Printer Friendly

Value Creation in U.S. Bank Mergers Before and After the Global Financial Crisis.

Introduction

Over the past several decades, the U.S. banking industry has experienced an unprecedented transformation as a result of liberalization and deregulation in financial markets and advances in technological innovation. During this time, the number of U.S. commercial banks has declined from 14,417 (1985) to 5,083. (1) Mergers and acquisitions were used as a strategic tool by bank managers to grow market share, diversify geographically or improve their competitive position in their respective markets. After witnessing such a big wave of mergers in past decades, both in terms of the number of deals and the market value involved in these deals, it is relevant to ask whether these transactions create significant value, and if they do, what the drivers of such value creation are. Unfortunately, empirical evidence does not provide unambiguous answers with respect to these questions. Furthermore, examination of the recent merger transactions is of importance because following the recent financial crises, we observed substantial changes in the way financial institutions operate due to increased legislative oversight and shareholder awareness. Since the investors became more cautious regarding financial markets and valuations in general, it is reasonable to believe that the short term market reactions to bank M&As may have changed.

In order to provide empirical contribution to the ongoing debate, this study explores the short term market reaction to M&A announcements that occurred during the 2000-2014 period in the U.S. banking industry and aims to test value creation capacity of M&A transactions around the Global Financial Crisis and financial regulations that followed, such as the Dodd-Frank Wall Street Reform and Consumer Protection Act of 2010. These legislative measures were designed to boost stability by improving accountability and transparency in the financial system and to curb the "too big to fail" perception, thus protecting the system from abusive banking practices as certain mergers and acquisitions are evaluated in terms of financial stability of U.S. banking system before approved by the authorities. These supervisory actions are more than likely to influence the characteristics of mergers and acquisitions both in terms of the managerial motives and the size of control premium, which determines the value created or destroyed by these deals.

With this goal in mind, we analyze the target, bidder and combined bank cumulative abnormal returns (CARs) utilizing standard event study methodology. In this respect, we test the three major hypotheses in M&A literature: 1) synergey hypothesis, 2) hubris hypothesis and 3) synergy and hubris hypothesis, as explained in Section 3. Then, we attempt to explain why differences in stock price reactions to M&A deals exist. Specifically, we investigate: 1) cash + mix vs. common stock only--financed transactions, 2) in-state (geographically focused) vs. inter-state (geographically diversified) mergers and 3) relative size of bidder and target banks. Finally, we explore whether the short term market reaction to bank merger announcements has changed before and after the Global Financial Crisis to better understand the influence of regulation on merger premiums.

The selection of the benchmark to measure normal returns is central to conducting the event study. In previous literature, S&P 500 Index was employed frequently as a benchmark in computation of abnormal returns (AR). To correctly measure the announcement impact of an M&A deal in banking industry, we need to control for confounding factors outside the banking industry. With this goal in mind, we employ the U.S. Banking Index alternative to S&P 500 Index as a benchmark. This approach may yield more accurate results relative to traditional approaches since the returns of the banks subject to M&A deals are more correlated with Banking Index returns than S&P 500 Index returns, and thus capture the actual effect of the deal. (2) In this respect, one minor contribution of this study to the current literature is the introduction of new benchmark in computation of ARs.

Our research contributes to the literature by examining the most recent merger deals, including those occurring after the Global Financial Crisis. We try to understand whether merger premiums have changed around the Crisis and recent regulation in the financial markets, such as Dodd-Frank Wall Street Reform and Consumer Protection Act. Although the impact of the Crisis on financial markets is extensively examined, to our best knowledge, there is no comprehensive study on short-term wealth effect of the U.S. bank mergers comparing pre- and post-Global Financial Crisis periods. In this sense, our sample period is relatively large (from 2000 to 2014) and comprehensive compared to earlier studies in relevant literature.

The current study is organized as follows. The next section presents a short overview of the salient related literature. Section 3 elaborates on the hypotheses to be tested. Section 4 outlines data and sample selection methodology. Section 5 introduces the model used to compute the abnormal stock price changes around the merger announcement dates. Section 6, with its sub-sections, presents the empirical results, and section 7 offers concluding remarks.

Literature Review

The widespread bank consolidation in the U.S. banking industry since the early 1980s has motivated a growing literature on the drivers and wealth effects of bank mergers. The short-term and long term impacts of M&A transactions from various aspects have been examined in depth by market researchers and academics. Since our goal in this section is not to provide a comprehensive review of the literature, we will mostly confine ourselves to a few related studies on the announcement effects of M&A transactions and refer the reader to several survey articles. (3)

M&A transactions are important corporate events that are likely to have a significant impact on the future operational efficiency and profitability of the company. M&A activities can substantially reduce operating costs since larger firms can operate more efficiently through the elimination of redundant facilities and personnel. Merger-related gains can also arise from increased market share and reduced competition. The increased market power can enable the surviving organization to earn higher profits by raising loan rates and lowering deposit rates (Pilloff and Santomero 1996).

An M&A transaction can benefit shareholders when the combined firm is more valuable than the sum of the values of the two separate pre-merger firms. The primary cause of the value gain comes from the performance improvement following the merger. Traditional event studies are designed to capture the financial market's expectation as to the overall performance results of mergers. They provide the most reliable evidence on the wealth effect for the bank shareholders, where abnormal stock market reaction around merger announcements is used as a gauge of value creation or destruction.

In spite of the more or less common methodology used in the event studies of the U.S. bank mergers, a great deal of variation exists among studies in sample and geographic coverage and period of time over which the market model is estimated and abnormal returns are computed. For example, sample size varies from 21 in Trifts and Scanlon (1987) to 558 in Becher (2000). The period of time over which abnormal returns are computed is only the announcement date and the day before (-1, 0) in Cornett and Tehranian (1992), whereas the same period is -40 to +20 weeks in Trifts and Scanlon (1987).

Considering the shareholders of the target firm, a typical transaction is said to be value-enhancing. However, evidence regarding the gains of bidding firms exhibits a somewhat mixed picture. Earlier research typically reports significant positive abnormal returns to the shareholders of the target firms and significant negative returns to shareholders of bidder firms. (4,5) A few studies find positive abnormal returns or no significant abnormal returns for the bidders. (6,7) Shareholders of the combined firms generally experience significant positive returns around the transaction date. (8) Despite substantial diversity among voluminous studies, the findings point to the favorable conclusion that merger and acquisitions are in general value-enhancing for the shareholders. However, since the value of the bidder typically falls, the market believes that bidding firms tend to overpay for the deal in anticipation of merger benefits that are not likely to be realized.

For example, based on 29 merger deals in the U.S. over the 1979-1985 period, Neely (1987) reports 36.22% positive ARs for target firms and no statistically significant gains for bidder banks. Using a sample size of 21 deals, Trifts and Scanlon (1987) find average gains of 21.4% for targets and average losses of 3.25% for bidders for the period of 1982-1985. Cornett and De (1991a) study 189 U.S. banks (152 bidders and 37 targets) during the period of 1982-1986 and find average gains of 9.76% for targets and a loss of 0.44% for bidders. Houston and Ryngaert (1994) analyze 153 bank mergers over the period of 1985-1991. Using a 5-day event window period, they find CARs of -2.32%, 14.39% and 0.38% for bidder banks, target banks and combined, respectively. A later study by Houston and Ryngaert (1997), using 209 mergers over the period of 1985-1992 and in a 6-day (-4, +1) event-window period, finds 0.24% and 20.4% for bidder and target CARs, respectively.

Using a relatively larger sample size, Becher (2000) examines 558 U.S. bank mergers over the period 1980-1997 and finds that target banks enjoy positive returns. According to Becher (2000), bank mergers posit synergistic gains, and mergers in this industry do not take place just to create empires for chief executive officers (CEOs). Over a 36-day (-30, +5) event window, CARs are 22.64%, -0.10% and 3.03% for target banks, bidder banks and combined, respectively. Over an 11-day (-5, +5) event window, CARs are 17.10%, -1.08% and 1.80% for target banks, bidder banks and combined, respectively.

The evidence for European banks is broadly consistent with these results. Cybo-Ottone and Murgia (1996) analyze 26 European mergers taking place between 1988 and 1995 in 13 European banking markets. Their results indicate significantly negative abnormal returns for target firms and no significant abnormal returns for the acquirers, suggesting a transfer of wealth from acquirer to target shareholders. In a more recent study, Asimakopoulos and Athanasoglou (2013) examine the impact of announced M&As on stock prices for a sample of European banks for a period of 15 years (1990-2004). They find that, overall, an M&A announcement does not create value for the shareholders of bidders.

Another issue tackled by these studies is the impact of geographically focusing (in-state) or diversifying (inter-state) transactions on the merger premiums. A generally accepted view is that geographically focusing (in-state) transactions can produce more efficient post-merger entities since they provide the opportunity to eliminate overlapping offices as well as to combine back-office operations and administrative functions. A study of cross-border M&As involving U.S. banks by DeLong (1999) finds that more value is created by the cross-borders deals. Several studies that investigate abnormal returns in respect to inter-state and intra-state mergers generally find no difference (Hannan and Wolken 1989; Hawawini and Swary 1990; Baradwaj, Dubofsky and Fraser 1992; Cornett and Tehranian 1992 and Toyne and Tripp 1998).

The way an M&A transaction is financed can also impact the yield to shareholders. Regarding the method of payment, most studies find that the method of payment plays an important role in explaining acquiring firms' stock return. For instance, Travlos (1987) finds that shareholders of the bidding firm experience significant negative abnormal returns for the deals that are financed with stocks but zero or positive abnormal returns when it is financed with cash. Antoniou and Zhao (2004) reports that bidder's returns are lower when the transaction is financed with stocks compared to cash or combined offers. Other studies (Huang and Walking 1987; Franks, Harris and Mayer 1988 and Eckbo and Langohr 1989), confirm that the choice of payment method has an impact on the profitability of a takeover.

Testable Hypotheses

Literature offers several hypotheses to explain motivations behind mergers and acquisitions that can broadly be categorized under value-creating and non-value creating motivations. In this study, we test three alternative hypotheses explaining the possible reasons for mergers and acquisitions as outlined in Becher (2000): the synergy hypothesis, the hubris or empire building hypothesis and the combined synergy and hubris hypothesis.

According to the synergy hypothesis, M&As take place when the combined firm value is greater than the sum of the values of the individual firms. The additional value is the synergistic gain arising from increase in operational or financial efficiencies obtained by combining the resources of the bidder and target firms. Accordingly, synergy hypothesis predicts CARs of target firms should be positive, CARs of bidder firms should be non-negative and CARs of the combined should be positive. As an example of the non-value creating motivations, the hubris or empire building hypothesis posits that bidder firms overpay to acquire the target firm due to either bidder management beliefs that synergies between target and bidder exist when in fact they do not exist or when the management of bidder firm is self-driven to realize a merger or acquisition in order to build an empire rather than create a synergy. The hubris or empire building hypothesis would predict that, on average, CARs to target firms are positive, CARs to bidder firms are negative and the CARs to the combined firm are non-positive (Roll 1986).

In sum, while the synergy hypothesis argues that mergers are wealth creating events, the hubris or empire building hypothesis claims that M&As may be the result of managerial hubris and empire building rather than a synergistic motive. A third alternative hypothesis put forth by Becher (2000) is that mergers and acquisitions are a result of both the synergy and hubris hypotheses. Accordingly, CARs of the target and of combined firm should be positive along with negative CARs of the bidder firms, implying that positive synergies may be associated with an M&A transaction. However, bidder firms might overpay to obtain these synergies.

Using the standard event study methodology, we initially test the abovementioned three hypotheses for target, bidder and combined firms for different periods. In order to examine whether value creation characteristic of merger events has changed around the Global Financial Crisis, we test pre-Crisis vs. post-Crisis CARs for the target, bidder and combined banks using the null hypothesis of [H.sub.0]=[CARs.sub.pre-Crisi]=[CARs.sub.post-Crisi]

Data and Sample Selection

Initially, we retreived a global list of 15,847 bank M&A deals data from the year of 2000 to 2014 from SNL Financial database. In SNL Financial data, there are four different country classifications: Actual Acquirer Country, Buyer Country, Target Country and Seller Country. Having included only U.S.-based banks for all four classifications, our sample size reduced to 8,622. Daily stock return data was needed to compute abnormal returns for the bidder and target banks. After including only publicly traded banks, our sample size dramatically reduced to 604. For the purpose of our analysis, only commercial banks and bank holding companies were included in the sample. This further reduced our final sample size to 450.

We utilized Center for Research in Security Prices (CRSP) database to obtain the return data for each bank's security. After applying our inclusion criteria to the analysis, our final sample size was 214 bidder and target banks for the period of 2000-2014.

According to Pilloff and Santomero (1998), a selection bias stems from either including in the sample only major M&A deals during the period surrounding the deal of interest or excluding from the sample M&As banks that had multiple mergers in the same year or over a given time period. Because of these criteria, transactions that are most relevant to analysis of M&A deals might be omitted in the sample. Since our sample selection method does not have such inclusion criteria, our analysis is not subject to such selection biases.

Average number of M&A transactions over 15 years covered in this study is roughly 14 transactions per year. The highest number of transactions took place in the years of 2004, 2006 and 2007, with 30, 27 and 28 deals, respectively. The lowest number of M&A transactions took place in the years of 2001 and 2002, with single transaction in each. Average target-to-bidder ratio for the whole sample is 15.64%. This number implied that from 2000 to 2014, on average, market value of bidder bank is 6.39 times larger than the value of target bank in our sample. The average target-to-bidder ratio for the 125 deals in pre-Crisis period (2000-2007) is 16.3%, while the same ratio for the 74 deals in post-Crisis period (2010-2014) is 14.7%; either bidder banks got bigger or target banks got smaller or both happened together following the Crisis.

Empirical Model and Methodology

Event study methodology has been used frequently to assess the effect of a particular event on the returns of a firm's common stock. In a typical study, first the market model is estimated using historical data, and then the estimated market model's parameters are used to determine the size and direction of the price changes. In this study, we examine the value creation around the announcement of a bank merger and acquisition by using the method outlined in Brown and Warner (1985). According to the efficient market hypothesis, the market incorporates all available information immediately and fully in stock prices. Thus, prompt correction or balancing will be coming into the prices after the announcement of an M&A event.

Abnormal return represents the gain or loss for shareholders, which could be explained by many factors, including an M&A transaction. It is called an abnormal return in a sense that it deviates from what an investor would normally expect to earn or lose for accepting a certain level of risk in normal market conditions. The null hypothesis of our study is that such an M&A event has no impact on the return generating process or the abnormal return is to be zero.

In order to estimate the expected return of each security, this study uses the market model, which relates return of a corresponding security to the return of the market portfolio. The market model assumes that there is a stable linear relationship between the market return and the security return. The linear relationship in the pre-event estimation period or the market model is defined as:

[R.sub.it] = [[alpha].sub.i] + [[beta].sub.i][R.sub.mt] + [[epsilon].sub.it] (1)

where, [R.sub.it], is the rate of return on the stock of bank i on day t, [R.sub.mt] is the return on the U.S. Banking Index on day t (market portfolio), [[alpha].sub.i] is the intercept, [[beta].sub.i] is stock i's sensitivity to the market's return and [[epsilon].sub.it] is the zero mean disturbance term at time t. (9) This regression analysis is performed in the estimation window to determine the market model parameters.

Using the estimates in Equation (1), expected return for stock i on day t is defined as:

[mathematical expression not reproducible] (2)

where, [[??].sub.it] is expected rate of return on the stock of bank i on day t, [R.sub.mt] is the return on the U.S. Banking Index on day t (market portfolio), [[??].sub.i] is the intercept and [[??].sub.1] is stock i's sensitivity to the market's return. Then, the following equation is utilized to compute the abnormal returns or risk-adjusted returns in the event period:

[AR.sub.it] = [R.sub.it] - [[??].sub.it] (3)

where, [AR.sub.it] is the abnormal return for bank i on day t, [R.sub.it] is the actual return on the stock of bank i on day t and [[R].sub.it] is the expected return of bank i on day t.

Then, the CARs over the event period are computed as the sum of the arithmetic means of the cross-sectional abnormal returns of each day over the event window period. For instance, if the event window is 3-day (-1, +1), then ARs are computed for each day (-1, 0 and +1) and the sum of ARs ([AR.sub.-1] + [AR.sub.0] + [AR.sub.+1]) for security A provides us CAR for Security A. The CARs under 3-day (-1,+1), 5-day (-2,+2) and 36-day (-30,+5) event windows are computed.

In order to explore whether the abnormal returns have changed over time, the event study analysis is conducted for the whole sample period (2000-2014) and the sub-sample periods of 2000-2007 (pre-Crisis period), 2008-2009 (Crisis period) and 2010-2014 (post-Crisis period). In order to test whether a merger is value creating, we computed the combined CARs using the methodology suggested by Houston and Ryngaert (1994):

Combined Cumulative Abnormal Return (CCAR) = [([V.sub.ib][CAR.sub.ib])+[([V.sub.it][CAR.sub.it])]/([V.sub.ib]+[V.sub.it])] (4)

where, [V.sub.ib] is the market value of bidder bank i on the first day of the event window and [V.sub.it] is the market value of target bank i on the first day of the event window. The value of each bank is computed by multiplying the market value of the bank's stock with the bank's number of shares outstanding. [CAR.sub.ib] represents the CAR for bidder bank i over the event window, and [CAR.sub.it] represents the CAR for target bank i over the respective event window.

Finally, in an attempt to discover the main drivers of target, bidder and combined CARs, a regression analysis is utilized. In the regression analysis, the dependent variable is the CARs, and the independent variables are relative size, payment method, geographic dummy variable and dummy variable for each year. The relative size is computed as the natural logarithm of target-to-bidder ratio. Market value for each bank is computed by multiplying the stock price of each bank with the number of shares outstanding on the first day of corresponding M&A transaction's event window. The regression model can be written as follows:

[CAR.sub.i] = [[alpha].sub.0] + [[beta].sub.1] (ln(size.)) + [[beta].sub.2] ([payment.sub.i]) + [[beta].sub.3] ([geographic.sub.i]) + [[beta].sub.2]([year.sub.i]) + [[epsilon].sub.i] (5)

ln(size.) = the ratio of the natural logarithm of market value of target bank to the market value of bidder bank ratio;

[Payment.sub.i] = the payment method dummy variable; if the M&A transaction is financed by Cash+Mix, it equals 1, and if the M&A transaction is financed by Common Stock Only, it equals 0;

[Geographic.sub.i] = the geographic location dummy variable; if the M&A transaction occurs in the same state (in-state), it equals 1, and if the M&A transaction occurs in different states (inter-state), it equals 0;

[Year.sub.i] = dummy variable of the year that M&A transaction took place; and

[[epsilon].sub.i] = error term

Empirical Results

Usually, the target shareholders demand a fairly large premium to sell their shares to the bidder firms because a typical merger is expected to create significant corporate value in the post-merger firm. In an efficient market, this premium should be immediately reflected in the target firm's share price. Average wealth effects for the overall sample and for various sub-samples classified by different event windows are presented in Table 1.

Overall, M&As announced between 2000 and 2014 create substantial positive CARs (statistically significant at the 1% level) for the target and combined firms. Overthe entire sample period, the CARs to the target banks are on average 23.41% (3-day event window), 23.14% (5-day event window) and 26.04% (36-day event window), respectively, with all three at 1% significance level. These results are in line with the previous studies that report shareholders of the target banks earn significant positive returns around the announcement dates. Although we do not report the results to save space, we should note that our results are robust to the selection of market index (S&P 500 Index vs. U.S. Banking Index) as a benchmark to compute abnormal returns. (10)

The CARs to target banks within 3-day (-1, +1) event window for 2000-2014, 2000-2007, 2008-2009 and 2010-2014 periods are 23.41%, 19.55%, 25.99% and 29.48% (all statistically significant at the 1%), respectively. For the same periods, the CARs to target banks within 5-day (-2, +2) and 36-day (-30, +5) event windows are also similar. Target banks results are consistent with the synergy hypothesis, hubris hypothesis and hubris and synergy hypothesis as all three hypotheses expect target banks to have positive CARs.

Panel B of Table 1 displays the results for the bidder banks. For the full 2000-2014 period, the CARs to their shareholders are negative under each event window and statistically significant within 3-day (-1, +1) and 5-day (-2, +2) event windows. The CAR values are -1.41% (significant at 1 %), -1.07% (significant at 5 %), and -1.07% within the 3-day (-1, +1), 5-day (-2, +2) and 36-day (-30, +5) event windows, respectively. These results are in line with the findings of prior studies that the shareholders of the bidder firms experience a loss around the announcement of an M&A.

The CARs to bidder banks in 2000-2014, 2000-2007, 2008-2009 and 2010-2014 periods are -1.41% (significant at 1%), -2.06% (significant at 1%), -4.09% and 0.24%, respectively, within 3-day (-1,+1) event window. For the same periods, the CARs to bidder banks within 5-day (-2, +2) event window are -1.07% (significant at 5%), -2.09% (significant at 1%), -4.19% and 1.29%, respectively. Within 36-day (-30, +5) event window, the CARs to the bidder banks in 2000-2014, 2000-2007, 2008-2009 and 2010-2014 periods are -1.07%, -2.11% (significant at the 1%), -5.72% (significant at the 10%) and 1.64%, respectively. Bidder results are consistent with hubris hypothesis and hubris and synergy hypothesis, as these hypotheses expect bidder banks to have negative CARs. However, our overall results for the banks are not consistent with the synergy hypothesis as this hypothesis expects the bidder banks to realize non-negative CARs.

Panel C of Table 1 summarizes CARs to the combined entity are positive and statistically significant at the 1% level in all event windows for the full period. CARs to combined firm turn out to be 2.24%, 2.52% and 3.29% and all statistically significant at 1% for 3-day (-1, +1), 5-day (-2, +2) and 36-day (-30, +5) event windows, respectively. These results exhibit that the combined firm experiences a positive but small return around the announcement of a merger or acquisition and suggests a wealth transfer from the bidder banks to the target banks. This finding is also substantiated by Becher (2000), Anderson, Becher and Campbell (2004) and Delong and DeYoung (2004).

Overall, the results obtained by utilizing U.S. Banking Index return data point out that target banks realize a positive return, bidder banks realize a negative return and the combined experiences a positive return around the merger. These results imply that the target banks increase their values at the expense of the bidder banks and the overall result is positive for the combined entity.

OVERALL RESULTS WITH RESPECT TO THE HYPOTHESES

Our research directly tests three hypotheses as to why banks might engage in takeover activities. As mentioned before, the most essential motive of companies engaging in mergers and acquisitions is the synergy hypothesis. The synergy hypothesis proposes that the value of the combined firm is higher than the sum of the individual firm values (Bradley, Desai and Kim 1988; Seth 1990; Maquiera, Megginson and Nail 1998 and Hubbard and Palia 1999). The combined firm can improve revenues by increasing operational efficiency and lowering cost of capital. Operating and/or financial synergy created from the deal not only allows the bidder bank to cover the costs of the takeover but also provides the target bank shareholders with a sizable premium for their shares. Therefore, under synergy-motivated transactions the combined entity typically exhibits a positive net acquisition value.

The second category, the hubris hypothesis, suggests that managers make mistakes in evaluating target firms and engage in takeovers even when there is no synergistic gain. If a deal cannot create synergy gains, then the average increase in the target firm's market value should be more than offset by the average decrease in the value of the bidding firm. Thus, in a hubris-driven takeover, the premium paid to target shareholders represents a transfer between the target and the bidding banks. According to Roll (1986), the stock price of the acquiring firm should fall after the market becomes aware of the hubris motivated takeover bid since it does not represent an efficient allocation of wealth. Roll (1986) also states that the combined effect of the rising value of the target and the falling value of the acquiring firm should not be positive.

Table 2 compares our results produced using U.S. Banking Index Return with the expectation of each hypothesis. Looking at the comparison table, only CARs to the target banks satisfy all three hypotheses for the whole period as well as for the sub-periods, as CARs to target banks are positive across the board. CARs to the bidder banks are negative except in post-Crisis period of 2010-2014 as the regulation after the Crisis led to barely positive CARs for the bidders. Lastly, CARs to the combined are all positive, which satisfy both the synergy hypothesis and the hubris and synergy hypothesis.

PRE-CRISIS VS. POST-CRISIS RESULTS

In order to capture the differences in merger premiums before and after the Global Financial Crisis, we divide our sample into two periods and carried out the same analysis for the first period of 2000-2007 (representing pre-Crisis period) and the second period of 2010-2014 (representing the post-Crisis period).

Table 3 presents the results for the pre-Crisis and post-Crisis periods. Comparing the pre-Crisis and post-Crisis gains for the target banks, we can reject the null hypothesis of H0= CARspre-Crisis = CARspost-Crisis in all three event windows of 3-day (-1, +1), 5-day (-2, +2) and 36-day (-30, +5) at 1% significance level, meaning that the CARs to the target banks before and after the Global Financial Crisis of 2007-2008 are significantly different than each other. As seen from the Table 3, in all three event windows, CARs to the target banks increased by approximately 1000 basis points in post-Crisis (2010-2014) period compared to pre-Crisis period (2000-2007).

We also reject the same null hypothesis for the bidder banks in all three event windows, albeit with different significance levels (3-day event window at 10%, 5-day event window at 1% and 36-day event window at 5% significance level), indicating significantly different CARs to bidder shareholders around the Global Financial Crisis. The same conclusion also holds for the combined firm at 1% significance level for all three event windows. As seen from the Table 3, in all three event windows, CARs to the bidder and combined banks increased at least 230 basis points in post-Crisis (2010-2014) period compared to pre-Crisis period (2000-2007).

In Table 3, we also report the results of the F-test to understand whether variances of CARs have changed during the pre-Crisis (2000-2007) and post-Crisis (2010-2014) periods. Equality of variances between two periods is rejected at 1% significance level for the targets and combined in all three event windows. Equality of variances between two periods is rejected at 1% significance level within 5-day and 36-day event windows and at 10% significance level within 3-day event windows for bidder banks.

Our findings also reveal that M&As taking place before the Global Financial Crisis period (2000-2007) realize lower gains for targets, bidders and combined firms compared to post-Crisis period (2010-2014). This difference can be attributed to improvement in overall economic and banking conditions, reflecting the transition from the recovery following the financial crisis to a more prudent and reliable market environment in the post-Crisis period. Improved economic conditions make potential targets more attractive due to their healthier portfolios and strengthen potential acquirers, giving them greater ability to acquire new banks. The passage of the Dodd-Frank Act and its sub-implementing regulations imposed certain restraints on bank mergers to limit the "too big to fail" banks' exposures and bolstered the financial stability of the U.S. banking industry. These regulatory initiatives might have created stronger banks, a sounder banking system and more transparent and efficient capital markets, causing a change in M&A premiums as reflected in short-term market reaction to deal announcements before and after the Global Financial Crisis.

METHOD OF PAYMENT

M&A transactions can be financed using different ways, such as cash, securities or some combination of the two. The M&A literature posits that choice of payment method used to finance the M&A deal can affect the abnormal returns to the target, bidder and combined firms. For example, Travlos (1987) reports negative abnormal returns for the bidders when the deal is financed with stocks but zero or positive abnormal returns when it is financed with cash. Antoniou and Zhao (2004) reports that bidder's returns are lower when the transaction is financed with stocks compared to cash or combined offers. Other studies (Huang and Walking 1987; Franks, Harris and Mayer 1988 and Eckbo and Langohr 1989) confirm that the choice of payment method has an impact on the profitability of a takeover.

Table 4 displays the number and percentage of the M&As that are financed by Cash + Mix and Common Stock. For overall sample period, 72% of M&As are financed by Cash + Mix and 28% of deals are financed by Common Stock.

Table 5 exhibits the CARs for targets, bidders and combined firm with respect to the payment methods (Cash+Mix vs. Common Stock) over three different event windows (3-day, 5-day and 36-day) and different sub-periods. Panel A of Table 5 displays the CARs to target banks for different sample periods. CARs of Cash+Mix and Common Stock transactions to the target banks within 3-day (-1, +1) event window for the periods 2000-2014 (whole period), 2000-2007 (pre-Crisis period), 2008-2009 (Crisis period) and 2010-2014 (post-Crisis period) are 24.84% vs. 20.55%, 21.52% vs. 15.15% (difference is statistically significant at 5% level), 27.80% vs. 28.32% and 29.79% vs. 28.45%, respectively. CARs of Cash+Mix and Common Stock transactions to the target banks for 5-day (-2, +2) and 36-day (-30,+5) event windows exhibit similar patterns. Overall, we can infer that, cash involved transactions produce higher positive yields for the shareholders of the target banks compared to common stock only-financed transactions, although the difference is only statistically significant for the pre-Crisis period.

Panel B of Table 5 presents the results for the bidders for different sample periods and event windows. For example, the results of 3-day window CARs of Cash+Mix and Common Stock transactions to the bidder banks for the periods 2000-2014 (whole period), 2000-2007 (pre-Crisis period), 2008-2009 (Crisis period) and 2010-2014 (post-Crisis period) are -1.91% vs. 0.49%, -2.05% vs. -1.54%, -5.81% vs. 5.52% and -0.96% vs. 3.15%, respectively. For the bidder banks, the CARs common stock-financed transactions are higher than cash-financed transactions CARs (albeit the differences are not statistically significant), which may indicate the market tends to reward the bidder banks when common stock is used in financing the M&A transaction.

The similar results are presented in Panel C of Table 5 for the combined entity. Irrespective of event windows and time periods used in the analysis, common stock-financed transactions' CARs turn out to be higher than cash-financed transactions' CARs but not statistically significant as in the case of bidder banks.

GEOGRAPHIC LOCATION

In this section, we attempt to answer the question as to whether the targets, bidders and combined abnormal returns are affected by geographically focusing (intra-state or in-state) versus geographically diversifying (inter-state) mergers. If a bidder banks attempts to acquire or merge with a target bank in the same state, the transaction is called In-state or Intra-state Merger. If the target bank is in a different state, then the transaction is called Inter-state Merger. Table 6 demonstrates the summary statistics for our deals sample based on geographic location. For the overall 2000-2014 period, 56.07% of the M&A's are geographically focusing (in-state/intra-state) whereas 43.93% of the M&A's are geographically diversifying (inter-state).

Table 7 exhibits the CARs for targets, bidders and combined with respect to the geographic diversification (inter-state vs. in-state) over three different event windows and sub-periods. As shown in Panel A, for the target banks, in-state transactions produce larger CARs than inter-state deals for all periods except the Crisis period (2008-2009), but the differences in CARs are not statistically significant. The results for the bidder banks are mixed for different periods, and there is no statistically significant difference between in-state and inter-state deals. Panel C displays the results for the combined entity where the only significant difference between in-state and inter-state deals (1.93% vs. 0.20%) exists in the pre-Crisis period of 2000-2007. This conclusion holds under different event windows.

REGRESSION ANALYSIS

In this subsection, we apply multivariate regression analysis for the whole period in an attempt to explain the price reaction to the merger announcement by the relative size of the merging banks, payment type (Cash+Mix vs. Common Stock) and geographic focus (in-state vs. interstate). The regression results for targets, bidders and combined are presented in Table 8, with respect to different event windows.

Panel A of Table 8 exhibits the regression results for target banks. Under 3-day event window, relative size of target to bidder ratio is negatively correlated with the target CARs at 10% significance level, meaning that as the relative size of target to bidders increases, the CARs for target banks decrease. For the same window, geographic location dummy variable (l=in-state, 0=inter-state) turns out to be positive and significant at 10% level, indicating that instate M&As yield more positive yields for the shareholders of the target banks. The method of payment dummy variable is not significantly related to the CARs for the target banks.

Panel B of Table 8 exhibits the regression results of bidder banks. Within 3-day event window, relative size of target to bidder ratio and geographic location dummy variable are both negatively correlated with the CARs, but they are not statistically significant at any conventional level. Under the same event window, method of payment dummy variable (1=cash + mix, 0= common stock) is found to be negative and significant at 5% level, meaning that shareholders of the bidder banks experience lower abnormal returns when cash is involved in financing the M&As transaction. The regression results are in line with the method of payment CAR results outlined in the "Method of Payment" section.

The results of the regression analyses for the combined entity is presented in Panel C of Table 8. Within all event windows (3-day, 5-day and 36-day), relative size of target to bidder ratio is highly significant (1% level) and positively correlated with the CARs, meaning that as the relative size increases, the yield to the shareholders of the combined entity increases as well. The method of payment and the geographic location dummy variables do not produce any significant results for the combined firms.

Summary and Conclusion

One of the most prominent features of the U.S. economy during the past several decades has been the changing structure of the financial sector. Banks and other financial institutions tried to take advantage of opportunities presented by large scale deregulation and technological In this study, we examine the wealth effects of U.S. bank mergers spanning a period of 15 years (2000-2014) by utilizing a standard event-study methodology and the U.S. Banking Index as the market return in estimation of market parameters. According to the overall results, M&A announcements on average create significant value for the shareholders of the target and the combined banks at the expense of the shareholders of acquirer banks.

We find significantly different abnormal returns between pre-Crisis (2000-2007) and post-Crisis (2010-2014) periods. Our results suggest that the CARs to targets, bidders and combined banks increased significantly following the Global Financial Crisis. New financial regulations put in place following the Global Financial Crisis could be one reason for significantly higher gains in the post-Crisis period, as the new regulations could reduce the risk levels with more conservative rules. Another reason could be that stronger and healthier banks surviving the Crisis could increase the quality of target pool for the acquirers.

We also examine different drivers of abnormal returns on M&A transactions. In terms of method of payments, cash involved transactions yield higher gains than common stock-financed transactions for the target bank shareholders. For the bidder banks and the combined, common stock-financed transactions yields higher gains than cash-financed transactions, indicating the market rewards the bidder banks when common stock is used in financing the M&A transaction. This conclusion is also supported by the results of the regression analyses. Negative and significant coefficient on the payment type dummy variable suggests that shareholders of the bidder banks experience lower abnormal returns when cash is involved in financing the M&A transaction.

With respect to the geographic diversification for targets and combined, in-state (geographically focused) transactions yield higher CARs than inter-state (geographically diversified) deals. Regression analysis reveals similar results for target banks. The positive and significant coefficient on the geographic location dummy variable indicates that instate M&As yield more positive yields for the shareholders of the target banks. For the bidder banks, the relationship is not significant within different event windows and the periods. Our results are in line with Delong (2001), as geographically focused M&As may be rewarded more than inter-state M&As by the capital markets, mostly due to operational efficiency gains such as the enhancement of managerial efficiency, reduction of overhead costs, maximization of market power and creating value through reduction of over-investment or economies of scale and scope.

In addition, we test three hypotheses in M&A literature: synergy hypothesis, hubris hypothesis and hubris and synergy hypothesis. Target banks results are more robust and consistent with synergy hypothesis, hubris hypothesis and hubris and synergy hypothesis as all three hypotheses expect target banks to have positive CARs. Bidder bank results are consistent with hubris hypothesis and hubris and synergy hypothesis because these hypotheses expect bidder banks to have negative CARs. However, our overall results for the bidder banks are not consistent with the synergy hypothesis because this hypothesis expects the bidder banks to realize non-negative CARs. Combined bank results are consistent with synergy hypothesis and hubris and synergy hypothesis; these hypotheses expect combined firm to have positive CARs. However, our overall results for the combined banks are not consistent with the hubris hypothesis, as this hypothesis expects the combined to realize non-positive CARs.

Before concluding the study, several caveats are in order. First, because event studies require stock market data to estimate market model parameters, our study of mergers and acquisitions involves only relatively large, publicly traded banks as bidders and targets. Since there are many more mergers transactions that involve banking firms that are not publicly traded, our results are not necessarily representative of all bank mergers. Second, the market model we use to estimate abnormal returns is not free of criticism for misspecification problems in studies of security performance. Finally, our analysis is based on a methodology that captures short term valuation effects of M&A transactions. The analysis presented in this study can be extended to examine the potential long-term value gain implications of merger transactions for future research in this strand of literature.

References

Alien L., and A.S. Cebenoyan. "Bank Acquisitions and Ownership Structure: Theory and Evidence." Journal of Banking & Finance, vol. 15, no. 2,1991, pp. 425-448.

Amihud, Y., G. DeLong, and A. Saunders. "The Effects of Cross-border Bank Mergers on Bank Risk and Value." Journal of International Money and Finance, vol. 21, no. 6, 2002, pp. 857-877.

Anderson, C.W., D.A. Becher, and TL. Campbell. "Bank Mergers, the Market for Bank CEOs, and Managerial Incentives." Journal of Financial Intermediation, vol. 13, no.1, 2004, pp. 6-27.

Antoniou, A., and H. Zhao. "Long-run Post-takeover Stock Return: The Impact of Overlapping Return, Takeover Premium, and Method of Payment." Centre for Empirical Research in Finance (CERF) Durham Business School University of Durham, 2004.

Asimakopoulos, I., and P.P. Athanasoglou. "Revisiting the Merger and Acquisition Performance of European Banks." International Review of Financial Analysis, vol. 29(C), 2013, pp. 237-249

Baradwaj, B., D. David, and F. Donald. "Bidder Returns in Interstate and Intrastate Bank Acquisitions." Journal of Financial Services Research, vol. 5, no. 3, 1992, pp. 261-273.

Becher, D. "The Valuation Effects of Bank Mergers." Journal of Corporate Finance, vol. 6, no. 2, 2000, pp.189-214.

Becher, D. "Bidder Returns and Merger Anticipation: Evidence from Banking Deregulation." Journal of Corporate Finance, vol.15, no. 1, 2009, pp. 85-98.

Bradley, M., A. Desai, and E.H. Kim. "Synergistic Gains from Corporate Acquisitions and Their Division between the Stockholders of Target and Acquiring Firms." Journal of Financial Economics, vol. 21, no.l, 1988, pp. 3-40.

Brown, S., and 1 Warner. "Using Daily Stock Returns: The Case of Event Studies." Journal of Financial Economics, vol. 14, no. 1, 1985, pp. 3-31.

Cornett, M., and S. De. "Common Stock Returns in Corporate Takeover Bids: Evidence from Interstate Bank Mergers." Journal of Banking & Finance, vol. 15, no. 2, 1991, pp. 273-295.

Cornett, M., and S. De. "Medium of Payment in Corporate Acquisitions: Evidence from Interstate Bank Mergers." Journal of Money, Credit, and Banking, vol. 23, no. 4, 1991, pp. 767-776.

Cornett, M., and H. Tehranian. "Changes in Corporate Performance Associated with Bank Acquisitions." Journal of Financial Economics, vol. 31, no. 2, 1992, pp. 211-234.

Cybo-Ottone, A., and M. Murgia. "Mergers and Shareholder Wealth in European Banking." Journal of Banking & Finance, vol. 24, no. 6, 2000, pp. 831-859.

DeLong, G. L. "Domestic and International Bank Mergers: Shareholder Gains from Focusing Versus Diversifying." Brookings-Wharton Papers on Financial Services: 2000, edited by Robert E. Litan and Anthony M. Santomero, Brookings Institution Press, 2000, p. 143.

DeLong, Gayle. "Stockholder Gains from Focusing Versus Diversifying Bank Mergers." Journal of Financial Economics, vol. 59, no. 2, 2001, pp. 221-252.

DeLong, Gayle, and Robert DeYoung. "Learning by Observing: Information Spillovers in the Execution and Valuation of Commercial Bank M&As," Working paper WP 2004-17, Federal Reserve Bank of Chicago, 2004.

Desai, Anand S., and Roger D. Stover. "Bank Holding Company Acquisitions, Stockholder Returns, and Regulatory Uncertainty." The Journal of Financial Research, vol. 8, no. 2, 1985, pp. 145-156.

Dodd, P., and R. Ruback. "Tender Offers and Stockholder Returns: An Empirical Analysis." Journal of Financial Economics, vol. 5, no. 3, 1977, pp. 351-373.

Eckbo, B.E., and H. Langohr. "Information Disclosure, Method of Payment, and Takeover Premiums: Public and Private Tender Offers in France." Journal of Financial Economics, vol. 24, no. 2, 1989, pp. 363-403.

Esty, B., B. Narasimhan, P. Tufano. "Interest-rate Exposure and Bank Mergers." Journal of Banking & Finance, vol. 23, no. 2, 1999, pp. 255-285.

Franks, J.R., R.S. Harris, and C. Mayer. "Means of Payment in Takeovers: Results for the United Kingdom and the United States." Corporate Takeovers: Causes and Consequences, University of Chicago Press, 1988, pp. 221-264.

Hannan, T.H., and D.D. Wolken. "Returns to Bidders and Targets in the Acquisition Process: Evidence from the Banking Industry." Journal of Financial Services Research, vol. 3, no. 1, 1989, pp. 5-16.

Hart, D.R., and V.P. Apilado. "Inexperienced Banks and Interstate Mergers." Journal Economics Business, vol. 54, no. 3, 2002, pp. 313-330.

Hawawini, G.A., and I. Swary. Mergers and Acquisitions in the U.S. Banking Industry. North-Holland, 1990.

Holdren, Don P., Helen M. Bowers, and W. Doe Mason. "Bank Holding Company Acquisition Activity: Evidence from Pre- and Post-Deregulation Periods." The Financial Review, vol. 29, no. 2, 1994, pp. 275-92.

Houston, D.F., C.M. James, and M.D. Ryngaert. "Where Do Merger Gains Come from? Bank Mergers from the Perspective of Insiders and Outsiders." Journal of Financial Economics, vol. 60, no. 2-3, 2001, pp. 285-331.

Houston, J.F., and M.D. Ryngaert. "The Overall Gains from Large Bank Mergers." Journal of Banking and Finance, vol. 18, no. 6, 1994, pp. 1155-1176.

Houston, J.F. and M.D. Ryngaert. "Equity Issuance and Adverse Selection: A Direct Test Using Conditional Stock Offers." Journal of Finance, vol. 52, no. 1, 1997, pp. 197-219.

Huang, Y.S., and R.A. Walking. "Differences in Residuals Associated with Acquisition Announcements: Payment, Acquisition Form, and Resistance Effects." Journal of Financial Economics, vol. 19, no. 2, 1987, pp. 329-349.

Hubbard, R.G., and D. Palia. "A Re-examination of the Conglomerate Merger Wave in the 1960s: An Internal Capital Markets View." The Journal of Finance, vol. 54, no. 3, 1999, pp. 1131-1152.

Hudgins, S.C., and B. Seifert. "Stockholders and International Acquisitions of Financial Firms: An Emphasis on Banking." Journal of Financial Services Research, vol. 10, no. 2, 1996, pp. 163-180.

James, C., and P. Weir. "Returns to Acquirers and Competition in the Acquisition Market: The Case of Banking." Journal of Political Economy, vol. 95, no. 2, 1987, pp. 355-370.

Jarrell, G.A., and M. Bradley. "The Economic Effects of Federal and State Regulations of Cash Tender Offers." Journal of Law and Economics, vol. 23, no. 2, 1980, pp. 371-407.

Kolaric, S., and D. Schiereck. "Shareholder Wealth Effects of Bank Mergers and Acquisitions in Latin America." Management Research: Journal of the Iberoamerican Academy of Management, vol. 11, no. 2, 2013, pp.157-177.

Loughran, T., and A.M. Vijh. "Do Long-term Shareholders Benefit from Corporate Acquisitions?" Journal of Finance, vol. 52, no. 5, 1997, pp. 1765-1790.

Madura, J., and K.J. Wiant. "Long-term Valuation Effects of Bank Acquisitions." Journal of Banking and Finance, vol. 18, no. 6, 1994, pp. 1135-1154.

Maquieira, C.P., W.L. Megginson, and L. Nail. "Wealth Creation Versus Wealth Redistributions in Pure Stock-for-Stock Mergers." Journal of Financial Economics, vol. 48, no. 1, 1998, pp. 3-33.

Neely, Walter P. "Banking Acquisitions: Acquirer and Target Shareholder Returns." Financial Management, vol. 16, no. 4, 1987, pp. 66-74.

Palia, D. "Recent Evidence of Bank Mergers." Financial Markets, Instruments, and Institutions, vol. 3, no. 5, 1994, pp. 36-59.

Pilloff, S.J. "Performance Changes and Shareholder Wealth Creation Associated with Mergers of Publicly Traded Banking Institutions." Journal of Money, Credit and Banking, vol. 28, no. 3, 1996, pp. 294-310.

Pilloff, S.J., and A.M. Santomero. "The Value Effects of Bank Mergers and Acquisitions." Working Paper, The Wharton School, University of Pennsylvania, 1996.

Pilloff, S.J., and A.M. Santomero. "The Value Effects of Bank Mergers and Acquisitions." Bank Mergers & Acquisitions, Springer US, 1998, pp. 59-78.

Rhoades, S.A. "A Summary of Merger Performance Studies in Banking, 1980-1993, and an Assessment of the Operating Performance and Event Study Methodologies," Staff Study 167, Board of Governors of the Federal Reserve System, 1994.

Roll, R. "The Hubris Hypothesis of Corporate Takeovers." Journal of Business, vol. 59, no. 2, 1986, pp. 197-216.

Seth, A. "Value Creation in Acquisitions: A Re-examination of Performance Issues." Strategic Management Journal, vol. 11, no. 2, 1990, pp. 99-115.

Siems, Thomas F. Bank Mergers and Shareholder Wealth: Evidence from 1995's Megamerger Deals. Financial Industry Studies Series, Federal Reserve Bank of Dallas, 1996.

Toyne, M.F., and J.D. Tripp. "Interstate Bank Mergers and Their Impact on Shareholder Returns: Evidence from the 1990s." Quarterly Journal of Business and Economics, vol. 37, no. 4, 1998, pp. 48-58.

Travlos, N.G. "Corporate Takeover Bids, Methods of Payment, and Bidding Firms' Stock Returns." The Journal of Finance, vol. 42, no. 4, 1987, pp. 943-963.

Trifts, Jack W., and Kevin R. Scanlon. "Interstate Bank Mergers: The Early Evidence." Journal of Financial Research, vol.10, no. 4, 1987, pp. 305-313.

Wall, L.D., and B.E. Gup. "Market Valuation Effects of Bank Acquisitions." Bank Mergers: Current Issues and Perspectives. Kluwer Academic Publishers, 1989, pp. 107-120.

Zhang, H. "Wealth Effects of US Bank Takeovers." Applied Financial Economics, vol. 5, no. 5, 1995, pp. 329-336.

iMDAT DOGAN

Energy Exchange Istanbul

H. SEMIH YILDIRIM

(1) Federal Reserve Bank of St. Louis, https://research.stlouisfed.org/fred2/series/USNUM (accessed on March 08, 2016)

(2) U.S. Banking Index return data is obtained from Bloomberg database.

(3) For a survey of event studies on bank mergers, see Rhoades (1994) and Pilloff and Santomero (1996).

(4) See, among others, Desai and Stover (1985), Dames and Weir (1987), Neely (1987), Trifts and Scanlon (1987), Wall and Cup (1989), Hawavini and Swary (1990), Cornett and De (1991), Cornett and Tehranian (1992), Houston and Ryngaert (1994), Madura and Wiant (1994), Zhang (1995), Hudgins and Seifert (1996), Pilloff (1996), Siems (1996), Becher (2000), Delong (2001), Hart and Apilado (2002), Delong and DeYoung (2004) and Asimakopoulos and Athanasoglou (2013).

(5) Hawavini and Swary (1990), Baradwaj, Dubofsky and Fraser (1992), Cornett and Tehranian (1992), Holdren, Bowers and Mason (1994), Madura and Wiant (1994), Palia (1994), Houston and Ryngaert (1994), Pilloff (1996), Siems (1996), Loughran and Vijh (1997), Esty, Narasimhan and Tufano (1999), Delong (2001), Amilhud, Delong and Saunders (2002), Delong and DeYoung (2004).

(6) Desai and Stover (1985), James and Weir (1987), Neely (1987), Cornett and De (1991) and Becher (2008).

(7) Trifts and Scanlon (1987), Alien and Cebenoyan (1991), Holdren, Bowers and Mason (1994), Becher (2000) and Hart and Apilado (2002).

(8) Becher (2000), Houston, Dames and Ryngaert (2001), Anderson, Becher and Campbell (2004), Zhang (1995), Hart and Apilado (2002), Dodd and Ruback (1977), Gregg and Bradley (1980) and Kolaric and Schiereck (2013).

(9) U.S. Banking Index obtained from Bloomberg is utilized as the market return in our analysis.

(10) For comparison, S&P500 CRSP equally-weighted index data is utilized.
TABLE 1
Cumulative Abnormal Returns (CARs)

                3-day (-1, +1)    5-day (-2, +2)       36-day (-30, +5)
                Event Window      Event Window          Event Window
Year           CARs    p-value   CARs    p-value       CARs    p-value
               (%)               (%)                   (%)

Panel A:
Target Banks
2000 - 2014    23.41    .0001    23.14    .0001        26.04    .0001
2000 - 2007    19.55    .0001    19.35    .0001        22.10    .0001
2008 - 2009    25.99    .0013    25.41    .0015        28.07    .0012
2010 - 2014    29.48    .0001    29.07    .0001        32.29    .0001
Panel B:
Bidder Banks
2000 - 2014    -1.41    .0041    -1.07    .0354        -1.07    .1333
2000 - 2007    -2.06    .0001    -2.09    .0001        -2.11    .0003
2008 - 2009    -4.09    .2005    -4.19    .1823        -5.72    .0570
2010 - 2014     0.24    .8350     1.29    .2707         1.64    .3320
Panel C:
Combined
2000 - 2014     2.24    .0001     2.52    .0001         3.29    .0001
2000 - 2007     1.05    .0025     0.97    .0087         1.55    .0048
2008 - 2009     2.68    .3310     2.69    .2728         3.39    .4095
2010 - 2014     4.20    .0004     5.12    .0001         6.22    .0001

This table represents the CARs results with respect to U.S. Banking
Index utilized. P-values test the statistical significance of the CARs.

TABLE 2
Comparison of Hypotheses in Sub-Periods
Actual Results Compared with Expected Results (3-day Event Window)

                                     2000-2014       2000-2007

Hubris                Expected
                      Results
           Target     Positive       [check] (***)   [check] (***)
           Bidder     Negative       [check] (***)   [check] (***)
           Combined   Non-positive   X (***)         X (***)
Synergy
           Target     Positive       [check] (***)   [check] (***)
           Bidder     Non-negative   X (***)         X (***)
           Combined   Positive       [check] (***)   [check] (***)
Hubris &
Synergy
           Target     Positive       [check] (***)   [check] (***)
           Bidder     Non-negative   [check] (***)   [check] (***)
           Combined   Positive       [check] (***)   [check] (***)

                    2008-2009                2010-2014

Hubris

                   [check] (***)            [check] (***)
                   [check]                  X
                   X                        X (***)
Synergy
                   [check] (***)            [check] (***)
                   X                        [check]
                   [check]                  [check] (***)
Hubris &
Synergy
                   [check] (***)            [check] (***)
                   [check]                  X
                   [check]                  [check] (***)

(*), (**), (***) denote statistical significance at the 10%, 5%, and
1 % level, respectively.

TABLE 3
Pre- and Post-Crisis Cumulative Abnormal Returns (CARs)

                        3-day (-1.+1)
                        Event Window
Year            CARS      f-vatue     t-value           p-value
                (%)                   of t-test:        of
                                      pre-Crisis        t-test:
                                      vs. post          pre-Crisis
                                      Crisis            vs.
                                                        post
                                                        Crisis

Panel A:
Target Banks
2000 - 2007     19.55      2.50         -2.95 (***)     0.0039
2010 - 2014     29.48
Panel B:
Bidder Banks
2000 - 2007     -2.06      8.95         -1.93*          0.0576
2010 - 2014      0.24
Panel C:
Combined
2000 - 2007      1.05      6.47         -2.65 (***)     0.0096
2010-2014        4.20

                                5-day (-2. +2)
                                Event Window
Year                 CARS   f-value   t-vatue        p-value
                     (%)               of            of
                                       t-test:       t-test:
                                       pre-Crisis    pre-Crisis
                                       vs. post      vs.
                                       Crisis        post
                                                     Crisis

Panel A:
Target Banks
2000 - 2007         19.35   2.79      -2.83 (***)    0.0055
2010 - 2014         29.07
Panel B:
Bidder Banks
2000 - 2007          2.09   7.19      -2.79 (***)    0.0064
2010 - 2014          1.29
Panel C:
Combined
2000 - 2007          0.97   5.60      -3.52 (***)    0.0007
2010-2014            5.12

                                 36-day (-30.+5)
                                 Event Window
Year            CARs       f-value      t-value         p-value
                (%)                     oft-test:       of
                                        pre-Crisis      t-test:
                                        vs. post        pre-Crisis
                                        Crisis          vs. post
                                                        Crisis

Panel A:
Target Banks
2000 - 2007     22.10      2.70        -2.72 (***)      0.0077
2010 - 2014     32.29
Panel B:
Bidder Banks
2000 - 2007     -2.11      5.09        -2.12 (**)       0.0362
2010 - 2014      1.64
Panel C:
Combined
2000 - 2007      1.55      4.82        -2.85 (***)      0.0054
2010-2014        6.22

This table displays the CARs for targets, bidders and combined around
the announcement date of a bank merger or acquisition. (*), (**), (***)
denote statistical significance at the 10%, 5%, and 1% level (t-test),
respectively. For F-test, H0=Variances are equal. P-value represents
the significant of difference.

TABLE 4
Summary Statistics of the Sample with Respect to Method of Payments

Time Period      Cash+Mix      Common     Cash+Mix      Common
                               Stock                    Stock

2000 - 2014      154           60         72%           28%
2000 - 2007       89           36         71%           29%
2008 - 2009       10            5         67%           33%
2010 - 2014       55           19         74%           26%

Cash + Mix indicates any combination of financing that includes cash.
Common stock indicates that the M&A was financed by stock only.

TABLE 5
CARs to Targets, Bidders and Combined with Respect to Method of Payment

                                3-day (-1.+1)
                                Event Window
                   Cash            Common           f-value    t-value
                   + Mix           Stock
                   (%)             (%)

Panel A:
Targets
2000-2014          24.84          20.55             1.58      -1.20
2000-2007          21.52          15.15             1.83      -2.26 (**)
2008-2009          27.80          28.32             2.36       0.04
2010-2014          29.79          28.45             2.58      -0.16
Panel B:
Bidders
2000-2014          -1.91           0.49             5.70       1.57
2000-2007          -2.05          -1.54             3.81       0.57
2008-2009          -5.81           5.52             1.13       1.68
2010-2014          -0.96           3.15            14.37       0.95
Panel C:
Combined
2000-2014           1.82           3.98             5.85       1.39
2000-2007           1.15           1.27             1.49       0.15
2008-2009           0.76          13.48             8.36       1.49
2010-2014           3.12           6.62             8.25       0.88

                               5-day (-2, +2)
                               Event Window
                 Cash       Common     f-value      t-value
                 + Mix      Stock
                 (%)        (%)

Panel A:
Targets
2000-2014       24.51      20.48       1.57         -1.11
2000-2007       21.27      14.95       1.84         -2.30 (**)
2008-2009       28.61      26.93       3.92         -0.11
2010-2014       28.99      29.25       2.20          0.03
Panel B:
Bidders
2000-2014       -1.21       0.01       3.64          0.88
2000-2007       -1.99      -2.18       2.93         -0.19
2008-2009       -3.21       3.50       3.13          0.98
2010-2014        0.41       3.52       7.66          0.78
Panel C:
Combined
2000-2014        2.37       3.79       5.38          0.90
2000-2007        1.17       0.67       1.30         -0.61
2008-2009        3.04      12.04       6.37          0.98
2010-2014        4.19       7.55       8.90          0.87

                                 36-day (-30, +5)
                                  Event Window
                Cash        Common      f-value      t-vatue
                + Mix       Stock
                (%)         (%)

Panel A:
Targets
2000-2014       27.68       22.18        2.34        -1.22
2000-2007       24.01       15.50        1.30        -2.48 (**)
2008-2009       39.29       34.16        8.35        -0.17
2010-2014       31.49       31.70        3.02         0.02
Panel B:
Bidders
2000-2014       -1.62        0.04        5.63         0.68
2000-2007       -2.54       -3.75        1.17        -1.00
2008-2009        2.07       11.15        4.41         0.68
2010-2014       -0.82        4.32       11.50         0.87
Panel C:
Combined
2000-2014        2.89        4.22        5.36         0.51
2000-2007        1.25       -0.27        1.27        -1.38
2008-2009       12.02       19.34        5.35         0.33
2010-2014        3.89        8.75        8.85         0.90

This table displays the CARs for targets, bidders and combined around
the announcement date of a bank merger or acquisition. (*), (**), (***)
denote statistical significance at the 10%. 5%, and 1% level (t-test),
respectively.

TABLE 6
Sample Statistics of the Sample with Respect to Geographic Location

Time Period     In-state     Inter-state     In-state      Inter-state
                                                           (%)

2000 - 2014     120          94              56%           44%
2000 - 2007      71          54              57%           43%
2008 - 2009      10           5              67%           33%
2010 - 2014      39          35              53%           47%

This table presents the number and percentage of M&As by geographic
location. In-state is defined as those M&As where the bidder and target
banks are headquartered in the same state. Inter-state indicates that
the bidder bank is not headquartered in the same state as the target
bank.

TABLE 7
CARs to Targets, Bidders and Combined with Respect to the Geographic
Diversification

                               3-day (-1.+1)
                               Event Window
Year         In-state    Interstate    f-value   t-value      p-value
             (%)         (%)

Panel A:
Targets
2000-2014   25.11         21.75         1.08      -1.14         .2540
2000-2007   21.10         17.88         1.30      -1.09         .2772
2008-2009   24.19         35.55         5.14       0.61         .5711
2010-2014   32.73         25.76         1.34      -1.15         .2531
Panel B:
Bidders
2000-2014   -1.37         -1.08         3.94       0.30         .7608
2000-2007
            -2.22         -1.49         1.09       1.15         .2524
2008-2009   -4.96          3.83         2.33       1.25         .2323
2010-2014    1.17         -1.16         7.46      -1.04         .3053
Panel C:
Combined
2000-2014    3.10          1.55         1.93      -1.58         .1156
2000-2007    1.93          0.20         2.75      -2.52 (**)    .0132
2008-2009    2.55          9.91         4.70       1.06         .3104
2010-2014    5.45          2.44         5.59      -1.36         .1795

                           5-day (-2, +2)
                           Event Window
Year        In-state   Interstate   f-value   t-value       p-value
            (%)        (%)

Panel A:
Targets
2000-2014    24.69      21.71        1.05      -1.00          .3179
2000-2007    20.99      17.43        1.33      -1.23          .2195
2008-2009    24.31      35.53        4.72       0.57          .5946
2010-2014    31.52      26.32        1.30      -0.83          .4087
Panel B:
Bidders
2000-2014    -0.78      -0.92        3.66      -0.15          .8837
2000-2007
             -2.30      -1.70        1.72       0.83          .4072
2008-2009    -2.96       3.01        2.37       0.87          .4010
2010-2014     2.55      -0.28        5.28      -1.26          .2126
Panel C:
Combined
2000-2014     3.55       1.77        1.79      -1.77 (*)      .0777
2000-2007     1.82      -0.01        3.26      -2.68 (***)    .0086
2008-2009     3.94      10.23        6.53       0.66          .5392
2010-2014     6.61       3.32        5.05      -1.53          .1308

                                36-day (-30. +5)
                                Event Window
Year            In-state     Interstate   f-value  t-value      p-vatue
                (%)          (%)

Panel A:
Targets
2000-2014       28.01         23.75        1.35    -1.25         .2140
2000-2007       23.84         18.56        1.31    -1.66         .1001
2008-2009       25.49         61.76        4.28     1.83 (*)     .0900
2010-2014       36.24         26.32        1.26    -1.45         .1519
Panel B:
Bidders
2000-2014       -0.79         -1.62        1.64    -0.53         .5989
2000-2007        2.88         -2.90        2.07    -0.02         .9850
2008-2009        2.06         11.17        3.87     0.68         .5078
2010-2014        2.27         -1.46        4.95    -1.16         .2500
Panel C:
Combined
2000-2014        4.51          1.68        1.36    -1.66         .0993
2000-2007        2.23         -1.05        2.53    -3.58 (***)   .0006
2008-2009        8.31         26.77        8.24     1.11         .2854
2010-2014        7.68          2.30        2.76    -1.79 (*)     .0787

This table displays the CARs for targets, bidders and combined around
the announcement date of a bank merger or acquisition. (*), (**), (***)
denote statistical significance at the 10%, 5%, and 1% level (t-test).
respectively. For F-test, H0=Variances are equal. P-value represents
the significant of difference.

TABLE 8
Regression Analysis Summary Statistics

Dependent Variable:                         3-day (-1.+1)
CARs for                                    Event Window
Independent                      Estimate      t-statistic      p-value
Variables

Panel A:
Target Banks
Constant                        0.16014         2.48 (**)       .0141
ln(MV of target/MV of          -0.01932        -1.76 (*)        .0808
bidder)
Method of payment               0.03582         1.12            .2662
dummy variable
Geographic location             0.05076         1.73 (*)        .0859
dummy variable
[R.sup.2]                      16.27%
Panel B: Bidder Banks
Constant                        0.01665         0.71            .4800
ln(MVof target/MV of           -0.00141        -0.35            .7251
bidder)
Method of payment              -0.02546        -2.18 (**)       .0306
dummy variable
Geographic location            -0.00108        -0.10            .9196
dummy variable
[R.sup.2]                       7.77%
Panel C: Combined
Constant                        0.07747         3.58 (***)      .0004
ln(MV of target/MVof            0.01475         4.18 (***)      .0001
bidder)
Method of payment              -0.01790        -1.62            .1060
dummy variable
Geographic location             0.00977         0.98            .3261
dummy variable
[R.sup.2]                      18.41%
Number of Obs.                214

Dependent Variable:                       5-day (-2, +2)
CARs for                                  Event Window
Independent                      Estimate     t-statistic    p-value
Variables

Panel A:
Target Banks
Constant                         0.15934      2.52 (**)      .0124
ln(MV of target/MV of           -0.01220     -1.12           .2643
bidder)
Method of payment                0.03377      1.04           .3010
dummy variable
Geographic location              0.04255      1.43           .1546
dummy variable
[R.sup.2]                       15.80%
Panel B: Bidder Banks
Constant                         0.01381      0.61           .5438
ln(MVof target/MV of            -0.00149     -0.38           .7046
bidder)
Method of payment               -0.1345      -1.15           .2524
dummy variable
Geographic location              0.00140      0.13           .8965
dummy variable
[R.sup.2]                       10.16%
Panel C: Combined
Constant                         0.07620      3.66 (***)     .0003
ln(MV of target/MVof             0.01513      4.25 (***)     .0001
bidder)
Method of payment               -0.00968     -0.87           .3845
dummy variable
Geographic location              0.01037      1.04           .3005
dummy variable
[R.sup.2]                       21.51%
Number of Obs.                 214

Dependent Variable:                  36-day (-30,+5)
CARs for                             Event Window
Independent                  Estimate       t-statistic      p-value
Variables

Panel A:
Target Banks
Constant                     0.16898        2.29 (**)        .0230
ln(MV of target/MV of       -0.00080       -0.06             .9497
bidder)
Method of payment            0.05443        1.43             .1543
dummy variable
Geographic location          0.04397        1.26             .2079
dummy variable
[R.sup.2]                   12.84%
Panel B: Bidder Banks
Constant                     0.01001        0.29             .7751
ln(MVof target/MV of        -0.00302       -0.50             .6169
bidder)
Method of payment           -0.1596        -0.88             .3780
dummy variable
Geographic location          0.00882        0.53             .5937
dummy variable
[R.sup.2]                   10.70%
Panel C: Combined
Constant                     0.08158        2.32 (**)        .0216
ln(MV of target/MVof         0.01690        2.81 (***)       .0055
bidder)
Method of payment           -0.00637       -0.34             .7347
dummy variable
Geographic location          0.01832        1.08             .2794
dummy variable
[R.sup.2]                   16.97%
Number of Obs.             214

This table presents the results of the regression analysis of the
cumulative abnormal return (CAR) for targets, bidders and combined. CAR
is regressed against the natural logarithm of the target-to-bidder
ratio, a method of payment dummy variable (1 if financed by any
combination of cash involved, and 0 if financed by common stock only)
and a geographic location dummy variable (1 if the M&A is inter-state
and 0 if in-state). In order to provide a snapshot of regression
analysis and save some space, we do not include the dummy variable for
each year on this table. Please keep in mind [R.sup.2]s encompass dummy
variable of each year. (*), (**), (***) denote statistical significance
at the 10%, 5%, and 1 % level, respectively, p value tests the
significance of the relationships.
COPYRIGHT 2017 Creighton University
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Dogan, Imdat; Yildirim, H. Semih
Publication:Quarterly Journal of Finance and Accounting
Article Type:Abstract
Geographic Code:1USA
Date:Jun 22, 2017
Words:11180
Previous Article:Organizing a Successful Academic Finance Conference: Thoughts, Heuristics and Statistical Evidence.
Next Article:A Test of the Functional Fixation Hypothesis Using Derivative Financial Instruments.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters