. Dev. Min Max Liquidity creation ratio 61, OBS liquidity creation ratio 61, BRI 61, Ln Total Asset 61, Equity Ratio 61, ROE 61, Ln Zscore 61, Equity Volatility Ratio 61, Ln HHI 61, Multi BHC Dummy 61, Merger Dummy 61,

34 Table 3: Correlation Matrix LCR BRI Ln GTA Equity Ratio ROE Ln Zscore Equity Volatility MBHC Dummy Merger Dummy LCR 1 BRI *** 1 Ln GTA *** *** 1 Equity Ratio *** *** *** 1 ROE *** *** *** *** 1 Ln Zscore * *** *** *** *** 1 Equity Volatility *** *** *** *** *** *** 1 MBHC dummy 0.141*** *** *** *** *** *** *** 1 Merger Dummy *** *** *** *** *** *** *** 0.132*** 1 34

35 Table 4: Exogeneity tests PANEL A: EFFECT OF BANK RISK (1) (2) (3) (4) Lag 0 Lag 1 Lag 3 Lag 5 Ln Zscore (-0.913) (-1.084) (-1.135) (-1.034) Control Variables YES YES YES YES Observations State Fixed Effect YES YES YES YES Year Fixed Effect YES YES YES YES PANEL B: EFFECT OF BANK LIQUIDITY CREATION (1) (2) (3) (4) Lag 0 Lag 1 Lag 2 Lag 5 Liquidity Creation (0.277) (0.185) (-0.589) (-0.788) Control Variables YES YES YES YES Observations State Fixed Effect YES YES YES YES Year Fixed Effect YES YES YES YES Dependent variable is deregulation which equals 1 if the state deregulated and 0 otherwise. Robust z-statistics in parentheses, *** p<0.01, ** p<0.05, * p<0.1. Control Variables include local deposit HHI index and means of bank size, bank profitability (ROE) in the market. Error terms are clustered at state-year level.

36 Table 2: T-tests for parallel trends T= -1 T= -2 T= -3 Difference Wilcoxon P-value Difference Wilcoxon P-value Difference Wilcoxon P-value LCR Ln GTA ECAP * ROE Ln-Zscore EQV * Ln HHI *** p<0.01, ** p<0.05, * p<0.1 Figure 1: Parallel trends Notes: Figure 1 illustrates the behaviour of yearly changes in the dependent variables, for five years preceding the IBBEA. The treatment group is presented with the solid line, whereas the control group is depicted as the dash line. 36