The effect of mergers and acquisitions on acquirer's long-term stock performance : evidence from the European Banking Industry in 2006-2009

(1)

The Effect of Mergers and Acquisitions on Acquirer’s Long-term Stock

Performance:

Evidence from the European Banking Industry in 2006-2009

Student name: Siyu Wang Student number: 10966382 Supervisor: Dr. J.E. (Jeroen) Ligterink

Finance and Organization - Finance Faculty of Business and Economics

(2)

Statement of Originality

This document is written by Siyu Wang who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

1. Introduction

There were unprecedented numbers of mergers and acquisitions (M&As) in the European banking sector during the recession period, both in the forms of within EU countries domestically as well as cross-border between countries. There were over 1.5 millions of M&A announcements worldwide since 1985s, more than 15 thousands happened in the year 2007 and 2008, half of them occurred in Europe. As can be seen in table 1, both the total transaction value and the number of deals reached the peak during the recession period.

Table 1

The number of all M&A deals and the total value of all transactions from 1985 to 2018 in Europe.

Note: Data are collected from the Thomson SDC Platinum.

M&As, in general, are the consolidations of companies or assets by buying or through other financial transactions on other firms (Kohli and Mann, 2012). To a large extent, M&As are based on the beliefs that gains are created through diminishing of competition and increase in market share (Piloff and Santomero, 1998). Another associated advantage is that the joint activities enable the acquirers to achieve cost restrictions, a realization of synergies as well as portfolio diversifications, thus, through the consolidations, the excess revenues are generated. Some also argue that M&As increase firm values and efficiency, move resources to their highest and best uses (Jensen, 1984). All these perspectives support that mergers and acquisitions

0 50 00 10 00 0 15 00 0 Nu mb er of D ea ls 0 50 00 00 10 00 00 0 15 00 00 0 Va lue of D ea ls 1980 1990 2000 2010 2020 Announcement Date

(4)

enable firms to generate more earnings, which are associated with greater stock price appreciation (Harkavy, 1953).

Table 2

The number of M&A deals in all year from 1985 to 2018 in Europe in different individual industries.

In addition, table 2 shows that the industry of other financials and banks engaged in M&As procedures the most since 1980s. This mainly due to the important role of banks in capital accumulations, firms’ growths, economic prosperities as well as its crucial policy implications, which further makes investigating the performance and efficiency of banks more interesting (Abbasoğlu, Aysan and Gunes, 2007).

The purpose of this thesis is to analyze the long-term stock performance of European acquiring banks three years after the M&A announcements by using the event study methodology. Indeed, this paper is not the first one aiming to find relationships between M&A-characteristics and the abnormal value creations. However, the uniqueness in this thesis is the attempt to look into details: how much abnormal returns are generated through the M&A processes, how different deal-specific factors influence the abnormal returns, and how these impacts different from cross-border and domestic mergers and acquisitions. In particular, this paper focuses on the

0 100000 200000 300000 400000

Value of Deals in Total Wireless

Transportation & Infrastructure Telecommunications Services REITs Publishing Professional Services Power Pharmaceuticals Others Other Financials and Banks Oil & Gas Metals & Mining Machinery Insurance Healthcare Equipment & Supplies Food and Beverage Chemicals Building/Construction & Engineering Brokerage Asset Management Alternative Financial Investments Aerospace & Defense

(5)

European banking industry, thus, requires that both acquirers and targets are European banks. All in all, this thesis is trying to evaluate the long-term abnormal returns for acquiring banks three-year after the event, and further test how deal-related factors impact the generation of abnormal returns.

The outline of this thesis is introduced as follows. The theoretical backgrounds of bid-related characteristics on abnormal returns and the predictions of these impacts are presented in Section 2 based on several pieces of literatures. In Section 3, data collection and its criteria, long-term abnormal return methodology and cross-sectional analysis are described. The empirical results and robustness checks regard the main regression are presented in Section 4, followed by a summary of hypotheses. Discussion and conclusion are illustrated in Section 5.

2. Literature Review

In this section, the theoretical background of M&As value creation and individual deal-characteristics will be reviewed. The corresponding expectations of the directions of these impacts will be made based on previous studies.

2.1 Efficient Market Hypothesis

The efficient market hypothesis essentially claims that the stock market is extremely efficient in processing information about individual stocks and the whole security market (Malkiel, 2003). This is also associated with the idea: random walk. The logic behind this idea is that the stock prices immediately and fully reflect all available information in the market. Thus, the stock prices are independent and unpredictable. However, the efficient market hypothesis does not always hold, especially in the long run. Fama and French (1988), Poterba and Summers (1988) argue that the stock market is not perfectly efficient, stock prices are dependent and this relation is often negative, which indicated that the existence of abnormalities and a forecastable pattern of stock returns in the long-term horizon.

2.2 Previous Research on M&As

Often, researchers use event study to evaluate the M&A performance, which measures the impact of a specific event on firm value (MacKinlay, 1997). The

(6)

abnormal return is then the difference between actual firm return and the expected rate of return at time t, where the expected return is estimated based on asset models. The abnormal return is the part that cannot be explained by risk factors in the asset model, economists then conclude these unexplainable return as the contribution of M&As to the firm value creation.

Most papers on the M&A stock performance focus on the short-run abnormal returns surrounding the announcements. These results show that the short-run cumulative abnormal returns are generally negative and significant around several days before and after the announcements for the acquiring firms (Rani, Yadav and Jain, 2015; Kohli and Mann, 2012; Wansley, Lane and Yang, 1983; Piloff and Santomero, 1998). However, these short-run negative stock performance may not represent the long-term performance persistency (Focarelli and Panetta, 2003). They argue that, firstly, cost cuttings take time, implementing plans into practice and realization of synergies require time. Secondly, merging disparate workforces is not an easy task, differences in communication styles, customer needs, and distribution channels, etc. could also harm the exchange of information, may thus let M&A be a lengthy program. Overall, it is essential to take these time lags into consideration when studying the effect on the shareholders’ value created by M&A. Interestingly, based on this argument, Andre, Kooli and L'her (2004) find an abnormal return of -0.745% at 0.01 significant level 36 months after the events. Andrade, Mitchell and Stafford (2001) and Gupta and Misra (2007) also obtain significant negative abnormal returns of -5% and -1.84% three years after the announcement in the U.S. market.

Hypothesis 1: Abnormal returns created by banking M&As are negative three-year after the event.

2.3 Deal-Specific Characteristics and Predictions from the Prior Literatures 2.3.1 Cross-border M&As

Cross-border M&As can be treated as one of the direct investment strategies a firm can choose. The number of all cross-border M&A deals increased from 48% in 1991 to 75% in 2007 (table 3). This could also be an evidence that cross-border M&A plays a dominant role in international capital flows and the world globalization process (Kang and Johansson, 2000; Grave, Vardiabasis and Yavas, 2012). Campa and

(7)

Hernando (2004) further argue that, due to the decrease in barriers between EU countries, type of cross-border M&As would happened more frequently.

Table 3

The number of cross-border M&A deals and the total value of all from 1985 to 2018 in Europe.

Seth, Song and Pettit (2002) as well as Kohli and Mann (2012) believe that cross-border mergers and acquisitions outperform domestic ones theoretically, as besides the normal advantages created by M&As, they also subject to the benefits of market imperfections. Market imperfection theory explains how multinational corporations are able to take advantages of market distortions through cross-border M&As, which let shareholders to earn excess wealth (Kohli and Mann, 2012).

A common argument in the banking industry is that mergers and acquisitions, especially cross-border ones, have the potential to reduce banks’ risks of insolvency (Amihud, DeLong and Saunders, 2002). This type of risk reduction is carried by the M&As activities and is associated with geographic diversification. Thus, it became one of the reasons why mergers and acquisitions was a trend in 2007 and 2008 in the

0 5 0 00 1 0 00 0 1 5 00 0 N u m b e r o f D e a ls 0 5 0 0 00 0 1 0 0 0 00 0 1 5 0 0 00 0 2 0 0 0 00 0 V a lu e o f D e a ls 1970 1980 1990 2000 2010 2020 Announcement Date

(8)

banking sector, that was, the urgent need to reduce the risks emphasized on them by the economic crisis (Focarelli and Panetta, 2003). Moreover, cross-border mergers and acquisitions present significant opportunities for firms, not only to diversify their activities geographically, gain access to valuable resources but also to expand the bank’s current businesses (Chen and Young 2010). Hence, the cross-border banks are supposed to benefit more following this trend. Cybo-Ottone and Murgia (2000), Houston and Ryngaert (1994), Kohli and Mann (2012) find positive CAARs in cross-border M&As in 12 months period post events.

However, some researchers argue that the cross-border acquirers are not always outperforming domestic ones, the value creations are less affected by nationalities of the involving firms, instead, relying more on other deal related characteristics (Campa and Hernando, 2004; Chen and Young, 2010; Dutta and Jog, 2009). Additionally, the country barriers still exit in Europe, which introduce a likelihood of decrease in the value generation (Campa and Hernando, 2004). Chen and Young (2010), Dutta and Jog (2009) and Campa and Hernando (2004) find significantly lower abnormal returns of -0.44%, -0.7% and -8.2% respectively at the 5% significant level for cross-border M&As. Amihud et al. (2002) find significantly negative CARs (+10, +260) days after the announcements for the European cross-border acquirers. The existing literatures give different opinions regarding value creations of domestic and cross-border acquiring firms, thus, the direction of the impact of the cross-border factor on abnormal return is ambiguous.

Hypothesis 2: Abnormal returns generated by cross-border M&As and domestic ones are significantly different in the long run.

2.3.2 Payment Methods of M&As

The M&A-related abnormal returns can be strongly impacted by choices of financing methods. Bidders often face a tradeoff between corporate control threats concerns issuing equity and rising financing distress costs of issuing debt, which implies a choice between debt financing and stock financing (Faccio and Masulis, 2005). Also, in the research of Myers and Majluf (1984), stock issuance always reduces stock prices, and this impact is significant. Furthermore, given that shareholders are not willing to issue new equity because of the dilution of voting power combined with a

(9)

reduction of stock price, the equity financing will give a bad signal of firm’s condition. Thus, the stock price will be even lower than expected.

In addition, Dutta and Jog (2009) and Faccio and Masulis (2005) further argue that, the decision on payment method of M&A is also related with the market values of both the bidder and target. If the bidder believes that its own current market value is over-valued, then they might tend to offer cash or a larger proportion of cash to signal its health condition to the market (Fuller et al., 2002). Moreover, if the bidding firm thought that the target’s market value is fair or there is a huge uncertainty related to the target’s intrinsic value based their own information, they will offer stock instead (Myers and Majluf, 1984). Franks et al. (1991), Loughran and Vijh (1997), Fuller et al. (2002) and Dutta and Jog (2009) report significant negative abnormal returns of -0.7%, -0.5%, -6.1% and -4.3% respectively for all equity-financing deals in a threeyear horizon after the events. Houston and Ryngaert (1994) get a significant result of -2.49% for the European banking industry.

Nevertheless, Wansley, Lane and Yang (1983) argue that tax benefits are stronger for M&As involving equity issuance, as the equity gain taxes will be deferred until the date that all stocks are sold, whereas, for cash financing M&As, the relevant taxes will be collected in that particular accounting year. Based on this, they hypothesize that the cumulative abnormal returns would be higher for stock financing M&As. Nonetheless, the research result is not in line with their hypothesis. Hence, a negative effect of equity financing on abnormal return is expected.

Hypothesis 3: Abnormal returns are lower for equity financing acquirer banks than cash/mixed financing ones in the long run.

2.3.3 Large or Small Targets

Small targets may have minor effects on the acquirers’ returns (Fuller et al., 2002). As the synergies and benefits carried by small targets are small compared to large targets (Dutta and Jog, 2009). Also, the management issues and all relevant expenses are different in every aspects. Thus, different impacts of small and large targets on acquirers’ abnormal returns are expected. Fuller et al. (2002) find that the cumulative abnormal returns are increasing from small to large. Dutta and Jog (2009) also give a

(10)

conforming result, that the large targets bring on average 65% higher abnormal returns than small ones. Instead, Franks et al., (1991) find an contrast result, illustrates that large targets appear to have negative effects on post-merger firm performance.

Hypothesis 4: Abnormal returns are higher for large targets.

2.3.4 Public or Private Targets

Capron and Shen (2007) and Mantecon (2008) introduce a phenomenon called “the private firm discount”. The private targets generally are purchased at a 20-30 percentage discount. This mainly due to the different information availability on private vs. public targets (Capron and Shen, 2007). Usually, the information quantity and quality for private targets are worse than public ones. Information on public firms are widely spread, whereas the managers of private firms are able to control the information they are willing to share.

However, this less available information characteristic benefits the value creation for acquirers. Firstly, the bidder competition will be lower for private targets. The lack of visibility, transparency, and market price associated with the private target give bidder more bargaining power (Koeplin, Sarin and Shapiro, 2000; Faccio and Masulis, (2005). Furthermore, the information asymmetry puts bidders under the risk of overpay. Thus, both Shleifer and Vishny (1986) and Dutta and Jog (2009) agreed on the idea that offers for private targets would be discounted due to adverse selection and illiquidity. Thirdly, the lower publicity on the M&A process makes the private information less likely to be dissipated. Mantecon (2008) find a significant negative result of -5.5% for deals with public targets. Thus, private target is expected to have a positive effect on abnormal returns.

Hypothesis 5: Abnormal returns are higher for private targets.

2.3.5 M&As happened in Great Recession

All benefits behind M&As, no matter cross-border or domestic ones, are particularly alluring to firms when times are tough. That is why the deal amounts and values have surged in the recession period (Amihud et al., 2002). In addition, Grave et al. (2012) further claim that, the M&A landscape was fully changed by the Great Recession. Strong companies will behave to buy other companies to create more competitiveness and cost-efficient, thus, theoretically, generate more shareholder value (MacKinlay, 1997). Meanwhile, targets will often agree to be purchased when they face insolvency

(11)

problems, which also encourage the M&As to take place. Thus, the M&As happened in recession period are expected to have unique impacts on value creation.

Hypothesis 6: Value creations are different between M&As happened in the Great Recession and in other periods.

2.3.6 Other Characteristics

In addition to the factors mentioned above, there are several other M&A-related characteristics that have impacts on the abnormal returns. For example, relatedness of bidder and target, hostile or friendly offers, etc.. However, in this thesis, these factors are omitted, as all deals are related and friendly in our sample set. So, no result can be derived for those factors.

3. Data and Methodology

3.1 Data

This thesis is investigating the effects of M&A-related factors on the value creations for the European acquiring banks three-year after the events. The events defined in this research are the announcement dates of M&As. These announcement dates and deal characteristics are collected on Thomson SDC Platinum database. The deal values, acquirers’ and targets’ equity values are reached on Bureau van Dijk database. Additionally, the monthly stock prices of acquiring banks are obtained in CRSP database. The Kenneth French’s website is used for the Fama-French 3 research factors.

Samples satisfy the following criteria:

a. The M&As are announced between 01/01/2006 and 12/31/2009. b. All deals included in the sample must be completed.

c. The acquirer has share price information available at least since the M&A announcement date, otherwise will be removed from the sample.

d. Both bidder and target have their accounting information available three years after the event, otherwise will be removed from the sample.

e. Both bidder and target are located in Europe. f. Both bidder’s and target’s mid industry are banks.

(12)

g. All deal values are above €400,000.

h. All deals that lack of relevant information will be excluded from the sample.

After the primary collection of data, there are 632 deals left in the European banking industry provided by the Thomson Reuters DataStream. After filtering the sample gradually, there are 35 deals remaining that satisfy all sample criteria.

Table 4(1)

Statistic description

Bid-characteristics Domestic (N) Cross-border (N) All (N)

Stock financing M&As 6 8 14

Large targets 12 13 25

Public targets 17 17 34

Crisis 10 10 20

Table 4(2)

Variable Observations Mean SD Min. Max.

Firm returns 1295 -0.0042 0.1581 -0.5656 0.8835 Market returns 1295 0.1670 7.5076 -22.02 13.67 SMB 1295 0.0211 2.2267 -4.98 4.88 HML 1295 -0.2461 0.7006 -4.2 7.42 Risk-free rate 1295 0.0682 0.1218 0 0.44 3.2 Methodology

In order to examine the long-term performance of mergers and acquisitions on the acquiring banks’ stock prices, the event study methodology is employed. Event study plays an vital role in testing whether the nonzero abnormal stock return is persistent with a particular corporate event, especially in a long run period (Khotari and Warner, 2007). Generally, there are two methods that are available for testing the long-horizon abnormal returns, buy-and-hold abnormal return (BHAR) and the calendar-time portfolio approach.

BHAR is one of the most famous and widely used approaches in analyzing the long-term M&A effect on acquiring firms. The BHAR measures the return difference

(13)

between the event firm and the market benchmark. However, Mitchell and Stafford (2001) argue that BHAR subjects to new listing bias, rebalancing bias and skewness problem, which destroy the reliability of BHAR approach in calculating the long-term abnormal return. In addition, an appropriate benchmark is more critical for the long-horizon test, the estimation of the long-term abnormal return is more sensitive to the choice of benchmark than a short-run analysis (Khotari and Warner, 2007). Khotari and Warner (2007) further argue that even a small error in risk adjustments can lead to a significant difference in the results of abnormal returns. Moreover, Franks et al. (1991) state that the CRSP equally-weighted and value-weighted single indexes are inefficient, that is why many studies wrongly concluded that long-term abnormal return is significant. Therefore, in this thesis, the calendar-time portfolio approach is employed.

3.2.1 Calendar-time Portfolio Approach

The calendar-time portfolio approach looks into the long-term abnormal return by measuring stock performance in calendar time relative to some benchmarks (Mitchell and Stafford 2001). The calendar-time portfolio approach outperforms BHAR in several aspects. Firstly, both skewness and cross-correlation problems are less severe for the calendar-time portfolio approach than BHAR, which often lead to false rejection of the null hypothesis, even there is no existence of abnormal returns (Kothari and Warner, 2007). Furthermore, the misspecification of the expected return appeared more commonly under BHAR compared with the calendar-time portfolio approach (Fama and French, 1988). Therefore, the power of the calendar-time portfolio approach is stronger relative to BHAR. The model (Fama and French, 1988) is shown below:

𝑅_#,%− 𝑅_',%= 𝛼_#+ 𝛽_,× .𝑅_/,%− 𝑅_',%0 + 𝑠#𝑆𝑀𝐵%+ ℎ#𝐻𝑀𝐿%+ 𝜖#,% (1) The left-hand side stands for the monthly return of firm i (𝑅#,%) over the risk-free rate (𝑅_',%) at time t. The independent variable in this regression is the excess market return and two investment portfolios that mimic the common risk factors, namely, Fama-French 3 research factors: market factor (𝑅/,%), size factor (𝑆𝑀𝐵%), and book-to-market factor ( 𝐻𝑀𝐿_%) (Fama and French, 1988; Andre et al., 2004). 𝛼_# is the error

(14)

term, measures excess return relative to the risk adjustment factors, which is the average monthly abnormal return for bank i at time t.

Under this approach, firstly, the monthly calendar-time portfolio returns are calculated for banks experienced the event. Next, the monthly abnormal return is calculated as the intercept of the three factor model. Fama-French three research factors are available for the European market, which is convenient for our analysis.

The abnormal return is then calculated as follows:

𝐴𝑅_#,% = 𝑅_#,%− 𝛽<_#× .𝑅_/,%− 𝑅_',%0 − 𝑠̂#𝑆𝑀𝐵%− ℎ>#𝐻𝑀𝐿%+ 𝑅',% (2) Where, A𝑅#,% is the monthly abnormal returns for acquiring bank i in time t. 𝛽<# is the coefficient of market factor. 𝑠̂# is the coefficient of size factor. ℎ># is the coefficient of book-to-market factor. By using this approach, Andre et al. (2004), Dutta and Jog (2009), Franks et al., (1991) find significant 𝛼_# of -0.479%, 0.4% and -1.1% for Canadian and the U.S. acquiring firms respectively.

3.2.2 Cross-Sectional Regression Analysis

The cross-sectional regression of aggregate consolidation gains is conducted based on various deal-related characteristics illustrated in section 2 that have impacts on the abnormal returns theoretically and empirically. The purpose of this regression is to find some inter-group differences, and it is helpful when multiple hypotheses exist (MacKinlay, 1997). The abnormal returns generated under the calendar-time portfolio approach is the dependent variable for the regression. The main deal-specific factors are the independent variables (i.e., cross-border, payment methods, large or small targets, public or private targets and crisis). In addition, two other dummies are included: the time dummies, which allow us to compare the abnormal return by years.

The firm-specific characteristics are treated as control variables in this analysis. The reason is that, according to MacKinlay (1997), the abnormal return generated by the occurrence of the event not only depends on the valuation effects of the event but also relates to the forecast of the likelihood of the event occurring by rational investors, which depends on the firm-specific factors. Therefore, adding these firm-relative control variables could make the regression more robust.

(15)

The control variables considered in this paper are deal value, relative deal size, relative equity size and return on equity (ROE). The relative size of deal value and equity of the target to the acquirer’s equity are crucial to the value creation, which measure the relative effects of target’s shareholders on the acquirer’s shareholders (Fuller et al., 2002; Gupta and Misra, 2007; Dutta and Jog, 2009; Capron and Shen, 2007; Franks et al., 1991). Besides, relative equity size is the numeric measure of the large/small factor, which is indispensable when evaluating the value of abnormal return. Moreover, ROE measures the firm’s profitability, which shows how effective the firm is in generating profits by using the funds invested by shareholders. Thus, it also has enormous impacts on the firm returns.

The main regression is presented below:

A𝑅_#,%=

𝛼# + 𝛽,∗ 𝐶𝑟𝑜𝑠𝑠𝑏𝑜𝑟𝑑𝑒𝑟#,% + 𝛽F∗ 𝑃𝑎𝑦𝑚𝑒𝑛𝑡#,% +𝛽M ∗ 𝐿𝑎𝑟𝑔𝑒#,%+ 𝛽O ∗ 𝑃𝑢𝑏𝑙𝑖𝑐#,%+ 𝛽T ∗ 𝐶𝑟𝑖𝑠𝑖𝑠_#,%+ 𝛽_U∗ log (𝐷𝑒𝑎𝑙 𝑣𝑎𝑙𝑢𝑒_#) + 𝛽_\∗ 𝑙𝑜𝑔.𝑅𝑒𝑙𝑎𝑡𝑖𝑣𝑒 𝐷𝑒𝑎𝑙𝑠 𝑆𝑖𝑧𝑒_#,%0 + 𝛽_^∗ 𝑙𝑜𝑔.𝑅𝑒𝑙𝑎𝑡𝑖𝑣𝑒 𝐸𝑞𝑢𝑖𝑡𝑦 𝑆𝑖𝑧𝑒_#,%0 + 𝛽_a∗ 𝑅𝑂𝐸_# + 𝛽_,c∗ 𝑀𝑜𝑛𝑡ℎ_#,,d,F+ 𝛽_,,∗

𝑀𝑜𝑛𝑡ℎ#,,MdFO+ 𝜀#,% (3)

The independent variables and control variables are given as follows:

Table 5

Illustration of explanatory variables.

Variables Definition of Variables

𝑃𝑎𝑦𝑚𝑒𝑛𝑡#,% dummy variable equals 1 if payment method of deal i is stock

issuance, equals 0 otherwise.

𝐶𝑟𝑜𝑠𝑠𝑏𝑜𝑟𝑑𝑒𝑟#,% dummy variable equals 1 if deal i is a cross-border deal, equals

0 otherwise.

𝑃𝑢𝑏𝑙𝑖𝑐#,% dummy variable equals 1 if the target engaged in deal i is public

traded.

𝐿𝑎𝑟𝑔𝑒#,% dummy variable equals 1 if the target engaged in deal i is large.

(16)

2008, equals 0 otherwise (year 2006, 2009). 𝑀𝑜𝑛𝑡ℎ#,% dummy variable equals 1,2,3: group the abnormal returns into 3

individual years. 𝑙𝑜𝑔.𝐷𝑒𝑎𝑙𝑠 𝑉𝑎𝑙𝑢𝑒#,%0 control variable logarithm of (deal value)

𝑙𝑜𝑔.𝑅𝑒𝑙𝑎𝑡𝑖𝑣𝑒 𝐷𝑒𝑎𝑙𝑠 𝑆𝑖𝑧𝑒#,%0 control variable logarithm of (deal value/

acquirer′s equity value) 𝑙𝑜𝑔.𝑅𝑒𝑙𝑎𝑡𝑖𝑣𝑒 𝐸𝑞𝑢𝑖𝑡𝑦 𝑆𝑖𝑧𝑒#,%0 control variable logarithm of (target equity value/

acquireru_{s equity value)}

𝑅𝑂𝐸#,% control variable return on equity for firm i

3.2.3 Robustness Checks

The robustness of the main cross-sectional regression could be examined by omitting or adding variables into the regression. The widely accepted reason of conducting robustness checks is to demonstrate that the main regression is robust no matter how assumptions and variables change (Du and Sim, 2016). Hence, follows the main regression analysis, the robustness checks will be conducted. The first robustness check looks into how important the choice of the sample period is to our analysis, the three-year research period is substituted by a 12-month period. Next, in the second robustness check variables are omitted or added gradually.

4. Empirical Results

4.1 Calendar-time Portfolio Approach

Table 6

Abnormal returns (αw) by groups of banks in period (1 month, 36 month).

Note: Regress Rw,y− Rz,y= αw+ β,× .R|,y− Rz,y0 + swSMBy+ hwHMLy+ ϵw,y, for all three groups

(17)

standard deviation in parentheses. ∗∗∗ significant at 0.01 level, ∗∗ significant at 0.05 level, ∗ significant at 0.1 level.

Table 6 shows significantly negative results 36 months following the events, for all three groups of banks, with abnormal returns of -4.18%, -10.29% and -7.39% respectively at the 1% significant level. This is in line with all literature that the abnormal return three-year after is significantly negative (Andre et al., 2004; Andrade et al., 2001; Chen and Young, 2010; Dutta and Jog, 2009; Houston and Ryngaert, 1994).

Table 7

Abnormal returns (αw) by periods for all banks.

Note: Regress Rw,y− Rz,y= αw+ β,× .R|,y− Rz,y0 + swSMBy+ hwHMLy+ ϵw,y, in sub period 1-12

months, 13-24 months, 25-36 months, two-year and three-year after the event respectively. ∗∗∗ significant at 0.01 level, ∗∗ significant at 0.05 level, ∗ significant at 0.1 level.

This table gives the trend of abnormal returns for acquiring banks. Table 7 illustrates the results of calendar-time portfolio approach for all M&A banks in all sub-periods: 1-12 months, 13-24 month, 25-36 months following the events, with abnormal returns of −0.1494∗∗∗ _,_−0.061∗∗∗ _, _{−0.0107 respectively. The abnormal return was} increasing and became from significant to insignificant since period 1-12 months to the period 25-36 months. Also, in the periods of two- and three-year post the events, banks were with the abnormal return of −0.1056∗∗∗_{and −0.0739}∗∗∗_{, which also} evident the trend of raising. Overall, the results show that the contribution of M&As on value creation in all sub-periods was negative and subjected to an increasing pattern.

(18)

4.2 Regression

Table 8

OLS regression for cross-sectional analysis.

Note: Standard coefficients for all independent variables with robust standard deviation in parentheses. Public dummy is omitted in domestic group. ∗∗∗ significant at 0.01 level, ∗∗ significant at 0.05 level, ∗ significant at 0.1 level.

Cross-border dummy: Cross-border dummy with coefficient of −0.066∗∗∗_{gives the} evidence that the abnormal returns of cross-border M&A banks are 6.6% statistically significantly lower than domestic M&A banks, which is in line with other empirical studies (Franks et al., 1991; Loughran and Vijh, 1997; Fuller et al., 2002; Dutta and Jog, 2009). To demonstrate this result, specifically in our sample, the country barriers and other deal-characteristics are in the dominant positions over the benefits carried by cross-border M&As, such as geographical advantages and market power.

Payment method dummy: The coefficient of payment method dummy is -0.019, and it

(19)

of banks financed by stock are 1.9% lower than those funded by cash or mixed for all M&As. These observations confirm the hypothesis of Myers and Majluf (1984), that, all stock issuances will lower the stock prices and the stock financing M&As will result in a lower level of abnormal returns. Notably, the coefficient for cross-border group (-0.019) is even greater than domestic ones (-0.032). It corresponds the idea of Faccio and Masulis (2005), they state that, due to home country bias and exposure to exchange risk, it is difficult to sell stocks to foreign investors, thus there is less demand for target’s stock. That is why less deals related to stock financing in cross-border consolidations. Hence, focus on this specific idea, cross-cross-border group has the potential to outperform domestic ones.

Large dummy: The Large dummy is significantly positive at a 0.05 significant level,

which is in line with the theory and hypothesis, that the larger targets would carry relatively more synergies, and then contribute more to the generation of abnormal returns. Interestingly, the coefficient (-0.011) is negative for cross-border group, which is in contrast with the hypothesis. This could be caused by the higher level of expenses associated with cross-border transportations. Besides, it is also corresponds with the study of Franks et al. (1991), who explain this phenomenon as the size-related bias.

Public dummy: The Public dummy gives insignificant coefficient of -0.021. It has

minor effect on value generation, which is in contrast to the expectation. The cause of this could be that the sample size is too small to evaluate the information effect.

Crisis dummy: Crisis dummy overall has an insignificant negative impact of -0.4% on

abnormal returns. It only shows a significant result of -6% at a 0.01 significant level for domestic consolidations, which demonstrates that the abnormal returns are significantly lower for domestic M&As occurred in 2007 or 2008 than those happened in other years (2006/2009). Instead, it has an insignificant negative coefficient of -0.02 for cross-border M&As.

Month dummy: The Month dummy gives differences between abnormal returns over

time. The abnormal returns generated in the first year and second year post the events are 14.1% and 4.6% significantly lower than the third year. It is also in line with the conclusion in table 7, that, the abnormal returns are in raising patterns.

(20)

4.3 Robustness Checks

Table 9

First robustness check: substituting the sample period from 36 months to 12 month post events.

From the first regression to the second, except for Crisis and Public dummy, none of the coefficients have significant changes. Several variables become relatively less significant in regression 2, such as Payment and Large dummies, suggests that the influences of these deal-specific factors on abnormal returns have decreased over time (Kohli and Mann, 2012). Moreover, the R-squared decreases from 0.355 to 0.196, means that the proportion of abnormal returns that can be explained by this model decreases. This is also reasonable according to Du and Sim (2016), as there are more firm-specific changes than the deal-relative factors that could affect the abnormal returns from time to time.

Table 10

(21)

As shown in table 10, by omitting variables of Public and Crisis dummies (1) or adding 100% Acquisition dummy (3) from and into the main regression (2), the adjusted R-squared barely changed. This gives a signal that the original regression is relatively robust, as its explanation power would not be influenced by the changing of treatments (Kohli and Mann, 2012).

4.4 Summary of Hypotheses

Table 11

Hypotheses Related variables Results

Hypothesis 1 Abnormal returns Supported

Hypothesis 2 Cross-border Supported

Hypothesis 3 Payment methods Supported

Hypothesis 4 Large Supported

Hypothesis 5 Public Rejected

Hypothesis 6 Crisis Rejected

Hypothesis 1 is supported, abnormal returns are significantly negative in our sample

(22)

Hypothesis 2 is supported, abnormal returns for cross-border M&As are significantly

different from and lower than domestic ones.

Hypothesis 3 is supported, abnormal returns of M&As financed by stock issuances are significantly lower than M&As that were financed by debt or mixed.

Hypothesis 4 is supported, large targets contribute more to the generation of abnormal

returns.

Hypothesis 5 is rejected, there is no statistical evidence shows that private targets are

outperforming public ones.

Hypothesis 6 is rejected, only M&As happened in 2007 or 2008 domestically had a

significantly lower abnormal return than in other years.

5. Conclusion

In this thesis, the long-term abnormal returns of European acquiring banks are examined by using a sample of 35 M&A deals during the period of 2006-2009. The research question of this thesis could be answered: how much abnormal returns are generated through the M&A processes, how different deal-specific factors influence the abnormal returns, and how these impacts different from cross-border and domestic mergers and acquisitions.

The long-term abnormal returns are calculated by using the Fama-French three-factor model, and the result of -7.39% is broadly consistent with relevant literatures discussed in this paper. Secondly, all abnormal returns in different sub-periods are obtained, results are from -14.94% to -1.07% and become from significant to insignificant. Both of these two statistical evidences show that the abnormal returns are in raising patterns. Then, abnormal returns for cross-border and domestic acquiring banks are calculated individually, domestic bidders outperform cross-border ones by 6.6%.

Further, a cross-sectional analysis of abnormal returns is conducted, several deal-related factors: Cross-border, Payment method, Large target, Public target and Crisis, are proposed in this paper. Some strong evidences are found: Cross-border dummy,

(23)

Payment dummy and Large dummy have statistically significant effects on abnormal returns, whereas Public and Crisis dummies are insignificant. Cross-border M&As underperform domestic ones three-year post the events in our sample. Interestingly, some of these impacts have the same significances and magnitudes in both cross-border and domestic groups, however, some are different. Such as, Payment dummy shows statistically significantly negative effect in both groups, whereas, Crisis dummy only has significant effect in the domestic group. Still, these results as a whole are in line with most of the papers on M&A performance studies. There are two robustness checks employed in this paper. First one substitutes the sample period to a shorter one, the second check is conducted by excluding or adding variables into the regression. Both robustness checks show that the main model in this paper is relative robust.

Still, due to the limited sample size, some variables cannot be interpreted by this paper, such as the effect of public targets on domestic abnormal returns. It is possible that the results obtained in this paper are subjected to the choice of benchmarks. Although the benchmark employed here is the three-factor model, which is more robust than equally-weighted and value-weighted single indexes. Besides, there are 11 independent variables in the cross-sectional regression, which raise the explanation power of this model. However, it is difficult to control all related variable that affect the abnormal returns. Thus, there are still some influential variables are left. Suggestions for future researches, a broader sample size is helpful in getting a more powerful result, same as suggested by Dutta and Jog (2009). Secondly, test abnormal returns by using several benchmarks will reduce the chance of expected returns misspecifications. Additionally, more control variables could be included and a more complex model could be conducted.

(24)

Reference:

Abbasoğlu, O. F., Aysan, A. F., & Gunes, A. (2007). Concentration, competition, efficiency and profitability of the Turkish banking sector in the post-crises period.

Amihud, Y., DeLong, G. L., & Saunders, A. (2002). The effects of cross-border bank mergers on bank risk and value. Journal of International Money and

Finance, 21(6), 857-877.

Andrade, G., Mitchell, M., & Stafford, E. (2001). New evidence and perspectives on mergers. Journal of economic perspectives, 15(2), 103-120.

Andre, P., Kooli, M., & L'her, J. F. (2004). The long-run performance of mergers and acquisitions: Evidence from the Canadian stock market. Financial

Management, 27-43.

Campa, J. M., & Hernando, I. (2004). Shareholder value creation in European M&As. European financial management, 10(1), 47-81.

Capron, L., & Shen, J. C. (2007). Acquisitions of private vs. public firms: Private information, target selection, and acquirer returns. Strategic management

journal, 28(9), 891-911.

Chen, Y. Y., & Young, M. N. (2010). Cross-border mergers and acquisitions by Chinese listed companies: A principal–principal perspective. Asia Pacific

Journal of Management, 27(3), 523-539.

Cybo-Ottone, A., & Murgia, M. (2000). Mergers and shareholder wealth in European banking. Journal of Banking & Finance, 24(6), 831-859.

Dutta, S., & Jog, V. (2009). The long-term performance of acquiring firms: A re-examination of an anomaly. Journal of Banking & Finance, 33(8), 1400-1412. Du, K., & Sim, N. (2016). Mergers, acquisitions, and bank efficiency: Cross-country

evidence from emerging markets. Research in International Business and

Finance, 36, 499-510.

Faccio, M., & Masulis, R. W. (2005). The choice of payment method in European mergers and acquisitions. The Journal of Finance, 60(3), 1345-1388. Focarelli, D., & Panetta, F. (2003). Are mergers beneficial to consumers? Evidence

from the market for bank deposits. American Economic Review, 93(4), 1152-1172.

Franks, J., Harris, R., & Titman, S. (1991). The post-merger share-price performance of acquiring firms. Journal of Financial economics, 29(1), 81-96.

(25)

Fama, E. F., & French, K. R. (1988). Dividend yields and expected stock

returns. Journal of financial economics, 22(1), 3-25.Fuller, K., Netter, J., & Stegemoller, M. (2002). What do returns to acquiring firms tell us? Evidence from firms that make many acquisitions. The Journal of Finance, 57(4), 1763-1793.

Grave, K., Vardiabasis, D., & Yavas, B. (2012). The global financial crisis and M&A. International Journal of Business and Management, 7(11), 56. Gupta, A., & Misra, L. (2007). Deal size, bid premium, and gains in bank mergers:

The impact of managerial motivations. Financial Review, 42(3), 373-400. Harkavy, O. (1953). The Relation between retained earnings and common stock

prices for large, listed corporations. The Journal of Finance, 8(3), 283-297. Houston, J. F., & Ryngaert, M. D. (1994). The overall gains from large bank mergers.

Journal of Banking & Finance, 18(6), 1155-1176.

Jensen, Michael C., The Free Cash Flow Theory of Takeovers: A Financial Perspective on Mergers and Acquisitions and the Economy. "The Merger Boom", Proceedings of a Conference sponsored by Federal Reserve Bank of Boston, pp. 102-143, October 1987. Available at SSRN:

https://ssrn.com/abstract=350422

Kang, N. H., & Johansson, S. (2000). Cross-Border Mergers and Acquisitions. Kohli, R., & Mann, B. J. S. (2012). Analyzing determinants of value creation in

domestic and cross border acquisitions in India. International Business

Review, 21(6), 998-1016.

Koeplin, J., Sarin, A., & Shapiro, A. C. (2000). The private company discount. Journal of Applied Corporate Finance, 12(4), 94-101.

Kothari, S. P., & Warner, J. (2007). Econometrics of event studies. Handbook of

empirical corporate finance, 1, 3-36.

Loughran, T., & Vijh, A. M. (1997). Do long-term shareholders benefit from corporate acquisitions?. The Journal of Finance, 52(5), 1765-1790. Lyon, J.D., Barber, B.M., Tsai, C., 1999. Improved methods for tests of long-run abnormal stock returns. The Journal of Finance 54, 165–201.

MacKinlay, A. C. (1997). Event studies in economics and finance. Journal of

economic literature, 35(1), 13-39.

Malkiel, B. G. (2003). The efficient market hypothesis and its critics. Journal of

(26)

Mantecon, T. (2008). An analysis of the implications of uncertainty and agency problems on the wealth effects to acquirers of private firms. Journal of

Banking & Finance, 32(5), 892-905.

Mitchell, M. L., & Stafford, E. (2001). Managerial decisions and long-term stock price performance. Journal of business, 73.

Myers, S. C., & Majluf, N. S. (1984). Corporate financing and investment decisions when firms have information that investors do not have. Journal of financial

economics, 13(2), 187-221.

Poterba, J. M., & Summers, L. H. (1988). Mean reversion in stock prices: Evidence and implications. Journal of financial economics, 22(1), 27-59.

Piloff, S. J., & Santomero, A. M. (1998). The value effects of bank mergers and acquisitions. In Bank Mergers & Acquisitions (pp. 59-78). Springer, Boston, MA.

Rani, N., Yadav, S. S., & Jain, P. K. (2015). Impact of Mergers and Acquisitions on Shareholders’ Wealth in the Short Run: An Event Study

Approach. Vikalpa, 40(3), 293-312.

Seth, A., Song, K. P., & Pettit, R. R. (2002). Value creation and destruction in cross-border acquisitions: an empirical analysis of foreign acquisitions of US firms. Strategic management journal, 23(10), 921-940.

Shleifer, A., & Vishny, R. W. (1986). Large shareholders and corporate control. Journal of political economy, 94(3, Part 1), 461-488.

Wansley, J. W., Lane, W. R., & Yang, H. C. (1983). Abnormal returns to acquired firms by type of acquisition and method of payment. Financial management, 16-22.

(27)

Appendix

Appendix 1: deal information

(28)

Appendix 2b:stata output all M&As, main regression

Appendix 3:correlation metrics

(29)

The effect of mergers and acquisitions on acquirer's long-term stock performance : evidence from the European Banking Industry in 2006-2009