Master’s Thesis 2018-2019 semester 2 M.Sc. International Business and Management
The impact of international assignment experience of the CEO on the performance of mergers and acquisitions and the moderation effect of gender
and TMT international experience in this relation.
University of Groningen, The Netherlands Faculty of Economics and Business
Nienke van Beek
h.g.van.beek@student.rug.nl Student number: 2733838
Date: 2019, June 3 Supervisor: dr. L. Ge Co-assessor: dr. S.R. Gubbi
Abstract 3
Introduction 4
Literature Review 7
Hypotheses 9
Hypothesis 1 10
Hypothesis 2 11
Hypothesis 3 12
Methods 13
Sample 13
Dependent variable 14
Independent variable 15
Moderators 15
Gender 15
Top management team international experience 15
Control variables 16
Analysis 17
Results 17
Descriptive statistics 18
Correlation analysis 18
Regression analysis 21
Robustness check 25
Discussion and conclusion 29
Managerial implications 30
Limitations and further research 31
Reference list 33
2
Abstract
The performance of merger and acquisitions is of significant importance for companies that strive for internationalization. Furthermore, it can be argued that the international experience of the CEO plays a significant role in this. Therefore, this thesis is about the influence of the international experience of the CEO on the performance of overseas merger and acquisitions.
Since the strategic decision making concerning merger and acquisitions, the upper echelons theory is of importance. Furthermore, based on the resource based view, CEOs are important resources for a company. In this thesis, their international experience is suggested as such a unique company resource for diversification. Moreover, the gender of the CEO and the international experience of the TMT have been studied as moderators. The statistical analysis of this study concludes that the international experience of the CEO does have a significant effect on the performance of a merger and acquisition, but this effect is negative. Further, the international experience of the TMT does not have a significant moderation effect, while the gender of the CEO has a significant moderation effect. The study is based on data from 2012-2017, containing 230 cross border M&A deals.
3
Introduction
Mergers and acquisitions are very important for companies. As mentioned by Custódio and Metzger (2013), merger and acquisitions (M&As) represent the biggest investments firms make.
This is also represented in the figures. As indicated by Grocer (2018), there was a record $2.5 trillion in mergers announced. This suggests the importance of M&As. However, this number assumably is not representing all the M&As firms were investing in. The performance of M&As is not always as expected. This is shown by the fact that the failure rate of M&A is high.
According to Christensen, Alton, Rising and Waldack (2011), studies have put the failure rate of M&As between 70% and 90%. As the literature on M&As show, there are different reasons why they would fail (Weber & Camerer, 2003; Cartwright & Schoenberg, 2006; Deng, 2010;
Christensen et al, 2011). Since the investments in M&A concern a lot of strategic risk, it is important to study how this decision-making process could be improved and how the additional attributes play a big role.
However, there is a difference between the characteristics of domestic M&As and cross-border M&As. Since M&As concern a lot of risk and have a high failure rate, research on the outcomes of cross-border M&As may be meaningful to the literature and management practices (Shimizu, Hitt, Vaidyanath & Pisano, 2004). That is where the focus on the international experience of the CEO comes in. As cross-border M&As are associated with more uncertainty and complexity than domestic M&As, knowledge, and experience about the context of an international M&A can be helpful (Shimizu et al., 2004). Furthermore, as noticed by Walters, Kroll and Wright (2006) prior experience of the CEO should improve the decision process. Moreover, Custódio and Metzger (2013) found that CEOs with experience in the target industry gained more returns than CEOs without that experience. However, how is this for the international experience of CEOs? There is little knowledge about how CEO’s international experience create value for the firms (Custódio and Metzger, 2013). Carpenter, Sanders & Gregersen (2001) state that executive international assignment experience is rare, valuable and hard to imitate and that it can ensure competitive advantages in some contexts. Moreover, managing overseas has been one of the most prominent challenges for U.S. businesses (Magnusson & Boggs, 2006). Does a CEO with
4
more international assignment experience also gain better overseas M&a performance returns than CEOs without this experience? How the international experience of the CEO can influence the merger and acquisitions field is not really clear. International assignment experience can ensure executives with the development of tools that are necessary to benefit from the global uncertainty in exchange rates, government policies, and competitors actions. Without this experience, a barrier to understanding the differences is hard to overcome (Carpenter, Sanders &
Gregersen, 2001). International experience is an important experience that affects the capability of the CEO. Therefore, it is important how this affects overseas M&A performance.
Therefore, to get answers to the questions mentioned this study will focus on the effect of the international experience of the CEO on the merger and acquisition performance of the MNE.
This results in the following research question that is central in this study:
How does the international experience of the CEO influence the merger and acquisition performance of the company and if so, under what conditions?
Firms that have prior experience from merger and acquisitions are likely to have subsequent overseas mergers and acquisitions (Collins, Holcomb, Certo, Hitt and Lester, 2009).
The prior international experience from a CEO could be useful in the decision making of an overseas M&A since the upper echelons theory suggest that CEOs make strategic decisions based on their experiences, values and their personality (Hambrick, 2007). Since it is suggested that prior international experienced of the CEO could have certain effects on the performance of M&As, the CEO could be a competitive advantage according to the resource based view.
There has also been a lot of research on the effect of gender on the behavior of manager (Levi, Li, and Zhang, 2014; Huang, Kisgen, 2013). However, how does this relate to the M&A performance through international assignment experience? Does a firm have a disadvantage when employing a certain gender? There are differences between male and female directors, which could result in differences in the M&A performance outcome. The main argument is that females are less risk taking then man (Byrnes, Miller & Schafer,1999). Therefore, women are expected to be less risk-taking, which could result in lower M&A performance.
5
As mentioned by Jaw and Lin (2009) the CEO might have different influences than the TMT.
They state that the job context, characteristics, and background are different from the executives in the TMT. Therefore, it is also useful to get insights about the international experience of the TMT and how this affects the performance of mergers and acquisitions. In this study, it is hypothesized that the international experience of the TMT could further enhance the relationship between the CEO international experience and the M&A performance since the TMT could support the CEO in the decision making process concerning M&As.
Wang, Holmes, Oh, and Zhu (2016) study the relationship between CEO characteristics, such as CEO experience, and firm outcomes. They state that it is not really clear how CEO characteristics have an influence on the firm outcomes, i.e. prior career experiences.
Furthermore, Ali, Kulik, and Metz (2011) note that there are inconsistent findings of research concerning gender. Therefore, this study contributes to this literature, in which it becomes evident how gender influences the relationship between CEO international experiences and M&A performance. Furthermore, Nielsen and Nielsen (2011) state that it is valuable to study different aspects of managerial backgrounds in international strategic decision-making.
Therefore, also the TMT is studied and the effect contributes to the literature about managerial backgrounds.
In the following section, I will review literature from previous studies and then develop hypotheses that will be summarized in a conceptual model. Afterward, I will discuss the methods section in which the sample and the different variables will be discussed, followed by the results.
In the end, there will be a discussion, the conclusion and the managerial implications from this study.
6
Literature Review
For the definition of merger and acquisitions, I make use of the definition provided by Buckley and Ghauri (2002). A merger and acquisition is when the ‘assets of two previously separate firms are combined to establish a new legal entity’ and the ‘control of assets is transferred from one company to another’.
A firm may engage in an M&A to explore a new opportunity or avoid a possible future threat.
Furthermore, an M&A may create opportunities for firms by learning new knowledge and the acquiring of new capabilities (Shimizu et al., 2004). The combination of the firms will ensure strategic goals more quickly and less expensive than without this M&A (Marks & Mirvis, 2011).
However, the performance of M&As is not always as expected. Marks and Mirvis (2011) state that most of the M&As result in failures in terms of financial performance and in negative consequences for the people and companies involved. Schraeder and Self (2003) look into different categories of challenges of M&As. The challenges after the completion of M&As mentioned in their paper are integration challenges, task challenges, demographic challenges, political challenges, and cultural challenges.
It is noticed that firms that have experience and knowledge from merger and acquisitions, both domestic and international, probably will also have subsequent international mergers and acquisitions (Collins, et al., 2009). They state that when completing an M&A, a firm is looking for new opportunities. They broadened their knowledge and networks in the international field.
When the subsequent deal is coming, the CEO will have knowledge and experience from the last time.
Organizational capabilities and organizational learning have been studied for many years (Huber, 1991; Anderson & Skinner, 1999). Since there are several processes when engaging in an M&A, all pre- and post-acquisitions processes are dynamic and involve learning. To smoothen these processes, prior experience can be useful (Shimizu et al., 2004), especially the absorptive capacity. The absorptive capacity as states by Zahra and George (2002) is ‘a dynamic capability pertaining to knowledge creation and utilization that enhances a firm’s ability to gain and sustain
7
a competitive advantage’. As an example, Nadolska and Barkema (2014) state that the experiences gained have an influence on the success of acquisitions.
However, Walters, Kroll, and Wrights (2007) state that key parties in an acquisition are the CEO and the directors. Therefore, the upper echelons theory is of importance (Magnusson & Boggs, 2006). The upper echelons theory has been a theory in which top executives have an important role in the big organizational outcomes (Carpenter, Geletkanycz & Sanders, 2004). As stated by Hambrick (2007), the central premise in the upper echelons theory is that CEO makes strategic decisions based on their personal interpretation and their personal interpretations are based on the experiences, values, and personalities of the CEO. Hambrick and Mason (1984) state that upper echelon characteristics are determinants for strategic decisions and have an effect on the performance of the organization. They further notice that a functional track may not have a major influence in strategic decision making of executives, but it is expected to have influence. As M&A is part of the largest investments a firm make with prominent strategic consequences, this is of significant importance. Moreover, as stated by Nadkarni and Barr (2008), top managers make strategic decisions based on their developed subjective representations of the environment.
Managers have the opportunity to create and select activities in which they have greater possibilities to have an effect on organizational outcomes (Finkelstein & Peteraf, 2016). When a CEO has experience from certain previous M&As, the CEO can determine better his opportunities and the choice of M&As. Besides, Carpenter, Sanders, and Gregersen (2001) state that it is supported by the upper echelons theory that international experience is valuable to obtain.
Moreover, a CEO may be seen as a unique organizational resource (Daily, Certo, & Dalton, 2000). This would place a CEO with international experience as a competitive advantage in the resource based view. As defined by Wernerfelt (1984) a resource is anything that is a strength or a weakness to a certain firm. Furthermore, Barney (1991) states that a competitive advantage involves a value creating strategy a company has that is at the same time not used by any current or potential competitor. However, in order to keep this advantage, it should be a sustained
8
competitive advantage. Barney (1991) state that in order for a competitive advantage to be sustained, a competitive advantage cannot be duplicated by current or potential competitors. As stated by Wernerfelt (1984) M&As are opportunities to trade resources that are not available at the market and to buy and sell resources in bundles. This is important since this study looks at the effect of the CEO’s international assignment experience as a resource for a certain organization on the M&A performance. As argued by Wright, McMahan, and McWilliams (1994) human resources are human capital that is in control of the firm through an employment relationship. They state that human resources always have the potential to lead to a sustained competitive advantage. Besides, there is much literature relating the characteristics of the CEO and the effect of these characteristics on the firm performance and firm outcomes.
Hypotheses
CEO international experience
Buckley and Ghauri (2002) state that firms can internationalize through different modes, including overseas M&As.
First, the CEO international experience could be positively related to the M&A performance, based on the upper echelons theory. Organizational outcomes can be seen as ‘reflections of the values and cognitive bases of powerful actors in the organization’ (Hambrick & Mason, 1984).
Carpenter, Geletkanycz and Sanders (2004) state that these cognitive bases are not observable.
However, the observable managerial characteristics are efficient proxies that provide proper measurements of the unobservable cognitive bases. Hambrick and Mason (1984) argue that the functional tracks and other career experiences are, among others, observable proxies for the unobservable cognitive constructs. As stated by Datta and Guthrie (1994), individual CEOs bring functional background experience with them to the firm. This functional background experience entails knowledge, attitudes, and skills that are partly the result of earlier similar functions.
Second, the CEO international experience could function as a competitive advantage to positively influence the M&A performance, based on the resource based view.
9
Also, Kolev and Haleblian (2018) state that firms learn from prior acquisitions. Important in this experience is the absorptive capacity. Cohen and Levinthal (1990) argue that the capability to evaluate and use outside knowledge is mostly the result of the level of prior related knowledge.
This could also be related to the prior international experience of the CEO. This international experience of a CEO could function as a competitive advantage because then the CEO is able to evaluate and use outside knowledge in the decision making concerning outside mergers and acquisitions. Moreover, as argued by Finkelstein, Hambrick, and Cannella (2009), international experience can help to create opportunities that result in improved performance. This is especially the case considering strategic context, in which it is an advantage to have this international experience. Especially since it could be used as a competitive advantage for the company. As noticed by Agrawal and Sensarma (2007), the reasons for firms to invest in M&As can be seen in the actual decision.
As suggested argued above, the CEO international experience has an influence on the strategic decision making process and could bring sustainable resources to the firm. Therefore, I predict that the international experience of the CEO will have a positive influence on the performance of the M&A, which a firm engage in.
Hypothesis 1
The greater the international experience of the CEO, the better the M&A performance.
Gender
As noticed by Huang and Kisgen (2012), there has been an increase in the number of females in top positions in U.S. firms. Several studies have looked into the position of female directors in top positions (Huse and Solberg, 2006; Bilimoria, 2000; Brennan and McCafferty, 1997). The characteristics of female leaders provide insight as to how the decision-making process may differ in their presence (Parola, Ellis and Golden, 2015). They state that female executives are more innovative, proactive, transformational, but also more cautious and risk averse than male executives. Parola et al. (2015) further argue that based on the upper echelons theory, the difference between the different genders of top managers results in possible differences in the
10
opinions, values, goals, and attitudes. As mentioned before, Hambrick and Mason (1984) argue that, based on the upper echelons theory, traits and experiences of the top manager will reflect in the organizational outcomes. Since gender is a characteristic in addition to the functional background and experiences of the CEO, this will have an outcome on the organizational outcomes. In general, males are much more likely to engage in risk taking than females (Byrnes, Miller & Schafer, 1999). Moreover, Charness and Gneezy (2012) state that the likelihood of women investing in risky assets is less likely than for men and this indicates that women are more risk averse than men. Furthermore, as concluded by Faccio, Merchica, and Mura (2016), female CEOs are less efficient in allocating capital as their male equivalent. As discussed earlier, M&As concern a lot of risks and often result in bad performance.
Ali, Kulik, and Metz (2011) argue, based on the resource based view, that human resources of a company are an essential type of competitive advantage because other resources can be easily be imitated by competitors. However, the resources a manager brings to the firm are different for females and males. Huang and Kisgen (2012) conclude that female executives are less likely to make acquisitions than their male colleagues. This is similar to the study by Levi, Li, and Zhang (2013), who also found that women are less likely to engage in mergers and acquisitions. This would indicate that women would take less risk in investments and thus also in overseas mergers and acquisitions since they are associated with a lot of strategic risks.
As the literature suggests that women tend to be less risk averse than men, I propose that the when the CEO is female, this would weaken the effect of the relation between the international experience of the CEO and the M&A frequencies. This results in the following hypothesis:
Hypothesis 2
When the CEO is female, this has a weakening effect on the influence of the international assignment experience of the CEO on the performance of overseas M&As.
TMT international experience
Based on the upper echelons theory, the TMT international experience matters when studying how it moderates the relationship between the CEO international experience and the M&A
11
performance. Hambrick, Humphrey, and Gupta (2015) argue that essential base of the upper echelons theory is that executives ‘perceive situations and alternatives through individualized lenses shaped by their personal attributes’, among others their professional experiences. They further state that TMTs make firm strategic decisions collectively. As argued by Jaw and Lin (2009) the TMT is different as the CEO. They state that the CEO role is a unique role and different from TMT roles. The international experience of top managers has been studied before (Reuber and Fischer, 2002; Nadolska and Barkema, 2014). Based on Lin and Rababah (2014), a high quality relationship between the TMT and CEO could be helpful in pursuing organizational goals. Jaw and Lin (2009) note that while the CEO in an important company strategy determinant in a turbulent environment, the TMT is the other essential determinant.
Moreover, based on the resource based view, the TMT would be a unique resource to the firm.
As argued by Field and Mkrtchyan (2017), board with a high level of acquisitions experience are better in making acquisition decisions. They further notice that firms should recruit directors based on the experience of acquisitions and the performance of those acquisitions. This experience would result in human resources for the firm, which is stated to be likely to result in a competitive advantage (Wright, McMahan and McWilliams, 1994).
Therefore, it is useful how the international experience of the TMT influences this relationship. I propose that the international experience of the TMT would further strengthen the relationship between the CEO international experience and M&A performance. This results in the following hypothesis:
Hypothesis 3
The diversity of the international experience of the top management team has a strengthening effect on the relationship of the CEO international experience on the M&A performance.
12
All the hypotheses can be summarized in the following conceptual model:
Methods
The data collection part of this study has been performed in collaboration with Lynn Ekeya, also MSc student International Business and Management.
Sample
The sample will consist of CEO centered in North America. This is consistent with the BoardEx database, from which the CEO profile data is obtained. Therefore, the date concerning the CEO and TMT will be collected from BoardEx.
The focus will be on companies that are in the Fortune 500. This ranking list is based on multinationals that are centered in the United States and which are ranked according to the size of their sales. The Fortune 500 ranking list is also used in earlier research concerning CEO international experience (Collins, et al, 2009; Daily, Certo, & Dalton, 2000; Hamori & Koyuncu, 2015).
In Zephyr the search strategy for the merger and acquisition deals had the following conditions:
the data had to involve cross border merger and acquisitions from S&P 500 companies within the time period 01/01/2012 till 31/12/2017. We only looked at acquirer M&A deals and all the deals that were completed. Furthermore, this study focuses on the manufacturing industry. To filter for the industry, the US SIC codes have been used. The manufacturing industry US SIC codes are
13
20-39. Focusing on this single industry ensured the deals were all of the same nature. After the collection of this data, all non-US acquirer companies were excluded.
The ISIN company codes that were left in this sample were matched with the BoardEx database.
This provided a list of directors that were employed by those companies. This resulted in 681 directors in our sample. For those directors, we collected first the work experience of them. This resulted in an excel sheet sorted by the director ID with all of their work experience until 31/12/2017. We then created a separate excel sheet with the director IDs and the directors’
names in which we manually counted the number of cross border countries they had work experience. This resulted in a number which displayed in how many cross border countries they worked. When the director did have work experience in an ‘Unknown’ country, also among other countries, the director was excluded from the sample. After this, the gender and the age of the director were collected by the Director ID from the BoardEx database and matched with this data. After completing this data, the directors of the companies were matched with the deal number and deal information. The directors that were employed between the announcement and completion date of the deal were then matched with the unique number per deal. Since the composition of the TMT was different for every deal, due to different periods in the time period, the directors employed were manually selected per deal. This left us with an overview of directors employed at the time of the deal. This was then matched with the company data concerning the deal.
After this, the TMT size (minus the CEO) has been calculated manually. Also, the diversity of the TMT has been calculated.
Furthermore, deals with missing or incomplete data were also excluded. The remaining sample contains 230 deals. These deals consist of companies in the S&P 500. Out of the 230 deals, there are deals of 91 S&P companies in our sample.
Dependent variable
The variable merger and acquisitions performance will be focused on the merger and acquisitions. This data will be collected through the Zephyr database. This database concerns all
14
company information from the most important companies worldwide. This variable will be measured through Tobin’s Q. The Tobin’s Q has earlier been used by Huang, Zhu, and Brass (2016) for measuring the performance of the merger and acquisition. The idea behind this Tobin’s Q is that it represents the ratio of market value to replacement cost (Lindenberg & Ross, 1981). Moreover, the performance of a merger and acquisition will not be displayed right after the completion of the merger and acquisition. Therefore, Tobin's Q at the end of the year after the year the merger and acquisition has been used. The formula used for calculating the Tobin’s Q is the following:
T otal assets
T otal assets + (common shares outstanding price close annual) − common ordinary equity*
Independent variable
The data regarding the international experience of the CEO will be collected through the BoardEx dataset. This database is previously used in research regarding CEO experience (Custódio and Metzger, 2013). The international experience of the CEO is measured as a continuous variable. The number it has stands for the number of cross border country a director or CEO has work experience in.
Moderators
Gender
Data regarding the influence of gender will also be collected by BoardEx. A dummy has been used for the gender of the CEO. A female CEO has the value ‘1’, whereas a male CEO has the value ‘0’.
Top management team international experience
Furthermore, the international experience of the TMT is proxied as the international experience diversity and is measured through the coefficient of variation. As noted by Abdi (2010), the
15
coefficient of variation tells the variance of a set of numbers independently of the unit of measurement used for these numbers.
Control variables
Jaw and Lin (2009) found that CEO position tenure had an important influence on the firm’s internationalization. As stated by Jaw and Lin (2009), a more tenured CEO may have more expertise in the foreign market and more knowledge in which direction the market or industry is going, to a better strategic decision making. However, it is also mentioned that for a long-tenured CEO, it is not attractive to internationalize, since the high risks involved. Also, CEO tenure is measured as the number of years the CEO has been CEO.
Furthermore, the age of the CEO is also a control variable. Yim (2013) states that CEO characteristics may change with age. They argue that the psychological and physiological differences between different ages have a direct effect on the CEOs tendency for acquisitions.
This control variable is measured as the age of the CEO.
Firm size is a common control variable regarding merger and acquisitions literature. Rienda, Claver, and Quer (2012) used firm size as a control variable and they state that the transaction cost theory predicts a relation between the firm size and the establishment mode, most of the times through acquisitions. Firm size is often associated with greater financial capacity (King, Dalton, Daily, Covin, 2004). As suggested by Nadolska and Barkema (2007), firm size should be used as a control variable. The size of the company is measured through the number of employees a company has.
As used in earlier research, another control variable will be firm age. Since it is assumed that older firms may have advantages due to earlier experiences that give them opportunities to grow internationally (Autio, Sapienza & Almeida, 2000). For this control variable, a continuous variable that represents the number of years a company exists has been used.
In accordance with Athanassiou and Nigh (2002), who studied the role of the TMT international experience on the firm’s internationalization, the role tenure of the TMT is a control variable.
According to Jaw and Lin (2009), the TMT tenure may have an influence on the firm’s internationalization. Keck (1997) states that shorter tenured TMT is more productive in
16
turbulent environments since they provide problem solving skills and fresh perspectives on strategic issues and how to implement it, while longer-tenured teams are more productive in stable environments since they prefer to uphold basic team functions. Since the decision making concerning M&A is strategic oriented, the tenure of the TMT will influence this. This variable has been valued in days for accurateness. A coefficient of variation has been used to measure the tenure diversity of the TMT.
Furthermore, TMT size has been used as a control variable. Jaw and Lin (2009) argue that TMT size is an important team characteristic. The TMT size is measured as the number of TMT members, excluding the CEO.
Analysis
It is the aim to analyze how the CEO international ( CEO IE) influences the performance of the M&A. Furthermore, the moderation effects of the CEO gender and the TMT international experience (TMT IE) are analyzed. In the analysis, the control variables will be included. With the use of SPSS, a linear regression analysis is performed to test the hypotheses. Since there was no concern of multicollinearity, the variables are not mean centered. This states that the dependent variable will predict the independent variable.
This results in the use of the following model:
β ...
Y = 0+ β1 * x1+ β2* x2+ + βp* xp + ε
Results
This section informs the results of the analysis of our sample. First of all, the descriptive statistics have been performed to show the number, the minimum and maximum value, and the mean with the standard deviation. This is followed by the correlation matrix. This matrix will show a value indication of the correlation between the different variables. At last, the regression models are shown, to indicate whether the results of the regression analysis are significant.
17
Descriptive statistics
The CEOs in this sample consisted of 200 (86,2%) men and 32 (13,8%) women, with an average age of 59,25 years at the time of the deal, with a standard deviation of 6,807. The mean of the international experience of the CEO is 1,31 years with a standard deviation of 1,648 years. The mean of the international experience of the TMT is 1,323 years with a standard deviation of 0,519 years.
N Minimum Maximum Mean Std. Deviation
Tobin's Q 2320 0,678 9,656 2,263 1,120
CEO
international
experience 230 0 8 1,32 1,651
CEO gender 230 0 1 0,140 0,347
TMT
international
experience 230 0 3 1,319 0,519
CEO tenure
(days) 230 184 6849,000 2316,91 1269,288
TMT tenure 230 0,000 1,164 0,530 0,195
Firm size 230 1,143 240,000 46,300 40,029
Firm age 230 2 137 55,48 37,721
TMT size 230 0 14 8,96 2,529
CEO age at
year of deal 230 41 77 59,13 6,713
Figure 2 - Descriptive statistics
Correlation analysis
A correlation analysis has been performed to indicate the correlation between the different variables used in the regression analysis. Mukaka (2012) states that correlation analysis is performed to measure the strength and direction of a linear relationship between two variables.
To examine the correlation, the Pearson correlation coefficient has been used.
18
Variable (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (1) Tobin's Q 1,000
(2) CEO international experience
-0,116 1,000
(3) CEO gender
-0,056 -0,132* 1,000
(4) TMT international experience
-0,079 0,026 -0,082 1,000
(5) CEO tenure
0,199** -0,062 -0,112 -0,058 1,000
(6)TMT tenure
0,057 0,085 0,089 -0,005 -0,138* 1,000
(7) Firm size -0,109 -0,186** -0,102 -0,052 -0,043 0,034 1,000
(8) Firm age -0,062 0,133* 0,014 0,033 -0,039 0,089 0,195** 1,000
(9) TMT size
-0,175*
*
0,1655* 0,051 0,052 -0,144* 0,041 -0,33 0,184** 1,000
(10) CEO age at year of deal
0,076 -0,163** -0,111 0,166** 0,249** -0,162* 0,277** 0,174** -0,165* 1,000
** Correlation is significant at the 0,01 level
* Correlation is significant at the 0,05 level
Figure 4 - Correlation matrix
As is stated by Lawrence and Lin (1989) the Pearson correlation coefficient indicates the deviation from each observation from the best-fit line. The correlation index should not be bigger than 0,8 since then the variables would correlate too much. The following table is provided by Mukaka (2012) as a rule of thumb.
Figure 3 - Correlation coefficient index table
19
The values of the correlation coefficient of the variables can be neglected since the correlation coefficients do not display any values above the 0,3. Therefore, multicollinearity is not a concern.
20
Regression analysis
Variable Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 5 Independent variable
CEO international experience
-0,081*
(0,046)
-0,093*
(0,047)
-0,079*
(0,046)
-0,063 (0,048)
-0,016 (0,158)
0,055 (0,160) Moderators
CEO gender
-0,256 (0,218)
0,157 (0,270)
0,159 (0,273) TMT international
experience
-0,153 (0,144)
-0,087 (0,215)
-0,038 (0,214) Interaction
CEO IE X CEO gender
-0,500**
(0,199)
-0,519**
(0,201) CEO IE X TMT
international experience
-0,046 (0,111)
-0,084 (0,111) Control variables
CEO tenure
0,000**
(0,000)
0,000**
(0,000)
0,000**
(0,000)
0,000**
(0,000)
0,000**
(0,000)
0,000**
(0,000)
0,000**
(0,000)
TMT tenure
0,664*
(0,401)
0,708*
(0,400)
0,760*
(0,402)
0,710*
(0,400)
0,663*
(0,399)
0,716*
(0,401)
0,676*
(0,399)
Firm size -0,003*
(0,002)
-0,004**
(0,002)
-0,004**
(0,002)
-0,004**
(0,002)
-0,004**
(0,002)
-0,004**
(0,002)
-0,004**
(0,002)
Firm age
-0,001 (0,002)
0,000 (0,002)
0,000 (0,002)
0,000 (0,002)
0,001 (0,002)
0,000 (0,002)
0,001 (0,002)
TMT size
-0,065**
(0,030)
-0,059**
(0,030)
-0,058*
(0,030)
-0,057*
(0,030)
-0,071**
(0,030)
-0,056*
(0,030)
-0,068**
(0,030)
CEO age at year of deal
0,011 (0,012)
0,008 (0,012)
0,007 (0,012)
0,011 (0,013)
0,001 (0,012)
0,012 (0,013)
0,005 (0,013)
Year 1 - 2012
0,066 (0,219)
0,056 (0,218)
0,072 (0,218)
0,042 (0,218)
0,062 (0,218)
0,051 (0,220)
0,106 (0,218)
Year 2 - 2013
0,017 (0,217)
0,008 (0,216)
0,038 (0,217)
0,005 (0,216)
0,010 (0,222)
0,005 (0,216)
0,113 (0,217)
Year 3 - 2014
0,211 (0,235)
0,252 (0,235)
0,295 (0,237)
0,223 (0,236)
0,167 (0,236)
0,238 (0,239)
0,266 (0,238) 21
Year 5 - 2016
0,035 (0,238)
0,038 (0,237)
0,045 (0,237)
0,017 (0,238)
-0,062 (0,2560
0,017 (0,238)
0,017 (0,235) Constants
R square 0,089 0,101 0,107 0,106 0,133 0,107 0,140
Adjusted R square 0,047 0,056 0,058 0,057 0,080 0,053 0,080
F Value 2,132 2,239 2,171 2,148 2,540 1,989 2,323
*** Significant at the 0,01 level
** Significant at the 0,05 level
* Significant at the 0,1 level
Figure 5 - Regression analysis
Model 1 describes the effect of the control variables on the merger and acquisition performance.
The tenure of the CEO shows significant effect on the M&A performance at the 5% level (B=0,000, t=2,245, p=0,026). Also, the TMT tenure show a significant effect at the 5% level (B=0,664, t=1,656, p=0,099). The firm size has a significant effect on the M&A performance (B=-0,003. t=-1,806, p=0,072). Furthermore, the TMT size also has a significant effect on the dependent variable at the 5% level (B= -0,065, t=-2,154, p=0,032). Moreover, the results do not show significance for the firm age (B=-0,001. t=-0,400, p=0,690) and CEO age at deal year (B=0,011, t=0,873, p=0,384). All the year control variables do not show significance as well.
Model 2 shows the effect of the independent variable on the merger and acquisition performance in addition to the five control variables. This is the model to test hypothesis 1. Hypothesis 1 states that the higher the international experience of the CEO, the better the merger and acquisition performance. This does have a significant result at the 10% level (B= -0,081, t=-1,759, p=0,080), but is negative. This would indicate that the international experience of the CEO has an influence on the M&A performance, but is negative. Therefore, hypothesis 1 is not supported.
Further, the control variable CEO tenure is significant at the 5% level (B=0,000, t= 2,217, p=0,028). In addition, the firm size aso shows significant results at the 5% level (B=-0,004, t=-2,113 p=0,036). Also, the control variable TMT size also shows significance at the 5% level (B=-0,059, t=-1,970, p=0,050). TMT tenure shows significance at the 10% (B=0,708, t=1,771, p=0,078). The other control variables do not show any significant effect. The firm age does not 22
have a significant effect on the performance of M&As (B=0,000, t=-0,110, p=0,912). The age of the CEO at the time of the deal also does not have a significant result (B=0,008, t=0,680, p=0,497). The year dummies also do not show significant results.
Model 3 shows the effect of the independent variable and the effect of the gender variable on the performance of merger and acquisition, in addition to the control variables. This model explains the direct effect of the variables on the merger and acquisition performance. The international experience of the CEO still has a significant effect, even a more significant effect, but slightly negative (B=-0,093, t=-1,963, p=0,051). However, the gender variable does not have a direct significant effect on the M&A performance (B=-0,256, t=-1,175, p=0,241).
Most of the control variables show significant effects. The CEO tenure shows significance at the 5% level (B=0,000, t=2,054, p=0,041). The firm size also shows significance at the 5% level (B=-0,004, t=-2,262, p=0,025).The TMT tenure shows significance at the 10% level (B=0,760, t=1,891, p=0,060). The results of the TMT size further indicate significance at the 10% level (B=-0,058, t=-1,942, p=0,053). The firm age, however, does not show significant results (B=0,000, t=-0,049, p=0,961). The year dummies still did not show significance.
Model 4 shows the effect of the independent variable and the effect of the TMT international experience variable on the M&A performance, in addition to the control variables. This model explains the direct effect of the variables on the performance of merger and acquisitions. The CEO international experience shows a slightly negative significant result at the 10% level (B=-0,079, t=-1,717, p=0,087). The TMT international experience does not show significant result (B=-0,153, t=-1,066, p=0,287). The control variable CEO tenure shows significance at the 5% level (B=0,000, t=2,105, p=0,036). The firm size shows significance at the 5% level (B=-0,004, t=-2,220, p=0,027). The TMT tenure shows significance in this model at the 10%
level (B=0,710, t=1,775, p=0,077). The TMT size also shows significant results at the 10% level (B=-0,057, t=-1,885, p=0,061). However, the firm age does not show significant results (B=0,000, t=-0,124, p=0,902) as well as the CEO age at the year of the deal (B=0,011, t=0,905, p=0,366). The year dummies are not significant.
23
Model 5 describes the effect of the CEO international experience, the effect of the CEO gender and the interaction effect of the CEO international experience and the CEO gender on the performance of M&A. This in addition to the control variables. This model test hypothesis 2, which states that when the CEO is a female, this will weaken the relationship between the CEO international experience and the M&A performance. This interaction effect shows negative significant results at the 5% level (B=-0,500, t=-2,517, p=0,013). Hypothesis 2 is supported and therefore accepted. This indicates that when the CEO is female, the relationship between the CEO international experience and the M&A performance is weakened. The results show insignificance for the CEO international experience variable (B=-0,063, t=-1,301, p=0,195) and for the direct effect of gender (B=0,157, t=0,579, p=0,563) on the M&A performance. The control variables CEO tenure ( B=0,000, t=2,431, p=0,016), firm size (B=-0,004, t=-2,207, p=0,028), TMT size (B=-0,071, t=-2,351, p=0,020) are significant at the 5% level, while TMT tenure is significant at the 10% level (B=0,663, t=1,662, p=0,098). Firm age (B=0,001, t=0,351, p= 0,726) and CEO age at year of deal (B=0,001, t=0,115, p=0,908) are not significant, as well as the year dummies.
Model 6 shows the effect of the CEO international experience, the effect of the TMT international experience and the interaction effect of the CEO and TMT international experience, in addition to the control variables. This model test hypothesis 3, which states that the international experience of the TMT, excluding the CEO, will strengthen the relationship between the CEO international experience and the performance of M&As. This interaction effect of CEO and TMT international experience is not significant (B=-0,046, t=-0,418, p=0,677). Therefore, hypothesis 3 is not supported and will be rejected. The results from the independent variable, the CEO international experience, on the dependent variable, M&A performance, is also not significant in this model (B=-0,016, t=-0,104, p=0,917). The direct effect of the TMT international experience on the M&A performance is also insignificant (B=-0,087, t=-0,403, p=0,688). The control variables CEO tenure (B=0,000, t=2,106, p=0,036), firm size (B=-0,004, t=-2,142, p=0,033) are significant at the 5% level, while TMT tenure
24
(B=0,716, t=1,788, p=0,075) and TMT size (B=-0,056, t=-1,856, p=0,065) are significant at the 10% level. Firm age (B=0,000, t=-0,097, p=0,923) and CEO age at year of deal (B=0,012, t=0,917, p=0,360) are not significant. The year dummies are also not significant.
Model 7 shows the complete model. However, in this model only the interaction effect of CEO international experience and CEO gender is negatively significant, at the 5% level. (B=-0,519, t=-2,579, p=0,011). The variable CEO international experience (B=0,055, t=0,341, p=0734), the interaction effect of the CEO international experience and the TMT international experience (B=-0,084, t=-0,761, p=0,447) are not significant, as well as the direct effect of CEO gender (B=0,159, t=0,538, p=0,560) and TMT international experience (B=-0,084, t=-0,761, p=0,447).
The control variables CEO tenure (B=0,000, t=2,325, p=0,021), firm size (B=-0,004, t=-2,196, p=0,029), TMT size (B=-0,068, t=-2,232 p=0,027) show significance at the 5% level, and TMT tenure shows significant results at the 10% level (B=0,676, t=1,693, p=0,092). The firm age (B=0,001, t=0,401, p=0,689) and CEO age at year of deal (B=0,005, t=0,368, p=0,713) are not significant. The year dummies are also not significant.
Robustness check
Since there is some inconsistency about the use of the variables for the international experience, a robustness check is necessary. In the first robustness check the international experience is ‘1’ if the CEO has international experience, and ‘0’ indicates that the CEO has no international experience.
25
