Master Thesis IB&M

UNIVERSITY OF GRONINGEN

Master Thesis IB&M

Board National Diversity and International

Diversification across Cultural Clusters

Prof. dr. D.J. (Dirk) Bezemer

d.j.bezemer@rug.nl

International Business & Mangement Faculteit Economie en Bedrijfskunde Global Economics & Management

Nettelbosje 2 9747 AE Groningen

Nederland

1 ABSTRACT

2 TABLE OF CONTENT

INTRODUCTION __________________________________________________________ 4 THEORY AND HYPOTHESES ______________________________________________ 6

International diversification __________________________________________________ 7 Board of directors _________________________________________________________ 8 National diversity ________________________________________________________ 10 The role of the CEO _______________________________________________________ 12 CEO’s nationality _______________________________________________________ 12 CEO’s power __________________________________________________________ 13

METHODOLOGY ________________________________________________________ 15

Sample _________________________________________________________________ 15 Variables _______________________________________________________________ 16 Dependent - International diversification ____________________________________ 16 Independent - Board national diversity ______________________________________ 17 Moderators ____________________________________________________________ 17 Control variables _______________________________________________________ 18 Method _________________________________________________________________ 19 Additional tests __________________________________________________________ 20 RESULTS ________________________________________________________________ 21 Descriptive statistics ______________________________________________________ 21 Fixed-effects (within) regression _____________________________________________ 22

DISCUSSION & CONCLUSION ____________________________________________ 25

Limitations and recommendations ____________________________________________ 27

3 LIST OF FIGURES AND TABLES

Figure 1. Conceptual model __________________________________________________ 15 Figure 2. Country clusters by Rhonen & Shenkar (1985). ___________________________ 34 Figure 3. Industry frequency __________________________________________________ 35 Figure 4. Hausman test in STATA _____________________________________________ 36 Figure 5. Heteroscedasticity test in STATA ______________________________________ 37

4 INTRODUCTION

The decision of where to establish a firm’s operations is a key element of a firm’s international strategy. As a result, the choice of location has become important in researching MNE’s. A common strategy to expand borders is Mergers & Acquisitions (M&As). M&A is a strategy which helps to create access to new unexplored markets with the possibility to grow. Besides opportunities, the firm will also face challenges when investing abroad. By means of cross-border M&As, the firm will get involved with other cultures with different norms, values and preferences. This makes strategic decisions very complex and more difficult compared to domestic investments. To better understand similarities and differences between countries, one can look at cultures. Ronen & Shenkar (1985) mapped nine country clusters based on previous cross-cultural literature. They used countries as unit of analysis since countries clearly delineate legal, political and social areas. These demarcated cultural clusters indicate a clearly defined environment on which a firm can anticipate its strategy on.

5

Since the board of directors normally consists of multiple individuals, it is possible to investigate the degree of national diversity within the board of directors. Diversity in origin is likely to lead to different strategic alternatives. Only a few studies have investigated the impact of national diversity on international diversification. For example, Caligiuri, Lazarova, & Zehetbauer (2004) found that board national diversity is positively related to four indices of firm-internationalization. Pisani, Muller, & Bogăţana, 2018 and Nielsen (2010) found a positive relationship between TMT internationalization and firm-internationalization. Also Punnett & Clemens (1999) found that national heterogeneous teams are more likely to consider international expansion in comparison to homogeneous teams. Hence, previous literature found positive effects between national diversity and firm-internationalization. However, firm-internationalization is often measured in terms of geographical dispersion of the firms’ operations. This study examines the influence of board national diversity on international-diversification across cultural clusters. Theoretically, it is expected that a national diverse board will increase the firms’ international diversification across cultural clusters. This is based on Upper Echelon Theory, whereby it is argued that international diversification is influenced by the nationality of the directors.

6

CEOs prefer to stay committed to the status quo (Haynes & Hillman, 2010). Safety of the current strategy compared to the risk of strategic change is a possible explanation for the CEO preferring the status quo. Therefore, it is expected that CEO power weakens the relationship between board national diversity and international diversification.

Overall, this study will provide more insight into the relationship between board national diversity and international diversification of the MNE moderated by the CEO’s nationality and power. As a result, the research question is formulated as “how does board national diversity influence the MNE’s international diversification across cultural clusters and do the CEO’s nationality and power moderate this relationship?”.

This study contributes to the current literature by selecting the board as unit of analysis and researching its effect on international diversification across cultural clusters. Second, this study may give more insight into the influence of board national diversity as a less researched Upper Echelon characteristic in the current literature. Last, this study not only focuses on group-level but also investigates the possible moderation of the CEO’s nationality and power. By doing so, the context in which the board takes strategic decisions is more delineated. Most literature focused on potential external moderators whereas this study looks at the internal context.

This study is organized as follows. First, relevant literature and theories are provided from which the hypotheses are formulated. Second, the sample, variables and method are described in the methodology. Third, the descriptive statistics and the fixed effects regression are carried out including a description of the empirical results. This study concludes with a discussion and conclusion followed by the study its limitations and recommendations.

THEORY AND HYPOTHESES

7

Hereafter follows the explanation about the possible moderation of CEO nationality and power. Consequently, the formulation of the second and the third hypothesis follows.

International diversification

The development of technologies and communications has changed the scope of doing business. As a result, firms expand their national borders to more distant markets. Hence, international diversification is a reflection of the distribution of a firm’s operations across different regions. International diversification is a process which consists of a set of key strategic decisions. These key decisions are the choice of location, the type of entry mode, degree of investment, and the degree of ownership (Beugelsdijk, Kostova, Kunst, Spadafora, & van Essen, 2018).

International success depends on the degree to which a company is able to deal with complexities which derive from differences between cultures, institutions and competitiveness in environments. Therefore, executives have to deal with very complex strategic decisions when internationalizing (Li & Lo, 2017). However, when a firm is successful in international diversification, then it can take advantage of the spread monetary risks, gain access to assets, technologies, local knowledge and local resources. On the other hand, expanding abroad involves additional costs due to economic and political hazards and the so-called liability of foreignness. Liability of foreignness can be defined as “all additional costs a firm operating in a market overseas incurs that a local firm would not incur” (Zaheer, 1995: 342-343). These costs are due to lack of market knowledge of the host country and the limited availability of information about opportunities and risks in that host country (Beugelsdijk et al., 2018). Not understanding consumer preferences, rules, and e.g. distribution channels, can lead to high costs. In addition, firm internationalization may cause organizational problems due to the increased firm size and the complexity which may result in increased monitoring costs. The question of how firms are able to cope with the complexity and uncertainties related to internationalization remains an important field of interest in the strategic management literature. Most cross-national research focused on geographic distance between the home and the host country. However, far less studies examined international diversification by means of cultural distance.

8

(e.g. United Kingdom, North America & Australia). This can be explained by historical colonization and immigration (Filippaios & Rama, 2011). Based on theory, the influence of cultural distance on firm internationalization is a key element of the well-known Uppsala internationalization process model. This model shows that firms tend to expand to markets/countries that are near to the home country that hold a similar culture. Once the company has gained experience in expanding to foreign markets in an incremental manner, it will slowly start entering other countries which are more geographically and cultural distant (Johanson & Vahlne, 1977). Therefore, cultural distance is an important factor for the international strategy of the firm.

Ronen & Shenkar (1985) divided countries into cultural clusters based on eight previous studies. They present nine country clusters (Nordic, Germanic, Anglo, Latin European, Latin American, Far East, Arab, Near East and Independent cluster) which include non-communistic countries. These clusters help to distinguish markets based on legislation, politics and social values. In this way, firms are able to match their resources with their international strategy.

This study will use the synthesis of cultural clusters by Ronen & Shenkar (1985) to examine international diversification across cultural clusters. Only a few studies investigated cultural dispersion of operations based on the cultural clusters by Ronen & Shenkar (1985). For example, Greve, Nielsen & Ruigrok (2009) and Carpenter & Fredrickson (2001) examined cultural dispersion of MNE’s subsidiaries. Barkema & Shvyrkov (2007) researched the spread of investments across cultural clusters. Last, Filippaios & Rama (2011) examined cultural dispersion of MNE’s affiliates.

The decision to invest in a unexplored market is determined by the board of directors. The next paragraph will explain that director characteristics are able to influence strategic choices and thus the international diversification of the firm.

Board of directors

9

partly a reflection of a upper manager's background characteristics (Hambrick & Mason, 1984). A key element of this theory is that the cognition, values and experiences of executives are important influencers of strategic decisions. In addition, this theory proposes that observable characteristics such as age, functional tracks, experience, education, financial position, socioeconomic roots, and group characteristics serve as proxies for strategic choices and therefore also for organizational performance (Tihanyi et al., 2000).

Originally, the Upper Echelon Theory selected the TMT as unit of analysis. However, Finkelstein, Hambrick & Cannella (2009) identified that the recent literature also applies the Upper Echelon Theory to the board and its influence on organizational outcomes.

The Board of Directors is a group of non-executives, executives and a chairman which are selected by the shareholders of the firm. Non-executive directors are seen as independent and objective individuals which do not perform executive tasks (Hillman, Cannella, & Paetzold, 2000). Unlike non-executive directors, executive directors are members of the executive management team. This team typically consists of a chief executive officer (CEO), a chief financial officer (CFO) and a chief operating officer (COO). These executives are responsible for strategic decision making. Besides the role of corporate governance, executives are also employees of the firm. In addition, the board includes a chairman. The chairman presides over meetings.

Furthermore, there are two types of governance structures, namely a one tier and a two tier board. Across countries, the governance structure of boards varies. This study focuses on MNEs in the United States (US) which carry a one tier (unitary) board structure.

10 National diversity

The previous paragraph described that the Upper Echelon Theory explains how an executive may interpret a given situation which is influenced by its cognitive base, values and demographic characteristics. This interpretation will eventually influence the strategic choice made by the director. One of the demographic characteristics that influences a directors’ cognitive base, values and behaviour is his nationality.

Nationality is one of the factors which predicts an individual’s psychological traits and behaviour. Nationality influences a person in several ways, namely its values, cognitive schema, physical behaviour, and language (Hambrick et al., 1998).

First, nationality influences a person’s values. A person’s values are defined as “a broad tendency to prefer certain states of affairs over others” (Hofstede, 1980: 19). These personal values are shaped by national culture. Previous studies found that underlying values within decision making tasks are greatly influenced by a manager’s national system of belief (Bendix, 1956; Sutton, Seymour, Carl, & Tobin, 1956; Chatov, 1973). Second, nationality influences a person’s cognitive schema. Cognitive schemas form a framework for having certain knowledge, considering alternatives, and looking at consequences (Hambrick et al., 1998). In addition, nationality also affects one’s interpretation and processing of new cognitive content. Third, physical behaviour such as non-verbal communication is also influenced by a person’s nationality. Last, nationality has a great influence on language. Besides the native language, someone’s nationality also influences how many other languages a person may speak. This is due to the geographic location of the place where the individual grew up (Hambrick et al., 1998). Thus, a person’s values, cognition and behaviour are partly a reflection of its nationality.

In a group, the underlying features of nationality tend to interact and reflect on the behaviour and characteristics of other group members’ nationality (Hambrick et al., 1998). Hence, national diversity in the board can serve as an indicator for the actual diversity of cultural values and country-specific knowledge (Greve et al., 2009). Therefore, national diversity in the board could influence where firms go.

11

conflicts are considered as improvement of decision making quality (Nielsen & Nielsen, 2011). As result of disagreement, directors are more aware of the alternative strategies and their consequences. Without disagreement, directors would be less likely to think of alternatives and the possible pros and cons of other options.

Besides different cognitive schemas, directors have also developed country specific institutional knowledge that can contribute in their role as board member (Nielsen & Nielsen, 2013). Formal institutional knowledge consists of a country’s explicit political and economic rules whereas informal institutional knowledge consists of social norms and values of a society. Together, country specific institutions form a framework of how people and organizations deal with the environment and its uncertainties (Nielsen & Nielsen, 2013). Consequently it can be argued that the greater national diversity within the board, the greater the country specific knowledge of the firm.

Furthermore, Thams (2013) argues that executives develop country specific skills. These country specific skills can be defined as “managers’ abilities that are developed in a particular national institutional context” (Thams, 2013:13). Besides that, Pisani et al. (2018) argue that an individual’s decision to work abroad comes from his/her cognition and attitude which influences the individual’s choices. The openness towards foreign countries of such a person leads to high interest in international opportunities. In the board, this international attention stimulates the communication and discussion of foreign markets and thus the development of international diversification (Bouquet & Birkinshaw, 2011).

Hence, the more different nationalities within the board, the more different country specific knowledge and skills, which helps a firm to reduce the liability of foreignness when entering a new market. These resources of institutional knowledge, experience, and specific know-how improves board decision-making and information input (Carter, D’Souza, Simkins, & Simpson, 2010). Therefore, the composition of nationalities affect internationalization decisions of the firm (Caligiuri et al., 2004).

12

TMT internationalization had a positive influence on firm-internationalization based on foreign sales ratio and ratio of international majority-owned affiliates. Nielsen (2010) found a positive relationship between TMT internationalization and FDI. Last, Punnett (1999) found that national diverse teams are more likely to consider international expansion in comparison to homogeneous teams.

In conclusion, national diversity is expected to positively influence international diversification because of the diversity in values, cognitive schemas, skills and abilities. These factors and the familiarity with different institutions helps to overcome the complexity and liability of foreignness when entering a new market. The more diverse the directors’ countries of origin, the greater the country knowledge from which the firm can benefit, which makes it easier to enter an unexplored market. Hence, board national diversity is expected to influence the selection of potential countries for the firm to internationalize. The above discussion about international diversification and board national diversity result in the following hypothesis.

Hypothesis 1: Board national diversity positively influences the MNE’s international

diversification across cultural clusters. Specifically, greater board national diversity will result in greater international diversification within different cultural clusters.

The role of the CEO

Of the few studies which did research board national diversity and its relationship with firm-internationalization, none of them included the CEO as a separate unit of analysis. This is remarkable because the CEO’s nationality and power are key influencers of strategic choices (Cannella, Finkelstein, & Hambrick, 2008). Moreover, the CEO is responsible for the conduct and performance of a firm. As a result, the CEO has a significant impact on strategic decision making. Hence, the potential moderating roles of the CEO’s nationality and power are discussed in the following paragraphs.

CEO’s nationality

13

interpreted by means of the CEO’s cognitive base, values and nationality as one of the upper echelon characteristics (Hambrick & Manson, 1984). Hereby is the CEO’s nationality a determinant of his/her strategic choices.

Zhou (1997) found that people who live in a country other than their country of birth, still have their birth country as reference framework. This could be an explanation why a person, who grew up in a certain culture from a young age, is influenced by his/her nationality when making strategic decisions in his/her later life (Thams, 2013). Thus, strategic decisions of native and foreign CEOs may differ. Globalization may lead firms to seek CEOs from outside the home country, since management skills may be country specific. “Firms engaging in geographical diversification can achieve superior returns when their diversification efforts match the profile of their human capital” (Thams, 2013: 22). Foreign CEOs are more likely to have a global perspective as a result of their own international experiences. In addition, executives who work abroad are able to withstand or recover quickly from setbacks and difficulties, are highly motivated and have a proactive attitude (Pisani et al., 2018). As a result of the CEO’s international experiences with multiple cultures, the non-native CEO is more likely to understand and appreciate foreign cultures. The foreign CEO is therefore able to bridge various national cultures of board members (Thams, 2013). Thams (2013) found that foreign CEOs are more likely to engage in more cross-border M&A changes over time relative to national CEOs.

In summary, a foreign CEO is expected to stimulate a national diverse board which leads to greater international diversification across cultural clusters. This results in the following hypothesis.

Hypothesis 2: The effect of board national diversity on international diversification is

positively moderated by the degree of the CEO’s nationality. Specifically, the CEO’s non-native nationality is expected to strengthen the relationship between board national diversity and international diversification across cultural clusters.

CEO’s power

14

& Johnson, 1997). CEO power is relative to the power distribution within the board whereby “less powerful members are prone not to voice their concerns or to be ignored if they do, leading to domination by the powerful members” (Greve & Mitsuhashi, 2007: 1203).

Finkelstein (1992) found four dimensions of power; (1) structural power, (2) ownership power, (3) expert power, and (4) prestige power. The first dimension of power is the degree of structural power which is determined by the executive’s formal organizational position. This type of power influences to which extent the CEO can control the board based on the hierarchy and organizational position within the MNE. Second, the degree of ownership influences the degree of control on strategic decisions. Executives who are family of the founders, or executives who have many shares can exercise greater power. The third dimension involves an executive’s degree of experience with uncertainties. The more often an executive is exposed to an uncertain situation, the better he knows how to deal with it, and therefore the greater his expert power to execute decisions. Lastly, prestige power involves the reputation or status of an executive. This reputation among the internal and external environment influences the perception of these stakeholders. Thus, executives can use their reputation to gain contacts or information from other parties which may be valuable for the company (Finkelstein, 1992). According to Finkelstein (1992), these four dimensions define the degree of decision making power.

15 status quo (Haynes & Hillman, 2010). In conclusion, a powerful CEO will reduce the effect of board diversity on international diversification due to its commitment to the status quo. Hence, it is expected that CEO power will weaken the relationship between board national diversity on international diversification across cultural clusters. This results in the following hypothesis.

Hypothesis 3: The effect of board national diversity on international diversification is

negatively moderated by the degree of CEO power. Specifically, when CEO power is high, the relationship between board national diversity and international diversification across cultural clusters weakens.

Overall, the formulated hypotheses are shown as conceptual model in figure 1.

Figure 1. Conceptual model

METHODOLOGY

As a result of the theory, three hypotheses are formulated. This chapter describes how these hypotheses will be analysed. First, the sample is described followed by the variables and the research method. Lastly, several additional tests are performed.

Sample

16

structures are similar. US MNE’s apply a one-tier board in which executive managers are part of the board of directors.

The relationship will be tested by means of two combined secondary databases. Demographic data on directors was collected by means of the BoardEx database. Second, Zephyr database provided M&A data. Data on M&A included completed deals from 2014 to 2016. The independent and control variables include data from 2013 to 2015. For the dependent variable t + 1 is applied because it is assumed that the period between a strategic investment decision by the board and the completion of that investment takes time.

To combine the data from the different databases, ISIN (International Securities Identification Number) is used. Firms with missing data and companies active in utilities and the financial sector were removed from the sample. So, only US MNEs which had complete data for 2013 to 2016 are in the sample. This resulted in a final sample of 127 US MNEs.

Variables

Dependent - International diversification

The dependent variable in this study is international diversification. Prior studies used various measurements regarding international diversification. Most studies use the ratio of foreign sales to total sales (FSTS) (Li & Lo, 2017; Caligiuri et al., 2004; Tihanyi et al., 2000) , the ratio of foreign assets to total assets (FATA) (Pisani et al., 2018; Caligiuri et al., 2004; Herrman & Datta, 2005), the ratio of foreign employees to total employees (FETE), the number of countries or the number of subsidiaries (Li & Lo, 2017; Pisani et al, 2018; Caligiuri et al., 2004). To measure international diversification pattern, this study follows one single item measurements used by Greve et al. (2009), namely cultural dispersion of the MNE. Based on the theory, the main interest of this study is to examine the firm’s international diversification across cultural clusters. Therefore, geographic and human capital dispersion are not included in the measurement of the dependent variable.

17

clusters of Rhonen & Shenkar (1985), this study adds Africa (including all African countries) as the tenth cultural zone. Following Carpenter & Fredrickson (2001), each M&A deal per cultural cluster counts for a value of 0.1. For example, a firm which completed four M&A deals in four different cultural clusters in the same year, generates a score of 0.4. If no M&A deal took place for a firm in a specific year, than the dependent variable has a value of 0. Furthermore, the measurement of the dependent variable does not distinguish whether a deal has taken place in the domestic cluster or within another cluster. So, a value of 0.1 may indicate a deal in the domestic cluster (Anglo) or could mean that this deal took place in another cluster.

Country presence applies when at least one M&A deal was completed in a certain country per year. No merger deals appeared in the examined sample. Thus, the data includes only completed acquisition deals in 2014-2016.

Note that as explained in the theory section, this study focuses on the international dispersion across cultural clusters by means of M&A deals. Cultural distance tends to vary for different types of industry sectors (Filippaios & Rama, 2001). Due to lack of time to take these differences into account and because of the essence of the research question, it was decided not to include the distance between cultures in the measurement.

Independent - Board national diversity

A common way to measure group national diversity is the Blau index. The Blau index is used as a measure for group heterogeneity. The Blau index includes the distribution of board directors across all present nationalities. The formula [B ∑ ], where p is the percentage of directors with nationality i. The higher the value of B, the greater is the heterogeneity on nationality of the board (Nielsen & Nielsen 2011; Greve et al., 2009; Rivas, 2012). Hence, a value of 0 indicates a homogeneous board whereas a value of 1 indicates a heterogeneous board of nationalities. A value of 0.5 indicates an equal amount of nationalities within the board.

Moderators

18

national diversity which leads to increased international diversification across different cultural clusters.

Finkelstein (1992) presented structural, ownership, expert, and prestige power as four valid and reliable dimensions to measure CEO power. Since the measurement of Finkelstein is very extensive, this study uses CEO tenure and duality as dimensions to measure CEO Power. Tenure of the CEO is an indicator of key knowledge and relationships built during the fulfilment of its position (Combs, Ketchen, Perryman, & Donahue, 2007; Greve et al., 2007). Tenure of the CEO is measured in the number of years the CEO holds its position in the board and therefore serves as a measurement of Finkelstein’s expert power. Tenure was split into two categories, namely CEOs with less than 5 years board tenure and 5 or more years board tenure. This was measured by means of a dummy variable which takes a value of 0 if the CEO had hold its position in the board for less than 5 years whereas a value of 1 indicates 5 or more years of experience within the board.

Duality is a part of structural power since it indicates the number of titles the CEO has. Duality is also measured by means of a dummy variable which takes the value of 1 when the CEO is also the chairman of the board and 0 if not (Abebe, Angriawan, & Liu, 2011; Combs et al., 2007). According to Combs et al. (2007) tenure, ownership & duality are most applied and supported in the current literature to measure CEO power. However, no ownership data was available and therefore it was not included as measurement for CEO Power. Tenure and duality values where summed, to provide a standardized measure for CEO Power (Abebe et al., 2011). Hence, CEO Power has a range of 0-2, where 0 represents low power, 1 medium power and a value of 2 indicates high power of the CEO. CEO Power is expected to weaken board national diversity and its influence on international diversification across different cultural clusters.

Control variables

19 Board size is included as a control variable since the group quantity may influence the group its heterogeneity degree (Barkema & Shvyrkov, 2006). Larger boards may be more heterogeneous since the board may include more diverse backgrounds (Tihanyi et al., 2000). Each individual of the board can be seen as an unique resource. Therefore, to limit the possible influence of board size is has been selected as a control variable. Board size is measured as the number of individual directors of the board.

Acquirer M&A experience is important to control the successful completion of a M&A deal which may be influenced by previous deals. Namely, with similar M&A deals the MNE can fall back on their routines which are gained through previous experience of the similar country cluster. In this study, acquirer M&A experience includes the total count of previous M&A deals across all possible cultural clusters.

Acquirer host M&A experience is measured as the sum of previous deals outside the domestic cluster (Anglo). Previous experience outside the home country may influence the international diversification decisions of the MNE. Experience and knowledge of doing business in different cultural clusters besides the domestic cluster, may decrease uncertainties and help to overcome the liability of foreignness which makes it easier to internationalize the firm’s operations.

For the M&A deals completed in 2014, acquirer M&A (host cluster) experience is counted from 2009 to 2013. For the M&A deals completed in 2015, acquirer M&A (host cluster) experience is counted from 2009 to 2014. For the M&A deals completed in 2016, acquirer M&A (host cluster) experience is counted from 2009 to 2015.

Acquirer industry is controlled by means of dummy variables. Financial institutions and utilities are excluded industry sectors since they have a different organizational structure (Greve et al., 2009). The industry classification of BoardEx was used, resulting in 27 unique sectors. The type of industry may influence the number of M&A deals, since M&A may be more common in some industries compared to the other ones. For a list of all included industries see table 3 in Appendix B.

Method

20

can be researched (Frees, 2004). Therefore, a panel data regression is carried out to test the relationship between board national diversity and international diversification including moderators. Panel data analysis includes two separate statistical techniques; the fixed effects model and the random effects model. The difference between these two techniques is “whether the unobserved individual effect embodies elements that are correlated with the regressors in the model, not whether these effects are stochastic or not” (Greene, 2008: 183). To decide which model fits the data best, the Hausman test was done. Doing this test, the null hypothesis includes the random effects model whereas the alternative hypothesis includes the fixed effects model. It tests if the unique errors are correlated with the regressors. If they are correlated, the null hypothesis is rejected and the fixed effects model should be used (Greene, 2008). From the Hausman test based on the data in this study, see figure 4 in Appendix C, it can be concluded that the null hypothesis can be rejected since Prob>chi2 = 0.0109. As a result, the fixed effects model is preferred because the unique errors (ui) are correlated with

the regressors. To conclude, the relationship between board national diversity on international diversification including the moderation of CEO nationality and CEO power will be tested by means of a fixed effects regression.

Additional tests

After the conclusion to use the fixed effects model for the panel data analysis, some additional test were done.

21

effect. In addition, acquirer host M&A experience is highly correlated with acquirer M&A experience because acquirer M&A experience includes acquirer host M&A experience terms. Therefore, both signs of multicollinearity are not seen as problematic.

Last, the data is tested on heteroscedasticity by means of a Breusch-Pagan/Cook-Weisberg test. Heteroscedasticity refers to data with unequal variability across predictor variables. Figure 5 in Appendix C shows that Prob > chi2 is smaller than 0.5 which indicates no sign of heteroskedasticity.

RESULTS

In this chapter the descriptive statistics and regression analysis are performed. This chapter concludes with a discussion of the empirical findings.

Descriptive statistics

Table 1 shows the descriptive statistics of 127 S&P companies sampled over 3 years. Hence, per variable there were 381 observations in total.

Looking at international diversification in table 1, it is shown that in 2015, there is a small increase of deals in one or more acquisitions across the cultural clusters compared to 2014. However, in 2016 the number of deals decreased compared to the previous years. In addition, there was less international diversification across the cultural clusters since the maximum amount of clusters was two in 2016. Hence, international diversification was greatest in 2015 followed by 2014 and the least in 2016. Second, board national diversity was greatest in 2013 on average. The greatest diversity score (0.740) was observed in 2015. However, striking is the strong decline of board diversity in 2015 on average. In this year, an increased number of foreign directors were replaced by Americans.

With regard to the moderators, it can be concluded that in 2015 there were less foreign CEOs on average compared to the previous years. In addition, CEO power was greatest in the first year.

22

Variable Year Observation Mean Std.Dev. Min Max

Int. Diversification 2014 127 .038 .059 0 .30

2015 127 .044 .063 0 .30

2016 127 .032 .053 0 .20

Board national diversity 2013 127 .480 .139 0 .702

2014 127 .471 .135 0 .719 2015 127 .096 .126 0 .740 CEO Nationality 2013 127 .228 .421 0 1 2014 127 .244 .431 0 1 2015 127 .094 .294 0 1 CEO Power 2013 127 1.024 .761 0 2 2014 127 .457 .531 0 2 2015 127 .921 .762 0 2

M1 (Board national diversity*CEO Nationality) 2013 127 .128 .240 0 .675

2014 127 .143 .255 0 .719

2015 127 .030 .107 0 .740

M2 (Board national diversity*CEO Power) 2013 127 .469 .372 0 1.254

2014 127 .198 .246 0 .797 2015 127 .083 .155 0 .941 (log)Firm size 2013 127 10.765 1.292 7.512 14.604 2014 127 10.729 1.324 7.481 14.604 2015 127 10.787 1.303 7.576 14.648 Board size 2013 127 16.44 2.284 9 24 2014 127 16.47 2.077 9 22 2015 127 16.78 2.149 12 26

Acquirer M&A experience 2014 127 2.276 3.023 0 19

2015 127 2.827 3.492 0 21

2016 127 3.378 4.086 0 23

Acquirer host M&A experience 2014 127 .528 1.097 0 6

2015 127 .654 1.306 0 7

2016 127 .787 1.551 0 8

Table 1. Descriptive Statistics

Based on figure 3 in Appendix B it can be concluded that industry sectors Pharmaceuticals and Biotechnology and Oil & Gas appeared most frequently in the sample.

Fixed-effects (within) regression

23

(a constant), cannot influence something over time. So, the fixed effect model only uses variation within companies.

In this section, four regression models are carried out in table 2. The first model (1) includes the dependent and control variables. In the second model (2), the independent variable is also included. The third model (3) includes the dependent, control, independent variables and both moderators. However, the moderators are not multiplied with the dependent variable. The last model (4) is complete by including the multiplication within the moderators.

Since all models are found to be significant (Prob > F = <0.05), it is enabled to interpret the results. Furthermore, the within coefficient of determination increases per model. In the complete model, 24,10% of international diversification is explained by the independent variables.

In the theory section three hypotheses were formulated which will now be discussed based on the empirical results found in table 2.

The first hypothesis included the main effect between board national diversity and its positive influence on the MNE’s international diversification across cultural clusters. Based on regression model 2, it can be concluded that board national diversity does not significantly influence international diversification. This indicates that board national diversity has no influence on the international diversification pattern of the US MNE’s across different cultural clusters. Hence, no support is found for hypothesis 1.

The second hypothesis included the moderation of the CEO’s nationality which was expected to strengthen the relationship between board national diversity and international diversification. This relationship was tested by means of regression model 4. Based on regression model 4, it can be concluded that no moderation effect of CEO nationality was found. Hence, no support is found for hypothesis 2.

24

Fixed-effects (within) regression

Int. diversification Model 1 Model 2 Model 3 Model 4

(log) Firm size 0.009 0.012 0.013 0.015

(0.03) (0.03) (0.03) (0.03)

BoD size 0.002 0.003 0.003 0.003

(0.00) (0.00) (0.00) (0.00)

Acquirer M&A experience 0.021** 0.026*** 0.026*** 0.026***

(0.01) (0.01) (0.01) (0.01)

Acquirer host M&A experience 0.025 0.026 0.025 0.027

(0.02) (0.02) (0.02) (0.02)

Board National diversity 0.022 0.012 -0.006

(0.01) (0.02) (0.02) CEO Nationality 0.002 -0.014 (0.01) (0.03) CEO Power -0.009 -0.013 (0.01) (0.01)

M1 (Board national diversity*CEO Nationality) 0.037

(0.06)

M2 (Board national diversity*CEO Power) 0.027

(0.02) Constant -0.179 -0.253 -0.249 -0.267 (0.36) (0.36) (0.36) (0.36) Number of observations 254 254 254 254 R-squared within 0.2080 0.2224 0.2316 0.2410 R-squared between 0.4970 0.4949 0.4987 0.4944 R-squared overall 0.3912 0.3878 0.3921 0.3896 F-test 8.074 6.977 5.167 4.164 Prob > F 0.000 0.000 0.000 0.000 *** p<0.01, ** p<0.05, * p<0.1

Table 2. Fixed-effects (within) regression table

25 DISCUSSION & CONCLUSION

The purpose of this study was to examine the relationship between board national diversity and its influence on international diversification across cultural clusters including the moderation of CEO nationality and power. These relationships were tested by means of three hypotheses.

26

which decreases the reliability and validity of a study since a single variable cannot measure the complex domain of the degree of internationalization (DOI).

The second hypothesis included CEO nationality as moderator. Theoretically, it was found that the cognitive framework of non-native individuals differs from the cognitive framework of native persons. They tend to have a more global mind-set as a reflection of their own international experiences. CEO nationality was expected to strengthen board national diversity and international diversification. However, no significant moderation effect of CEO nationality on board national diversity and international diversification was found.

In contrast to this study, Thams (2013) found significant positive evidence that non-native CEOs are more likely to implement more changes in cross-border M&A compared to native CEOs. However, Thams (2013) used a greater sample including firms from 36 different countries. Therefore, one of the reasons why Thams (2013) found different results could be because larger sample size generates more reliable results.

The last hypothesis examined the moderating role of CEO power. Research found that the CEO is a key person in strategic decision making. In this regard, the strength of the CEO’s power plays a crucial role. Previous literature argues that powerful CEOs are persistent to strategic change. Therefore, CEO Power was expected to weaken the relationship between board national diversity and international diversification. However, no evidence was found to support this moderation. This result is contradictory to previous empirical results. For example, Haynes & Hillman (2010) found significant support for CEO power to weaken the relationship between board diversity and strategic change. Also Golden & Zajac (2001) found that when the CEO is powerful, the board will have less effect on strategic change. Differences in the findings of this study and other empirical results may be caused by differences in the measurement of power. For example, Haynes & Hillman (2010) and Golden & Zajac (2001) both used four indicators to measure CEO power whereas this study only used two indicators.

27

First, it can be concluded that board national diversity does not influence the MNE’s international diversification across cultural clusters. Second, no evidence was found to infer that CEO nationality moderates the relationship between board national diversity and international diversification. Last, also CEO power had no significant moderation effect on the main relationship.

The findings of this study contribute to the current literature even though the alternative hypotheses could not be accepted. Previous studies which investigated firm-internationalization mainly focused on TMT as the unit of analysis. This study sheds light on the role of the Board of Directors on firm-internationalization. Moreover, very few studies have focused on cultural dispersion of the MNE. In addition to the Board of Directors, this study also contributes by highlighting the role of the CEO. Even though no moderation effect was observed, the results contribute to the understanding of the complex influence of the CEO on the composition of the board and its influence on the MNE’s international diversification pattern across cultural clusters.

Limitations and recommendations

The results of this study should be viewed in light of its limitations. First, the sample only included US MNE’s. Therefore the results are not generalizable for MNE’s located in different countries. Future research can cover this by including also MNE’s from different continents. Furthermore, to make results more reliable and powerful, the sample size is recommended to be increased.

The second limitation is the measurement of international diversification across cultural clusters. As mentioned previously, cultural distance between the cultural clusters is not taken into account. This results in a loss of the richness of the data. Furthermore, no distinction has been made whether a single M&A investment has been made in a domestic cluster or in a host cultural cluster. However, this study controlled the acquirer’s host cluster investment experience. Therefore, future research could add more insight into the cultural distances between the home and host cluster by including cultural distance (e.g. Kogut & Singh, 1988) in the measurement of international diversification.

28

that international diversified MNE’s attract more diverse directors. For example Greve et al. (2009) found evidence for MNE’s to match top executive positions to their strategies.

Fourth, Finkelstein (1992) emphasized the extensive measurement of CEO power. In this study, only two types of power are taken into account which may lead to different results than when all four measurements were applied. Future research may include all four CEO power measurements presented by Finkelstein (1992) to increase validity.

29 REFERENCES

Abebe, M. A., Angriawan, A., & Liu, Y. 2011. CEO power and organizational turnaround in declining firms: Does environment play a role? Journal of Leadership & Organizational Studies, 18: 260-273.

Barkema, H.G., & Shvyrkov, O., 2007. Does top management team diversity promote or hamper foreign expansion? Strategic Management Journal, 28: 663–680.

Bendix, R. 1956. Work and authority in industry. New York: Wiley.

Beugelsdijk, S., Kostova, T., Kunst, V. E., Spadafora, E., & Essen van, M. 2018. Cultural Distance and Firm Internationalization: A Meta-Analystical Review and Theorectical Implications. Journal of Management, 44 (1): 89-130.

Bouquet, C. & Birkinshaw, J. 2011. How global strategies emerge: an attention perspective. Global Strategy Journal, 1: 243–262.

Caligiuri, P., Lazarova, M., & Zehetbauer, S. 2004. Top managers' national diversity and boundary spanning: Attitudinal indicators of a firm's internationalization. Journal of Management Development, 23 (9): 848-859.

Canella, B., Finkelstein, S. & Hambrick, D., 2008. Strategic Leadership: Theory and Research on Executives, Top Management Teams, and Boards. 1 ed. New York: Oxford University Press.

Carpenter, M. A. & Fredrickson, J. W. 2001. Top management teams, global strategic posture and the moderating role of uncertainty. Academy of Management Journal, 44: 533–545.

Carter, D.A., D’Souza, F., Simkins, B.J., & Simpson, W.G. 2010. The Gender and Ethnic Diversity of US Boards and Board Committees and Firm Financial Performance. Corporate Governance: An International Review, 18(5): 396–414.

30

Combs, J., Ketchen, D., Perryman, A., & Donahue, M. 2007. The moderating effect of CEO power on the board composition-firm performance relationship. Journal of Management Studies, 44: 1299-1323.

Daily, C. M. & Johnson, J. L. 1997. Sources of CEO power and firm financial performance: a longitudinal assessment. Journal of Management, 23: 97–117.

Filippaios, F. & Rama, R. 2011. Cultural Distance and Internationalization: The World’s Largest Food and Drink Multinationals. Agribusiness, 27 (4): 399–419.

Finkelstein, S. 1992. Power in Top Management Teams: Dimensions, Measurement, and Validation. The Academy of Management Journal, 35 (3): 505-538.

Finkelstein, S., Cannella, B. & Hambrick, D., 2009. Strategic Leadership: Theory and Research on Executives, Top Management Teams, and Boards. 1 ed. New York: Oxford University Press.

Frees, E.W. 2004. Longitudinal and Panel Data. Madison: Cambridge University Press. Golden, B.R., Zajac, E.J. 2001. When will boards influence strategy? Inclination x power =

strategic change. Strategic Management Journal, 22: 1087-1111.

Greene, W.H. 2008. Econometric analysis. 6th ed. Upper Saddle River, N.J. : Prentice Hall. Greve, H., & Mitsuhashi, H. 2007. Power and glory: Concentrated power in top management

teams. Organization Studies, 28: 1197-1221.

Greve, P., Nielsen, S., & Ruigrok, W. 2009. Transcending borders with international top management teams: A study of European financial multinational corporations. European Management Journal, 27: 213– 224.

Hambrick, D.C. & Mason, P.A. 1984. Upper echelons: the organization as a reflection of its top managers, Academy of Management Review, 9 (2): 193-206.

31

Hambrick, D.C., Geletkanycz, M.A., & Fredrickson, J.W. 1993. Top executive commitment to the status quo: some tests of its determinants. Strategic Management Journal, 14 (6): 401-418.

Haynes, K.T. & Hillman, A. 2010. The effect of board capital and CEO power on strategic change. Strategic Management Journal, 31 (11): 1145-1163.

Herrmann, P. and Datta, D.K. 2005. Relationships between top management team characteristics and international diversification: an empirical investigation, British Journal of Management, 16 (1): 69-78.

Hillman, A.J., Cannella, A.A. & Paetzold, R.L. 2000. The resource dependence role of corporate directors: strategic adaptation of board composition in response to environmental change. Journal of Management Studies, 37(2): 235-255.

Hofstede, G. 1980. Culture's consequences: International differences in work-related values. Beverly Hills, CA and London: Sage.

Johanson, J. & Vahlne, J.-E. 1977. The Internationalization Process of the FirmA Model of Knowledge Development and Increasing Foreign Market Commitments. Journal of International Business, 8: 23–32.

Kogut, B. & Singh, H. 1988. The Effect of National Culture on the Choice of Entry Mode. Journal of International Business Studies, 19 (3): 411-432.

Li, PY, & Lo, FY. 2017. Top management teams’ managerial resources and international diversification The evidence under an uncertain environment. Management Decision, 55 (9): 1999-2017.

Nadolska, A. & Barkema, H.G. 2014. Good learners: how top management teams affect the success and frequency of acquisitions. Strategic Management Journal, 35 (10): 1483-1507.

Nielsen, B. B., & Nielsen, S. 2013. Top management team nationality diversity and firm performance: A multilevel study. Strategic Management Journal, 34 (3): 373-382. Nielsen, B.B. & Nielsen, S. 2011. The role of top management team international orientation

32

Nielsen, S. 2010. Top Management Team Internationalization and Firm Performance: The Mediating Role of Foreign Market Entry. Management International Review, 50 (2): 185-206.

Pisania, N., Muller, A., Bogăţana, P. 2018. Top Management Team Internationalization and Firm-level Internationalization: The Moderating Effects of Home-region Institutional Diversity and Firm Global Focus. Journal of International Management, 24: 239–256. Punnett, B.J. & Clemens, J. 1999. Cross-National Diversity: Implications for International

Expansion Decisions. Journal of World Business, 32 (2): 128-138.

Rivas, J.L. 2012. Diversity & internationalization: The case of boards and TMT’s. International Business Review, 21: 1–12.

Ronen, S. & Shenkar, O. 1985. Clustering Countries on Attitudinal Dimensions: A Review and Synthesis. Academy of Management Review, 10 (3): 435–454.

Salancink, G.R., & Pfeffer, J. 1977. Who gets power and how they hold on to it: A strategic contingency model of power. Organizational Dynamics, 5: 3–21.

Sullivan, D. 1994. Measuring the degree of internationalization of a firm. Journal of International Business Studies 25: 325–342.

Sutton, F., Seymour H., Carl, K., & Tobin, J. 1956. Value orientations and the relationship of managers and scientists. Administrative Science Quarterly, 10: 39-51.

Thams, Y. 2013. Foreign-born CEOs, Country-Specific Skills, Selection, and Strategic Consequences, FIU Electronic Theses and Dissertations. <http://digitalcommons.fiu.edu/etd/934> viewed 2 November 2018.

Tihanyi, L., Ellstrand, A.E., Daily, C.M. and Dalton, D.R. 2000. Composition of the top management team and firm international diversification, Journal of Management, 26 (6): 1157-1177.

Westphal, J. D., Zajac E.J. 1995. Who shall govern? CEO/board power, demographic similarity, and new director selection. Administrative Science Quarterly 40: 60-83. Zaheer, S. 1995. Overcoming the liability of foreignness. Academy of Management Journal,

33

34 APPENDIX A

Nine country clusters by Rhonen & Shenkar (1985).

35 APPENDIX B

Industry Code

Aerospace & Defence 1 Automobiles & Parts 2

Beverages 3

Chemicals 4

Clothing & Personal Products 5 Construction & Building Materials 6 Diversified Industrials 7

Electricity 8

Electronic & Electrical Equipment 9 Engineering & Machinery 10 Food & Drug Retailers 11 Food Producers & Processors 12

Forestry & Paper 13

General Retailers 14

Health 15

Household Products 16

Information Technology Hardware 17

Leisure & Hotels 18

Media & Entertainment 19

Oil & Gas 20

Pharmaceuticals and Biotechnology 21

Publishing 22

Renewable Energy 23

Software & Computer Services 24 Telecommunication Services 25

Tobacco 26

Transport 27

Table 3. Industry code table

36

Pairwise correlations

Variables -1 -2 -3 -4 -5 -6 -7 -8 -9 -10

(1) Int. Diversification 1.000

(2) Board national diversity -0.052 1.000

(3) CEO Nationality 0.076 0.393* 1.000 (4) CEO Power -0.053 -0.182* -0.253* 1.000 (5) M1 0.039 0.454* 0.959* -0.258* 1.000 (6) M2 -0.040 0.408* -0.064 0.672* -0.050 1.000 (7) logFsize 0.090 -0.106* -0.122* 0.030 -0.121* -0.017 1.000 (8) BoDsize 0.015 -0.113* -0.061 0.082 -0.077 0.018 0.264* 1.000

(9) Acquirer M&A experience 0.598* -0.111* 0.052 0.046 0.042 -0.027 0.080 -0.014 1.000

(10) Acquirer host M&A experience 0.559* -0.080 0.049 0.018 0.025 -0.021 0.112* -0.002 0.712* 1.000 * shows significance at the .05 level

APPENDIX C

Hausman test

Figure 4. Hausman test in STATA

Multicollinearity

Correlation Matrix

Table 4. Correlation Matrix

Prob>chi2 = 0.0109 = 21.42

chi2(9) = (b-B)'[(V_b-V_B)^(-1)](b-B) Test: Ho: difference in coefficients not systematic

B = inconsistent under Ha, efficient under Ho; obtained from xtreg b = consistent under Ho and Ha; obtained from xtreg CROSSexp .0268348 .0111076 .0157272 .0164729 TOTExp .0255232 .0067576 .0187657 .0075141 BoDsize .0033252 .0008942 .002431 .0029965 logFsize .0148725 .000948 .0139245 .0329874 M2_ .0274639 .0057489 .021715 .0129384 M1 .0371533 -.0424866 .0796399 .0345163 CEOPOWER -.0130999 -.0105759 -.0025239 .005238 CEONAT -.0143441 .0268067 -.0411509 .0201289 BLAU -.00647 -.0163136 .0098436 .0127205 fixed random Difference S.E.

(b) (B) (b-B) sqrt(diag(V_b-V_B)) Coefficients

37 Variance Inflation Factor (VIF)

Variable VIF 1/VIF

M1 10.89 0.091851

CEO Nationality 9.77 0.102321 Board national diversity 2.76 0.361754

M2 2.44 0.409083

CEO Power 2.31 0.433300

Acquirer host M&A experience 2.05 0.487412 Acquirer M&A experience 2.05 0.487896

log Firm size 1.09 0.917130

Board size 1.08 0.926714

Mean VIF 3.83

Table 5. Variance Inflation Factor test in STATA

Heteroscedasticity

Breusch-Pagan / Cook-Weisberg test for heteroskedasticity

H0: Constant variance

Variables: fitted values of MA chi2(1) = 49.72

Prob > chi2 = 0.0000