Effect of board diversity on M&A performance:
Evidence from European banking
Anna Ostrá
s2178125
University of Groningen
Abstract
Using a dataset of 158 European bank-to-bank M&A deals completed between 1998 and 2014, present study illustrates the effects of board diversity in the market for corporate control. A significant positive effect between gender and tenure diversity on the acquiring bank’s board and cumulative abnormal returns for the bank around the M&A announcement was found. No significant relationship was found for nationality and age diversity. Furthermore, initial evidence suggests that more gender diverse boards chose targets that are able to improve their profitability relatively more in the post-merger period.
Keywords: board diversity; gender diversity; nationality diversity; age diversity; tenure diversity; banks; mergers; corporate governance; performance
JEL classification: G30, G34, G21, J16
Spring 2015
Master Thesis MSc Finance
Faculty of Economics and Business
2
1. Introduction
Gender diversity in the top positions in banking has received a lot of attention lately. Mostly in the media, but increasingly also in academic research. EU commissioners Neelie Kroes and Viviane Reding, former UK minister Harriet Harman, and IMF director Christiane Lagarde voiced their concerns that the recent financial crisis could have played out differently if there were more women in the bank boards of directors to mitigate the masculine behaviour persistent in banking. This was later named the Lehman Sisters hypothesis (van Staveren, 2014). Consistently, in August 2013 the European Central Bank announced gender quotas for the top levels of management. By 2019 women should occupy at least 28 percent of senior management positions and 35 percent of middle management positions, which represents a doubling with respect to the last targets (Businessweek, 2013).
There has clearly been a lot of discussion about whether more women should be encouraged to take top positions in banking. However, is there empirical proof that higher gender diversity in the boardroom enhances bank performance? The evidence so far is rather limited, because many corporate governance studies examining the effects of board diversity specifically exclude the banking sector, due to its special nature (Mateos de Cabo, Gimeno, and Nieto, 2012). There are number of reasons, why corporate governance in banks should be studied separately. Firstly, the activities of banks are very opaque and banks are very privy to privileged information on their clients, making it more difficult for outsiders to monitor and assess the risks and value of the assets of the bank, as compared to non-financial firms (Hagendorff, Collins, and Keasey, 2007). Secondly, the presence of safety nets elicits additional risk-taking by managers and investors and simultaneously disincentives creditors and depositors from monitoring (Hagendorff, Collins, and Keasey, 2007). Thirdly, it can be argued that well-functioning corporate governance in banking is also important for other industries. Well-functioning banks are necessary for the smooth operating of financial system and efficient providing of funds and banks also play a crucial role in corporate governance of other firms, as creditors or equity holders (Staikouras, Staikouras, and Agoraki, 2007).
3
banking setting is so far limited and does not provide conclusive evidence on the positive effect. Pathan and Faff (2013) find positive effects of board gender diversity on the performance of the banks only in the pre-Sarbanes-Oxley Act period and Hagendorff and Keasey (2012) do not find significant effect of board diversity on the market reaction to M&A announcements of US banks. In a very recent study García-Meca, García-Sánchez, and Martínez-Ferrero (2015), find a positive effect of board diversity in terms of gender on bank performance using a sample of nine countries.
The aim of my thesis is to closely examine the effect of board diversity in European banking, answering the call for further research stated in Hagendorff and Keasey (2012). In their paper they concentrate on the value of board diversity in banking in the US and present evidence from the market of corporate control. They document a positive effect of occupational diversity, a negative effect of tenure and age diversity, and no significant effect of gender diversity on the board of directors on the cumulative abnormal return around a merger1 announcement. However, they argue that European boards are larger, sometimes employing a two-tier structure, and have substantial union representation (resulting in less independence). For these reasons they encourage work outside US on this topic and claim that this will help researches in understanding the inter-linkages between different environmental contexts, board set-ups and diversity in creating value for shareholders.
The research question I want to answer in my thesis is: Is diversity on the board of directors of European banks value-enhancing in terms of the reaction to mergers and acquisitions announcements? Mergers and acquisitions are clearly one of the most important decisions the board has to take and in addition, concentrating on one specific strategic decision and using the event study methodology partly solves the endogeneity issue2 inherent to corporate governance studies (Hagendorff and Keasey, 2012). Although the main focus of this study (and also of many other preceding studies) is gender diversity, following the recent calls to consider multiple dimensions of diversity simultaneously, as each aspect of diversity might have different impact on the performance (García-Meca, García-Sánchez, and Martínez-Ferrero, 2015), this study also examines the impact of some less observable and less enforced types of diversity: nationality, age, and tenure diversity. Using event study approach I show that gender diversity on bank boards has a positive effect on performance in Europe, validating the findings of García-Meca,
1
The terms merger, acquisition and M&A are used interchangeably in the context of the present paper.
2 Endogeneity arises when board characteristics and performance are jointly determined and causes doubts
4
García-Sánchez, and Martínez-Ferrero (2015).This would constitute a clear proof that gender diversity in boardroom is value-enhancing and banks could justify their attempts to encourage more women to take the top positions. In other words, this would be a contribution for the economical argument for board diversity, claiming that higher diversity truly leads to enhancement of the value and performance through higher efficiency and internal working of the board and larger networks. This is in contrast with the ethical argument for board diversity, which sees diversity as an end in itself, and links it to the notion of equality of representation and fairness (Hagendorff and Keasey, 2012).
My thesis contributes to both the literature on board diversity and financial performance, and the bank M&A literature. Firstly, it relates the effects of board diversity to a specific corporate strategy and the impacts on firm value connected with it, rather than to general performance, partially avoiding the endogeneity issue. Secondly, my study considers more types of diversity simultaneously and is to my knowledge the first study of this type using European bank data. Thirdly, it sheds light on which corporate governance characteristics may explain returns to banks acquisitions. This is quite important, as the understanding of antecedents of value creation in bank M&A literature remains rather limited to date (Hagendorff and Keasey, 2012).
The rest of this thesis is structured as follows. The next section reviews existing literature, and describes both the underlying theoretical framework – resource dependence theory and agency theory and the empirical findings on the relationship between diversity and performance in the area of general research, banking and M&A. Furthermore, after a short summary of the literature, hypotheses are developed. Section 3 provides details on the method used. A combination of an event study and an OLS regression is used in the present study. Section 4 gives a description of the data used, while Section 5 discusses the findings of the analysis. Section 6 supplements the findings with a complementary analysis and examines if more diverse boards exhibit better target-picking skills. Section 7 provides a conclusion, describes the limitations of the present study and gives suggestions for further research.
2. Literature review
5
research on the nationality, age and tenure diversity is discussed. Finally, findings of existing literature are summarized and based on them hypotheses are developed.
2.1. Theoretical framework
Corporate governance research regarding diversity is not based on a single coherent theory, but draws on more perspectives, including resource dependence theory and agency theory (Hagendorff and Keasey, 2012). Nevertheless, both provide a rationale for a positive relationship between board diversity and corporate performance and only differ in the view on how this is achieved.
Resource dependence theory regards the board of directors as a link between the firm and the resources needed to maximize its performance. The three main benefits provided by the board linkages are advice/counsel, legitimacy, and communication channels (Pfeffer and Salancik, 1978). Carter, Simkins, and Simpson (2003) argue that more diverse boards understand the complexities of the environment better, and can therefore give better advice or make better informed decisions. Furthermore, drawing on the ethical argument for diversity as stated in Van der Walt and Ingley (2003), where board diversity is linked to equality of representation and non-exclusion of certain groups, diverse boards are seen as fair and creating legitimacy for the firm. Moreover, more diverse boards imply larger networks and more contacts and communication channels, therefore linking the firm to more diverse set of external stakeholders, improving the efficiency of communication and creating value for the firm (Hillman, Cannella, and Paetzold, 2000).
Agency theory sees the main role of board of directors in protecting the interests of shareholders (principals) by monitoring management (agents) and eliminating the possibility of self-interested behaviour (agency costs) of the management (Jensen and Meckling, 1976). According to Fama (1980), a board can fulfil this monitoring role only if they provide impartial and high-quality advice. Prior research shows that more diverse groups are more creative and innovative and in turn their decisions are of higher quality (Maznevski, 1994). Diversity is also found to lead to more activist boards, which address many issues that might be neglected by homogenous boards (Ely and Thomas, 2001). However, increased monitoring might not always lead to better financial performance. Adams and Ferreira (2009) find that more gender diverse boards tend to monitor more, but this is not improving the overall performance of the firm.3
3
6
There might be more problems diverse boards have to face. Forbes and Miliken (1999) point out that more heterogeneous boards may encounter coordination problems due to different viewpoints or terminology/knowledge of its members. In European context, in nationality diverse boards the language might also be an issue. Piekkari, Oxelheim, and Randøy (2014) show that there are hidden costs of using a non-native language for the performance of the board. Furthermore, more diverse boards are likely to be less cohesive, resulting in less cooperation and possible conflicts among members (Forbes and Miliken, 1999).
2.2. Effects of gender diversity on performance in general research
Women are certainly different in some aspects relevant to financial and board decisions. Existing research shows that women seem to be more risk-averse (e.g., Sunden and Surette, 1998; Agnew, Balduzzi, and Sunden, 2003) and less overconfident (e.g., Barber and Odean, 2001; Niederle and Vesterlund, 2005) than men. Similarly, Rost and Osterloh (2010) show that the processing of information under conditions of uncertainty is better by women than by men. Therefore, diversity induced by higher participation of women can be expected to be value-enhancing along the lines of the arguments described above as both resource dependence and agency theory predict a positive effect of increased diversity on the corporate performance.
A positive effect of gender diversity in the boardroom on firm’s value in terms of Tobin’s Q was documented by Campbell and Mínguez-Vera (2008) for Spain and by Carter, Simkins, and Simpson (2003) for the United States, using a sample of 638 Fortune 1000 companies. Similarly, Erhardt, Werbel, and Shrader (2003) find a positive relationship between board gender diversity and firm performance in terms of ROA and ROI for a sample of 127 US companies. On the other hand, examining S&P 500, S&P MidCaps and S&P SmallCap firms, Adams and Ferreira (2009) find that gender diverse boards allocate more effort to monitoring, but the effect on firm performance is negative. In terms of the effects of female additions to the board, Campbell and Mínguez-Vera (2010) find a positive response of the stock price for Spain and Kang, Ding, and Charoenwong (2010) find the same effect for Singaporean firms. On the other hand, Farrell and Hersch (2005) do not find significant abnormal returns after the announcement of appointment of a female director for a sample of US firms.
2.3. Gender diversity in banking
7
previous section, this effect cannot simply be expected to hold in banking as well, and should be tested for separately.
Extant empirical literature on gender diversity in banking has mostly concentrated on risk-taking. For a sample of German banks, Berger, Kick, and Schaeck (2014) find that changes in board composition resulting in a higher proportion of female executives increase portfolio risk. Contrary to this, Muller-Kahle and Lewellyn (2011) find that US financial institutions acting as subprime lenders had relatively less gender diverse boards. Mateos de Cabo, Gimeno, and Nieto (2012) also observe that the proportion of women on a bank board is higher for lower-risk banks. Therefore, there is some evidence that gender diversity has an impact in the banking setting. Here an effect on risk-taking is documented; the effect on overall performance will be discussed below. Palvia, Vähämaa, and Vähämaa (2014) show that US banks with female CEOs hold more conservative levels of capital, controlling for bank risk. Furthermore, they also found evidence that small banks headed by a female CEO were less likely to fail than small banks headed by a male CEO during the financial crisis. Looking beyond the risk assessment at highest positions, Beck, Behr, and Guettler (2013) find that loans screened and monitored by female loan officers have a significantly lower arrear probability, controlling for loan officer’s experience, workload and differences in screening and selection.
Regarding the research on value implications of gender diversity on the board of directors in banking, Nguyen, Hagendorff, and Eshraghi (2015) do not find any significant value effect of the appointment of a female executive director in terms of cumulative abnormal returns after the announcement for a sample of US banks. Similarly, Pathan and Faff (2013) find that gender diversity in bank’s boardroom improves the performance only in the pre-Sarbanes-Oxley Act period (1997-2002), but that this effect diminishes in the post-Sarbanes-Oxley Act period (2003-2006), and the crisis period (2007-2011). García-Meca, García-Sánchez, and Martínez-Ferrero (2015) is the first study on bank board gender diversity to include also some European countries. Using a sample of 159 banks in nine countries, they find that gender diversity increases bank performance. In conclusion, the effect of gender diversity on the financial performance seems much less clear-cut in the banking literature compared to the general research.
2.4. Gender diversity in M&A
8
behaviour it can be expected that the reaction of the shareholders to the announcement of an M&A deal by a more diverse board will differ.
Research on the effects of board gender diversity with respect to mergers and acquisitions for firms outside of the banking sector found that a higher percentage of women on the board is negatively related to the number of acquisitions a firm is engaged in (Huang and Kisgen, 2013; Chen, Crossland, and Huang, 2014; Levi, Li, and Zhang, 2014), the acquisition size (Chen, Crossland, and Huang, 2014), the bid premium paid (Levi, Li, and Zhang, 2014) and positively related to announcement returns (Huang and Kisgen, 2013).
Banking M&A announcements are on average found to be value-destroying4 for the shareholders of the acquiring bank in the US (Hagendorff, Collins, and Keasey, 2007; DeYoung, Evanoff, and Molyneaux, 2009). The general consensus for European banking M&As is that there are positive abnormal returns for shareholders of the acquiring bank (Cybo-Ottone and Murgia, 2000; DeYoung, Evanoff, and Molyneaux, 2009), but newer studies, e.g. Campa and Hernando (2006) for period 1998-2002 and Beltratti and Paladino (2013) for period 2007-2011, find that there were zero abnormal returns for the shareholders of the acquiring bank around the announcement.
It seems that shareholders in Europe are also becoming more sceptic about the outcomes of mergers and acquisitions. Therefore, it is relevant to examine if this scepticism can be mitigated by sound corporate governance within the bank. As discussed previously, the aim of my paper is to look at a specific aspect of corporate governance, i.e. on the benefits of diversity of the board. In other words, do shareholders see M&A plans announced by diverse boards more favourably than plans of homogeneous (i.e. in this situation masculine) boards? Hagendorff and Keasey (2012) do not find a significant effect of gender board diversity on the abnormal returns for sample of 148 US bank merger announcements. However, the dynamics around mergers and acquisitions by European banks are very different from US banks and additional research on the relationship in a European context is still in its infancy. Hagendorff, Collins, and Keasey (2007) argue that the systematic differences in internal governance and external control mechanisms between the US (market-based economy with sophisticated investor protection rules) and Europe (mostly bank-based economies) may imply that identical corporate governance arrangements have different value implications in these two locations.
4 This is the result of the agency conflicts and non-profit maximization motivations for M&A (such as
9
2.5. Other types of diversity
The previous subsections discussed the effects of gender diversity on corporate performance. Here the findings on other types of diversity considered by this study – nationality, age and tenure, will be discussed.
While US literature tends to concentrate on ethnic diversity of the board (Carter, Simkins, and Simpson, 2003; Erhardt, Werbel, and Shrader, 2003), in Europe nationality diversity seems to be of higher importance (Oxelheim and Randøy, 2003; Ruigrok, Peck, and Tacheva, 2007). García-Meca, García-Sánchez, and Martínez-Ferrero (2015) argue that there are both positive and negative consequences of nationality diversity on the board. Given their different background, foreign directors bring different ideas, experience, and points of view to the table. Furthermore, they facilitate cross-border flows of knowledge and are able to link the firm to more potential investors. Oxelheim and Randøy (2003) show for a sample of Swedish and Norwegian firms, that having Anglo-American board members is value-enhancing. However, there are also some drawbacks; more nationality diverse boards are more likely to suffer from lower decision speed, misunderstandings and conflicts (Ruigrok, Peck, and Tacheva, 2007). Moreover, foreign directors might be less effective in monitoring the management, due to less extensive knowledge of country accounting rules, governance standards and management methods (Masulis, Wang, and Xie, 2012). In the context of European banking, which is also the focus of present study, García-Meca, García-Sánchez, and Martínez-Ferrero (2015) find that nationality diversity inhibits financial performance for a sample of 159 banks in 9 European countries.
10
2.6. Hypotheses development
Previous literature documents the beneficial effects of gender diversity on corporate boards on the performance of non-financial firms (Carter, Simkins, and Simpson, 2003; Erhardt, Werbel, and Shrader, 2003; Campbell and Mínguez-Vera, 2008; Campbell and Mínguez-Vera, 2010; Kang, Ding, and Charoenwong, 2010). The effect in banking seems to be very dependent on the region. For the US, gender diversity on the board does not seem to be value adding for banks (Hagendorff and Keasey, 2012; Pathan and Faff, 2013; Nguyen, Hagendorff, and Eshraghi, 2015). First evidence on European banking suggests that there could be a positive effect of gender diversity of the board on the bank performance (García-Meca, García-Sánchez, and Martínez-Ferrero, 2015). Combined with the fact, that M&As are an area of corporate decisions, where higher risk-aversion and lower overconfidence of women seems to make a difference on the outcomes (Huang and Kisgen, 2013; Chen, Crossland, and Huang, 2014; Levi, Li, and Zhang, 2014), an effect of gender diversity on the announcement return is expected, but the expected sign is unclear given the mixed results so far. Therefore, the hypothesis regarding gender diversity can be formulated as follows:
H1: Gender diversity on the board of a European acquiring bank impacts the abnormal returns
for the shareholders of the bank at the announcement of a bank merger or acquisition.
Literature on nationality diversity reaches conflicting results, documenting both benefits in terms of heterogeneity of ideas and cross-border knowledge transfer, but also costs in terms of coordination problems and missing local knowledge. In the context of European banking, a negative relationship between board nationality diversity and financial performance is expected, given the results of García-Meca, García-Sánchez, and Martínez-Ferrero (2015). Based on the theory and conflicting empirical evidence, the effect of nationality diversity on announcement returns can be positive or negative (with negative being slightly more likely here). Therefore the hypothesis regarding nationality diversity is formulated as follows:
H2: Nationality diversity on the board of a European acquiring bank impacts the abnormal
returns for the shareholders of the bank at the announcement of a bank merger or acquisition.
11
value-enhancing, pointing more into the direction of a negative effect. Nevertheless, the third hypothesis is stated as follows:
H3: Age diversity on the board of a European acquiring bank impacts the abnormal returns for
the shareholders of the bank at the announcement of a bank merger or acquisition.
Hagendorff and Keasey (2012) argue that more tenure diverse board combines the experience and expertise of longer-tenured directors with fresh perspective of newly appointed directors and therefore prevents the board from developing a certain “corporate mindset” and helps in critical appraisal of the merger proposal. On the other hand higher turnover might lead to less cohesion in the group, which is also documented by the findings of Hagendorff and Keasey (2012), who show that there is a negative relationship between tenure diversity and announcement returns. Therefore the last hypothesis to be tested can be formulated as follows:
H4: Tenure diversity on the board of a European acquiring bank impacts the abnormal returns
for the shareholders of the bank at the announcement of a bank merger or acquisition.
3. Methodology
This section describes the method used to test the hypotheses introduced in the previous section. It consists of two steps – an event study and an OLS regression. First of all, cumulative abnormal returns around the merger announcement are estimated using the event study methodology. The cumulative abnormal returns are then used as the dependent variable and the diversity measures as the explanatory variables in the subsequent OLS regression.
3.1. Event study methodology
Event study methodology will be used to test the proposed hypotheses. The rationale behind this method is that under the assumption of an efficient market, the stock price response to an M&A announcement relates to the estimated benefits of the deal. Following the methodology used in Hagendorff and Keasey (2012), I will estimate the following market model:
Rit = αi +βi Rmt +εit, (1)
12
Abnormal returns are defined as the difference between the actual returns on a day and the predicted returns using the estimated parameters ( and ), which are assumed to have been observed in case the event (i.e. the M&A announcement here) would not have happened.
=
Cumulative abnormal returns are estimated over an event window of 5 days (from 2 days before the announcement to 2 days after the announcement) and 21 days (from 10 days before the announcement to 10 days after the announcement), and are calculated as the sum of the abnormal returns on the days included in the respective event window.
=
3.2. Regression analysis
In the next stage these estimated cumulative abnormal returns are used as the dependent variable in an OLS regression.5 Explanatory variables are the board diversity measures. Gender diversity is defined as the percentage of female directors as in Erhardt, Werbel, and Shrader (2003), Adams and Ferreira (2009), Hagendorff and Keasey (2012) and García-Meca, García-Sánchez, and Martínez-Ferrero (2015). Nationality diversity is defined as the percentage of foreign directors, following García-Meca, García-Sánchez, and Martínez-Ferrero (2015). For age and tenure diversity, I construct the Pearson coefficient of variation, following Hagendorff and Keasey (2012). This is calculated as the standard deviation divided by the mean and captures the dispersion relative to the average on the board.
A number of control variables based on previous European banking M&A literature are introduced into the model. Profitability in terms of return on equity (ROE) of the acquiring bank in the previous year is found to have a positive effect on the abnormal returns (Hagendorff and Keasey, 2012; Beltratti and Paladino, 2013), and the relative size of the target to the acquirer is expected to have a negative effect (Campa and Hernando, 2006; Hagendorff and Keasey, 2012). From the deal descriptive variables, deal value is found to have a negative effect on the abnormal returns (Cybo-Ottone and Murgia, 2000; Hagendorff and Keasey, 2012) and domestic mergers (dummy variable taking value of 1, when acquirer and target are from the same country, 0 otherwise) are found to have a positive effect (Cybo-Ottone and Murgia, 2000; Campa and
5 Cornett, Hovakimian, Palia and Tehranian (2003) and Campa and Hernando (2006) claim that because
13
Hernando, 2006). The rationale behind these effects is that smaller and geographically focusing deals might be easier to integrate and should be therefore viewed more favourably at the announcement (Houston and Ryngaert, 1994). Furthermore, a dummy variable taking value of 1 when a bank completed more transactions within the examined period is introduced. DeLong and DeYoung (2007) show for a sample of 216 M&A deals of large US commercial banks that substantial learning effects are present, and therefore higher abnormal returns for repeated bidders are to be expected. Finally, a number of control variables describing the corporate governance need to be introduced. Boards with higher percentage of independent directors, i.e. directors that are in no way besides their directorship affiliated with the firm, are seen as making better acquisitions (Byrd and Hickman, 1992). This effect is also confirmed in the banking literature (Hagendorff and Keasey, 2012; Nguyen, Hagendorff, and Eshraghi, 2015), therefore a positive effect of the percentage of independent directors on the abnormal returns is expected. There is a consensus in the literature (see e.g. Hermalin and Weisbach, 2003) that large boards have negative effects on the firm value, because they are less efficient and agency problems (such as free-riding) are more likely to arise. This is also documented by Staikouras, Staikouras, and Agoraki (2007) for European banks, and therefore a negative relationship between board size6 and abnormal returns is expected. Furthermore, as the sample includes banks with both one-tier and two-tier boards, a dummy variable taking value of 1 for two-tier boards was included to account for the possible differences between these two structures.
The full econometric model to be tested can be formulated as follows:
C = α + β1 board diversity + β2 control variables + ε. (4)
4. Data
The M&A deals to be studied were identified in the Zephyr database, maintained by Bureau van Dijk. Following search strategy was used. The deal had to be a merger or an acquisition announced and completed between 01/01/1998 and 01/01/2015. This period was chosen to obtain as many observations as possible, as these are all the years covered in the Zephyr database. The deal had to result in a final stake of at least 50 percent for the acquirer, which is considered as a sufficiently high controlling stake in the M&A literature (e.g. Beitel, Schiereck, and Wahrenburg, 2004). Both the acquirer and the target have to be a bank (as defined by US SIC code 60). Furthermore, to be able to examine the returns, the acquirer has to be a listed bank and as the deal value is used as a control variable, it needs to be disclosed and available. Finally,
6
14
the acquiring bank has to be European, defined as EU (27) plus Norway and Switzerland, as in Beltratti and Paladino (2013), and there is no restriction on the location of the target. This search strategy resulted in 261 deals. Subsequently, 17 deals that included more than one target or more than one acquirer were eliminated. As suggested by Piloff and Santomero (1996) and performed by Cybo-Ottone and Murgia (2000) and Beitel, Schiereck, and Wahrenburg (2004) the sample is not purged from a few instances of repeated bidder activity. Piloff and Santomero (1996) argue that eliminating the deals by banks that were engaged in more than one transaction in given period would lead to an omission of the most relevant firms from the analysis, and therefore bias the results.
Data on the deals, such as the deal value, the deal announcement date, the location of the acquirer, and the target were also retrieved from Zephyr. Using the ISIN codes, data on the stock returns were retrieved from Datastream, together with the market index. As the market index the MSCI EUROPE Banks index was used.7 Data on bank variables (ROE and sizes of acquirer and target) were downloaded from Bankscope. As this information was missing for some banks, this resulted in an elimination of some additional deals, resulting in final sample of 158 deals. This constitutes about forty percent of the total sample8 of bank-to-bank M&A deals completed by European banks in the respective period. The geographic distribution of deals is illustrated in Table A1 in the appendix. France, Italy and Spain were the most active acquirers in my sample, with 20, 20 and 17 acquisitions completed within the observed period, respectively. Targets were located in 43 different countries, with the most being located in Italy (15), Denmark (13) and Norway (12).
The data on the composition of board (percentage of independent directors, percentage of female directors, percentage of foreign directors, ages and tenures of all board members), board type (one-tier/two-tier) and board size at the time of M&A announcement9 was hand-collected from the annual reports of the banks and BoardEx. When information on a specific director was missing in the annual report or BoardEx, this was retrieved from Lexis Nexis or Bankscope. Nevertheless, for some cases none of these four sources was able to provide the relevant information, resulting in missing observations on nationality diversity, age diversity, tenure diversity or percentage of independent directors for a few banks. Table 1 summarizes the variables used and provides an overview of the data sources.
7
I also tried estimating the market model using country specific bank indices constructed by Datastream, but the results were similar and are therefore not reported.
8
405 deals, also including deals with unknown value.
9
15 Table 1
Definitions and sources for the variables used.
Definition/calculation Source
5-day CAR Sum of the abnormal returns from 2 days before the announcement to 2 days after the announcement
Datastream
21-day CAR Sum of the abnormal returns from 10 days before the announcement to 10 days after the announcement
Datastream
Board diversity measures
Gender diversity Total number of female directors divided by total number of directors on the board BoardEx /AR Nationality diversity Total number of foreign directors divided by total number of directors on the board BoardEx /AR Age diversity Standard deviation divided by the mean age across the board BoardEx /AR Tenure diversity Standard deviation divided by the mean tenure across the board BoardEx /AR
Deal and bank variables
ROE Return on average equity in the year prior the deal announcement Bankscope Relative size (ln) Natural logarithm of the value of the assets of the target divided by the value of the assets
of the bidder
Bankscope
Deal value (ln) Natural logarithm of the value of the deal in thousand € Zephyr Domestic Dummy variable, 1 if bidder and target located in same country, 0 otherwise Zephyr Repeat bidder Dummy variable, 1 if bidder completed more deals in observed period, 0 otherwise Zephyr
Governance variables
Board independence Total number of independent directors divided by total number of directors on the board BoardEx /AR Board size (ln) Natural logarithm of the total number of directors on the board BoardEx /AR Two-tier board Dummy variable, 1 if the bank has two-tier board structure, 0 otherwise BoardEx /AR
Table 2 describes the market reaction to the announcement of the deal for the sample of 158 deals. Overall, in both event windows, there are no significant cumulative abnormal returns around the announcement. Approximately half of the banks experience positive CARs and the other half negative CARs; therefore it seems that European bank M&As are neither value-enhancing nor value-destroying in general, and specific bank or deal characteristics matter for the outcome. This is consistent with recent studies on European bank M&A (Campa and Hernando, 2006; Beltratti and Paladino, 2013) that also do not find any significant cumulative abnormal returns around the announcement.
Table 2
Cumulative abnormal returns around the announcement.
This table describes the market reaction to bank acquisition announcements. The sample consists of 158 European bank acquisitions announced between 1998 and 2014. Abnormal returns were calculated using the MSCI EUROPE Banks index as the market index. Tests of statistical significance are based on robust standard errors.
Event window N Average CAR (%) % Positive % Negative t-Test
-2 to +2 days 158 -0.1981 50.64 49.36 -1.24
16
Table 3 shows the descriptive statistics for the variables used. The average percentage of women on board is in line with García-Meca, García-Sánchez, and Martínez-Ferrero (2015) and percentage of foreigners is slightly lower (0.10 and 0.18, respectively). Age diversity and tenure diversity measures are in line with Hagendorff and Keasey (2012) (0.126 and 0.783, respectively). The average bank in the sample has a ROE of 15.2 percent in the year prior to the deal and value of the assets of the target constitutes on average 12 percent of the value of the assets of the bidder. Average deal value is 2.03 billion euro; however, the standard deviation is quite large; the sample includes very small and very large deals. Almost 40 percent of the deals are domestic and over 80 percent of the banks completed more than one acquisition throughout the observed period. Almost half of the directors on an average board are independent and a quarter of the banks in my sample use the two-tier board system. The average board is relatively large; it is comprised of 16 directors, which is quite standard in the banking industry. Table A2 in the appendix presents the pair-wise correlations between the variables. Most of the correlation coefficients are low; the highest correlation is 0.48 between the relative size of the target and bidder and the domestic dummy. Therefore multicollinearity does not seem to be an issue for my data.
Table 3
Descriptive statistics.
N Mean Median Standard deviation
Minimum Maximum
Board diversity measures
Gender diversity 158 0.109 0.067 0.127 0 0.667 Nationality diversity 117 0.148 0.133 0.143 0 0.615 Age diversity 125 0.138 0.135 0.032 0.076 0.266 Tenure diversity 138 0.790 0.786 0.282 0 2.285
Deal and bank variables
ROE 158 0.152 0.149 0.087 -0.217 0.378 Relative size10 158 0.120 0.017 0.253 0.000 1.807 Deal value (th. €) 158 2029979 206605 5160685 412.460 38900000 Domestic 158 0.392 0 0.490 0 1 Repeat bidder 158 0.804 1 0.398 0 1 Governance variables Board independence 132 0.440 0.474 0.251 0 1 Board size 158 16.247 16 5.702 5 32 Two-tier board 158 0.241 0 0.429 0 1 10
17
5. Results
This section reports the results of the analysis described in the previous section. First of all, univariate results of a simple comparison of the most and least diverse boards are reported and subsequently the results of the previously discussed OLS regression are shown.
5.1.Univariate results
The deals were ranked and divided into four quartiles based on the market reaction around the M&A announcement. Table 4 presents the board diversity measures for the lowest and the highest quartile according to the 5-day CARs, and Table 5 according to the 21-day CARs. If board diversity is beneficial for the decision-making process and therefore seen as value-enhancing by the shareholders, we should see significantly higher levels of board diversity in the highest quartile (i.e. the banks that experienced highest CARs around the announcement) as compared to the lowest quartile (i.e. banks that experienced very low, or more specifically very negative CARs around the announcement). This is true and consistent for both event windows for gender diversity. Banks in the highest quartile ranked according to 5-day CARs (21-day CARs) had on average of 13.8 (12.3) percent of female directors as compared to 8.7 (7.1) percent for the banks in the lowest quartile; a significant difference at the 5 percent level.
Table 4
Board diversity and the 5-day CAR around the merger announcement.
This table presents the board diversity measures for the lowest and the highest quartile according to the 5-day CAR. The sample consists of 158 European bank acquisitions announced between 1998 and 2014. Tests of statistical significance are based on standardized prediction errors. 5-day CAR is the sum of the abnormal returns from 2 days before the announcement to 2 days after the announcement. Abnormal returns were calculated using the MSCI EUROPE Banks index as the market index. Gender diversity is defined as the number of female directors and nationality diversity as the number of foreign directors, both scaled by board size. Age and tenure diversity are calculated as standard deviation divided by mean across the board. t-Statistics are in the parentheses. Statistical significance at the 10% and 5% level is denoted by * and **, respectively.
5-day CAR (%)
18 Table 5
Board diversity and the 21-day CAR around the merger announcement.
This table presents the board diversity measures for the lowest and the highest quartile according to the 21-day CAR. The sample consists of 158 European bank acquisitions announced between 1998 and 2014. Tests of statistical significance are based on standardized prediction errors. 21-day CAR is the sum of the abnormal returns from 10 days before the announcement to 10 days after the announcement. Abnormal returns were calculated using the MSCI EUROPE Banks index as the market index. Gender diversity is defined as the number of female directors and nationality diversity as the number of foreign directors, both scaled by board size. Age and tenure diversity are calculated as standard deviation divided by mean across the board. t-Statistics are in the parentheses. Statistical significance at the 5% level is denoted by **.
21-day CAR (%)
Lowest quartile (Q1) Highest quartile (Q4) Difference Q1- Q4 Gender diversity 0.071 (n=39) 0.123 (n=40) -0.053** (-2.030) Nationality diversity 0.145 (n=31) 0.113 (n=27) 0.032 (0.773) Age diversity 0.141 (n=32) 0.149 (n=33) -0.008 (-0.870) Tenure diversity 0.728 (n=36) 0.890 (n=36) -0.162** (-2.121)
This is consistent with the hypothesis that gender diversity enhances the decision-making capabilities of the board and that therefore more gender diverse boards enjoy higher abnormal returns around the merger announcement. On the other hand, both nationality and age diversity do not seem to have an impact on the market reaction upon the deal announcement. There is no significant difference in the nationality and age diversity between the banks in the highest and the lowest quartile. Finally, it appears that shareholders value the combination of experience of longer-tenured directors with fresh perspective of newly appointed directors and believe that more tenure diverse boards make better decisions. The boards in the highest quartile are significantly more tenure diverse (at 5 or 10 percent significance level, depending on the event window) than the boards of the banks in the lowest quartile.
5.2. Multivariate results
19 Table 6
Results of the regression analysis of the impact of board diversity on the 5-day cumulative abnormal returns around the announcement.
This table shows the results of the hierarchical regression analysis, using the sample of 158 European bank acquisitions announced between 1998 and 2014. In columns 1-4 each board diversity measure is regressed on the 5-day CAR separately, including only control variables. Column 5 reports the results for when all board diversity measures are included simultaneously, and in column 6 board independence is added as a control variable. 5-day CAR is the sum of the abnormal returns from 2 days before the announcement to 2 days after the announcement. Abnormal returns were calculated using the MSCI EUROPE Banks index as the market index. Gender diversity is defined as the number of female directors and nationality diversity as the number of foreign directors, both scaled by board size. Age and tenure diversity are calculated as standard deviation divided by mean across the board. ROE is the return on average equity in the year prior the deal announcement, relative size is calculated as the value of the assets of the target divided by the value of the assets of the bidder and deal value is measured in thousand €. Domestic dummy is 1 when bidder and target are located in the same country and 0 otherwise. Repeated bidder dummy is 1 when the bidder completed more acquisitions within the observed period and 0 otherwise. Board independence is the number of independent directors (scaled by the board size) and board size is the total number of directors. Two-tier dummy is 1 when the bidder has the two-tier board structure and 0 otherwise. Relative size, deal value and board size are introduced into the model in terms of natural logarithms. Statistical significance at the 10%, 5% and 1% level is denoted by *, ** and ***, respectively. Robust standard errors are in the parentheses.
20 Table 7
Results of the regression analysis of the impact of board diversity on the 21-day cumulative abnormal returns around the announcement.
This table shows the results of the hierarchical regression analysis, using the sample of 158 European bank acquisitions announced between 1998 and 2014. In columns 1-4 each board diversity measure is regressed on the 21-day CAR separately, including only control variables. Column 5 reports the results for when all board diversity measures are included simultaneously, and in column 6 board independence is added as a control variable. 21-day CAR is the sum of the abnormal returns from 10 days before the announcement to 10 days after the announcement. Abnormal returns were calculated using the MSCI EUROPE Banks index as the market index. Gender diversity is defined as the number of female directors and nationality diversity as the number of foreign directors, both scaled by board size. Age and tenure diversity are calculated as standard deviation divided by mean across the board. ROE is the return on average equity in the year prior the deal announcement, relative size is calculated as the value of the assets of the target divided by the value of the assets of the bidder and deal value is measured in thousand €. Domestic dummy is 1 when bidder and target are located in the same country and 0 otherwise. Repeated bidder dummy is 1 when the bidder completed more acquisitions within the observed period and 0 otherwise. Board independence is the number of independent directors (scaled by the board size) and board size is the total number of directors. Two-tier dummy is 1 when the bidder has the two-tier board structure and 0 otherwise. Relative size, deal value and board size are introduced into the model in terms of natural logarithms. Statistical significance at the 10%, 5% and 1% level is denoted by *, ** and ***, respectively. Robust standard errors are in the parentheses.
21
For the 5-day event window virtually none of the coefficients are significant and the R-squared of the regressions are very low. This points into the direction of possible leakage of the information by insiders prior to the announcement. Similar effect is documented by Cybo-Ottone and Murgia (2000) and Beltratti and Paladino (2013). For this shorter event window only a significant negative relationship between the deal value and the market reaction to the announcement is found, which is consistent with the assumption that smaller deals are easier to integrate and therefore seen more favourably by the shareholders.
The coefficient for the gender diversity is statistically significant and positive in the longer event window, both when included separately and simultaneously with the other measures of diversity. This is in line with the results of the univariate analysis and suggests that shareholders view M&A decisions made by more gender diverse boards more favourably, therefore providing further support for hypothesis 1. This is also consistent with the findings of García-Meca, García-Sánchez, and Martínez-Ferrero (2015) that board gender diversity has a positive impact on the financial performance of European banks. In contrast, Hagendorff and Keasey (2012) do not find any significant relationship between gender diversity and CAR around announcement for a sample of American banks. Therefore, it seems that the effect of gender diversity on financial performance is indeed location-specific.
The coefficient for nationality diversity is not statistically significant, so there is no support for hypothesis 2. It seems that shareholders attach neither positive nor negative value to announcements made by more nationality diverse boards. This could indicate that the nationality composition is not an important factor to them, or that they are aware of both benefits connected with foreign directors in terms of heterogeneity of ideas and cross-border knowledge transfer, but also of possible costs in terms of coordination problems and missing local knowledge. This is partially in conflict with the findings of García-Meca, García-Sánchez, and Martínez-Ferrero (2015) who document a negative relationship between nationality diversity and the general financial performance of a bank.
22
In line with the results of the univariate analysis, the coefficient for tenure diversity is significant and positive, both when included separately and simultaneously with other measures of board diversity, providing further support for hypothesis 4. This means that deals announced by more tenure diverse boards enjoy higher cumulative abnormal returns around the merger announcement. Therefore, this provides some evidence that shareholders value the combination of expertise of longer-tenured directors and fresh perspective of newly-appointed investors and believe this diversity provides superior decision-making dynamics. This is in conflict with the findings of the US literature, claiming that in terms of market reaction to the M&A announcement shareholders prefer experienced over diverse teams (Hagendorff and Keasey, 2012). Therefore, this again confirms the claim of Hagendorff, Collins, and Keasey (2007), that same corporate governance arrangements may have different value implication in these two respective regions.
There is a significant negative relationship between the market reaction to the deal announcement and the repeat bidder dummy variable, i.e. the market is more sceptic to deals announced by experienced bidders than to ones announced by first-time bidders. This is inconsistent with the expectation of learning effects for more experienced bidders. However, Beitel, Schiereck, and Wahrenburg (2004) also find a negative relationship between bidder experience and cumulative abnormal returns around the announcement date for a sample of European bank M&As.11 Furthermore, there is a significant negative relationship between the market reaction at the announcement and the two-tier board dummy, indicating there could be differences in the decision-making dynamics between the one-tier and the two-tier board structures, with possible value implications for the shareholders. This also supports the notion, that findings for one-tier structures (i.e. all of American bank literature) cannot be easily generalized to two-tier structures, which are present in some European countries.
The values of the R-squared in Table 7 are relatively low, but in line with the bank merger literature and illustrating the fact that many determinants of value creation for bank mergers are not yet known (Hagendorff and Keasey, 2012).
6. Complementary analysis
The M&A literature generally uses two approaches to evaluate the success of a merger or an acquisition (DeYoung, Evanoff, and Molyneaux, 2009). The first is examining the market
11 In my case it seems it is really the dichotomous distinction between first-time bidders and experienced
23
reaction and wealth creation at the announcement via an event study. The second commonly used approach is to evaluate the changes in operating performance in the post-merger period using accounting ratios. While the first approach relies on market's perception of the deal and its estimated benefits, the second examines to what degree these benefits materialize in the post-merger period and translate into actual changes in financial performance. To support my main findings from the event study, I conducted asimple exploratory analysis using the second approach. This combination of approaches is also used by others (e.g. Campa and Hernando, 2006), to increase the robustness of the outcome. The literature identifies two main sources of financial gains generated by M&As in the financial sector: improvements in operating efficiency and increases in market power (DeYoung, Evanoff, and Molyneaux, 2009). As the efficiency hypothesis is seen as the main driver for the consolidation in European banking (Campa and Hernando, 2006), I will concentrate on this aspect in the following. The “acquire to restructure” hypothesis, i.e. that targets are typically less efficient banks that are acquired to be restructured and made more profitable is further proven by Hernando, Nieto, and Wall (2009), Pasiouras, Tanna, and Gaganis (2011) and Caiazza, Clare, and Pozzolo (2012) for both domestic and cross-border targets in European bank M&A.
The first part of this study has shown that certain types of board diversity (gender and tenure) are seen by shareholders as enhancing the decision-making capabilities of the board and rewarded by higher cumulative abnormal returns around the announcement. In this part it will be examined if the same effects are present, when the actual changes in financial performance are considered, i.e. do more diverse boards pick targets that will experience larger improvements in their financial performance as result of the merger? Merger-induced performance gains are most commonly tested by comparing pre- and post-merger levels of accounting ratios (DeYoung, Evanoff, and Molyneaux, 2009). As is common in the literature, I will concentrate on profitability and cost-efficiency gains.
24
improvements in profitability (and efficiency) materialize in general between two and three years after the acquisition.12
To measure cost-efficiency, I use the cost-to-income ratio of the target, in line with Campa and Hernando (2006), Lanine and Vander Vennet (2007) and Correa (2009). Similar to profitability gains, cost-efficiency gains are measured as the change in cost-to-income ratio between one year before the acquisition and three years after the acquisition. It should be noted, that higher to-income ratio implies lower efficiency. Therefore, the improvement in cost-efficiency will be described by a negative number and deterioration in cost-cost-efficiency by a positive number, as the change in cost-to-income ratio.
Data on ROA and cost-to-income ratio for the targets were retrieved from Bankscope. As missing data was an issue for many targets, the final sample for this part of the analysis are 70 targets.
Subsequently, changes in ROA and cost-to-income ratio were calculated and targets were ranked and divided into four quartiles based on them. As more diverse boards are expected to choose targets that can be improved relatively more, it is expected to find higher levels of diversity in the highest quartile for profitability gains and in the lowest quartile for efficiency gains (due to the fact that cost-efficiency gains correspond to negative number for change in cost-to-income ratio).
Table 8 shows the results for profitability gains. It documents that targets that improved their profitability the most after the acquisition were acquired by banks with boards with significantly higher gender diversity (16 percent of female directors on average) as compared to targets that improved their profitability the least (7.7 percent of female directors on average). Therefore there is some evidence that shareholders are right when they see deals announced by more gender diverse boards more favourably. However, these findings should be interpreted with caution due to relatively small sample size. For all other types of board diversity there is no significant difference between the best and worst performing targets, so these do not seem to impact the capabilities of selection of a suitable target. This is in line with the findings from the first part for nationality and age diversity, but in conflict for tenure diversity.
12
Alternative ways of measuring the change, such as change between one year before and two years after or the difference between the average of three preceding years and three subsequent years, were also
25 Table 8
Bidder board diversity and target profitability gains.
This table presents the bidder board diversity measures for the lowest and the highest quartile according to profitability gains of the target. The sample consists of 70 European bank acquisitions announced between 1998 and 2014. Tests of statistical significance are based on standardized prediction errors. Profitability gains of the target bank are measured as change in return on assets (ROA) between one year before the acquisition and three years after the acquisition. Gender diversity is defined as the number of female directors and nationality diversity as the number of foreign directors, both scaled by board size. Age and tenure diversity are calculated as standard deviation divided by mean across the board. t-Statistics are in the parentheses. Statistical significance at the 5% level is denoted by **.
Change in ROA
Lowest quartile (Q1) Highest quartile (Q4) Difference Q1- Q4 Gender diversity 0.077 (n=17) 0.160 (n=17) -0.083** (-2.332) Nationality diversity 0.152 (n=16) 0.157 (n=12) -0.005 (-0.094) Age diversity 0.129 (n=15) 0.120 (n=14) 0.009 (0.632) Tenure diversity 0.829 (n=17) 0.879 (n=14) -0.050(-0.575)
Table 9 shows the results for cost-efficiency gains. Here no significant differences in board diversity of the bidder between the best performing and worst performing targets can be observed. Therefore it seems that banks with higher board diversity were not able to pick targets that were able to improve their efficiency due to post-merger restructuring more.
Table 9
Bidder board diversity and target cost-efficiency gains.
This table presents the bidder board diversity measures for the lowest and the highest quartile according to cost-efficiency gains of the target. The sample consists of 70 European bank acquisitions announced between 1998 and 2014. Tests of statistical significance are based on standardized prediction errors. Cost-efficiency gains of the target bank are measured as change in cost-to-income between one year before the acquisition and three years after the acquisition. Gender diversity is defined as the number of female directors and nationality diversity as the number of foreign directors, both scaled by board size. Age and tenure diversity are calculated as standard deviation divided by mean across the board. t-Statistics are in the parentheses.
Change in cost-to-income ratio
Lowest quartile (Q1) Highest quartile (Q4) Difference Q1- Q4 Gender diversity 0.105 (n=16) 0.071 (n=16) 0.034 (0.910) Nationality diversity 0.165 (n=15) 0.107 (n=12) 0.058 (0.870) Age diversity 0.117 (n=10) 0.138 (n=12) -0.021 (-1.202) Tenure diversity 0.814 (n=13) 0.813 (n=14) 0.001 (0.016)
26
7. Conclusion
This study used a unique hand-collected data set on 158 European bank-to-bank M&A deals, completed between 1998 and 2014, to illustrate the effects of board diversity with regard to a specific corporate strategy. It shows a significant positive relationship between gender and tenure diversity on the board of the acquiring bank and the cumulative abnormal returns surrounding the deal announcement, when a 21 day event window is assumed. This provides some evidence that shareholders see gender and tenure diversity as enhancing for the board decision-capabilities and therefore value-creating with regard to the M&A decision. Additional analysis provided some evidence supporting the correctness of this assumption of shareholders for gender diversity, by showing that target banks acquired by banks with more gender diverse boards were able to improve their profitability relatively more as a result of the post-merger restructuring. In contrast, the market for corporate control does not seem to be attaching any additional value to announcement made by relatively more nationality and age diverse boards. Overall, there is evidence that effects of board diversity on financial performance are location-specific and differ between the US and Europe.
These findings suggest that gender (and tenure) diversity improves the quality of decision-making on European bank boards and could be therefore used as one of the ways to improve the monitoring mechanisms. In the present study, I illustrate the benefits of diversity on M&A decisions, but for banks there are more situations in which the boards need to assess risks and make complex decisions, in which board diversity might prove to be value-adding. Given the results of this study, there is another piece of empirical evidence in support of the calls for higher gender diversity by many officials, discussed in the introduction of this study, and justification for the gender quotas set by the European Central Bank.
27
While the present study showed that gender and tenure diversity are value-creating for the shareholders of the acquiring bank, further research should try to examine and describe the processes behind this relationship. Agency and resource dependence theory provide many reasons why diversity should be value-enhancing, but to date there is little empirical research on how exactly diversity creates value and what the differences are in functioning between diverse and homogeneous boards. Additionally, there could be some merit in also considering the board diversity measures of the targets and adding them as control variables to the OLS regression. Low diversity on the board of the target and the hope for improvement in corporate governance as a result of the merger might be one of the driving forces behind the higher cumulative abnormal return for acquirers with more diverse boards. Furthermore, more thorough analysis using panel data and including control variables should be employed to explore the effect board diversity of the acquirer has on the post-merger improvement in the operating performance of the target. Finally, as the findings of my research suggest that the effects of board diversity are indeed location-specific and differ between US and Europe, further research could validate and more thoroughly assess this assumption, and possibly explore the effect in countries beyond these two regions.
References
Adams, R., Ferreira, D., 2009. Women in the boardroom and their impact on governance and performance. Journal of Financial Economics 94, 291-309.
Agnew, J., Balduzzi, P., Sunden, A., 2003. Portfolio choice and trading in a large 401(K) plan. American Economic Review 93, 193-215.
Altunbaş, Y., Marqués, D., 2008. Mergers and acquisitions and bank performance in Europe: the role of strategic similarities. Journal of Economics and Business 60, 204-222.
Barber, B., Odean, T., 2001. Boys will be boys: gender, overconfidence, and common stock investment. Quarterly Journal of Economics 116, 261-292.
Beck, T., Behr, P., Guettler, A., 2013. Gender and banking: are women better loan officers?. Review of Finance 17, 1279-1321.
Beitel, P., Schiereck, D., Wahrenburg, M., 2004. Explaining M&A success in European banks. European Financial Management 10, 109-139.
Beltratti, A., Paladino, G., 2013. Is M&A different during a crisis? Evidence from the European banking sector. Journal of Banking & Finance 37, 5394-5405.
28
Businessweek. 2013. Europe’s central bank goes for gender quotas. Businessweek, August 29, http://www.businessweek.com/articles/2013-08-29/europes-central-bank-goes-for-gender-quotas.
Byrd, J., Hickman, K., 1992. Do outside directors monitor managers? Evidence from tender offer bids. Journal of Financial Economics 32, 195-221.
Caiazza, S., Clare, A., Pozzolo, A., 2012. What do bank acquirers want? Evidence from worldwide bank M&A targets. Journal of Banking & Finance 36, 2641-2659.
Campa, J., Hernando, I., 2006. M&As performance in the European financial industry. Journal of Banking & Finance 30, 3367-3392.
Campbell, K., Mínguez-Vera, A., 2008. Gender diversity in the boardroom and firm financial performance. Journal of Business Ethics 83, 435-451.
Campbell, K., Mínguez-Vera, A., 2010. Female board appointments and firm valuation: short and long-term effects. Journal of Management & Governance 14, 37-59.
Carter, D., Simkins, B., Simpson, G., 2003. Corporate governance, board diversity, and firm value. Financial Review 38, 33-53.
Chen, G., Crossland, C., Huang, S., 2014. Female board representation and corporate acquisition intensity. Strategic Management Journal, forthcoming.
Cornett, M., Hovakimian, G., Palia, D., Tehranian, H., 2003. The impact of the manager-shareholder conflict on acquiring bank returns. Journal of Banking & Finance 27, 103-131.
Correa, R., 2009. Cross-border bank acquisitions: is there a performance effect?. Journal of Financial Services Research 36,169-197.
Cybo-Ottone, A., Murgia, M., 2000. Mergers and shareholder wealth in European banking. Journal of Banking & Finance 24, 831-859.
DeLong, G., 2003. Does long-term performance of mergers match market expectations? Evidence from the US banking industry. Financial Management 32, 5-25.
DeLong, G., DeYoung, R., 2007. Learning by observing: information spillovers in the execution and valuation of commercial bank M&As. The Journal of Finance 62, 181-216.
DeYoung, R., Evanoff, D., Molyneux, P., 2009. Mergers and acquisitions of financial
institutions: a review of the post-2000 literature. Journal of Financial Services Research 36, 87-110.
Díaz, B., Olalla, M., Azofra, S., 2004. Bank acquisitions and performance: evidence from a panel of European credit entities. Journal of Economics and Business 56, 377-404.
Ely, R., Thomas, D., 2001. Cultural diversity at work: the effects of diversity perspectives on work group processes and outcomes. Administrative Science Quarterly 46, 229-273.
Erhardt, N., Werbel, J., Shrader, C., 2003. Board of director diversity and firm financial performance. Corporate Governance: An International Review 11, 102-111.
Fama, E., 1980. Agency problems and the theory of the firm. The Journal of Political Economy 88, 288-307.
29
Forbes, D., Milliken, F., 1999. Cognition and corporate governance: understanding boards of directors as strategic decision-making groups. The Academy of Management Review 24, 489-505.
García-Meca, E., García-Sánchez, I., Martínez-Ferrero, J., 2015. Board diversity and its effects on bank performance: an international analysis. Journal of Banking & Finance 53, 202-214. Hagendorff, J., Collins, M., Keasey, K., 2007. Bank governance and acquisition performance. Corporate Governance: An International Review 15, 957-968.
Hagendorff, J., Keasey, K., 2009. Post‐merger strategy and performance: evidence from the US and European banking industries. Accounting & Finance 49, 725-751.
Hagendorff, J., Keasey, K., 2012. The value of board diversity in banking: evidence from the market for corporate control. The European Journal of Finance 18, 41-58.
Havrylchyk, O., Jurzyk, E., 2011. Inherited or earned? Performance of foreign banks in Central and Eastern Europe. Journal of Banking & Finance 35, 1291-1302.
Hermalin, B., Weisbach, M., 2003. Boards of directors as an endogenously determined institution: a survey of the economic literature. Economic Policy Review 9, 7-26.
Hernando, I., Nieto, M., Wall, L., 2009. Determinants of domestic and cross-border bank acquisitions in the European Union. Journal of Banking & Finance 33, 1022-1032. Hillman, A., Cannella, A., Paetzold, R., 2000. The resource dependence role of corporate directors: strategic adaptation of board composition in response to environmental change. Journal of Management Studies 37, 235-256.
Houston, J., Ryngaert, M., 1994. The overall gains from large bank mergers. Journal of Banking & Finance 18, 1155-1176.
Huang, J., Kisgen, D., 2013. Gender and corporate finance: are male executives overconfident relative to female executives?. Journal of Financial Economics 108, 822-839.
Jensen, M., Meckling, W., 1976. Theory of the firm: managerial behavior, agency costs and ownership structure. Journal of Financial Economics 3, 305-360.
Kang, E., Ding, D., Charoenwong, C., 2010. Investor reaction to women directors. Journal of Business Research 63, 888-894.
Lanine, G., Vander Vennet, R., 2007. Microeconomic determinants of acquisitions of Eastern European banks by Western European banks. Economics of Transition 15, 285-308.
Levi, M., Li, K., Zhang, F., 2014. Director gender and mergers and acquisitions. Journal of Corporate Finance 28, 185-200.
Masulis, R., Wang, C., Xie, F., 2012. Globalizing the boardroom – the effects of foreign directors on corporate governance and firm performance. Journal of Accounting and Economics 53, 527-554.
Mateos de Cabo, R., Gimeno, R., Nieto, M., 2012. Gender diversity on European banks’ boards of directors. Journal of Business Ethics 109, 145-162.
Maznevski, M., 1994. Understanding our differences: performance in decision-making groups with diverse members. Human Relations 47, 531-552.
