Does a firm’s strategy match its managers?

(1)

Master Thesis

Does a firm’s strategy match its managers?

An examination of the concept TMT internationalization and its

influence on firm internationalization

Submitted by

Christina Stefanie Vogt

B7059965, S3599167 c.s.vogt@students.rug.nl

MSc Advanced International Business & Management (Dual Award)

Supervision: Universities: Submission Date:

(2)

Abstract

Understanding the factors underlying firm internationalization has been a key endeavor in international business research over the past decades. Following an upper echelons perspective, this study focuses on the concept of top management team (TMT) internationalization as a central yet understudied driver of firm internationalization. By distinguishing between the two main TMT internationalization dimensions, namely international work experience and nationality diversity, it is examined whether they influence firm internationalization in disparate ways due to their unique impact on international strategic decision-making. Further, this study explores whether these influences change over time with continuous social interaction between TMT members. The results indicate that both TMT international work experience and TMT nationality diversity promote a firm’s level of international diversification. The moderating variable overlapping team tenure, however, appears to have no significant effect on these relationships. For this research, a new database is compiled, comprising complete information on the international careers, nationalities and tenure times of 744 top executives in 100 European multinational companies. The findings of this study contribute to literature by emphasizing the uniqueness of TMT international work experience and TMT nationality diversity for international strategic decision-making. Furthermore, this study entails insightful managerial implications for executive recruitment and development.

Keywords: TMT internationalization, Nationality diversity, International work experience,

(3)

Acknowledgments

(4)

In the past decades, the accelerating globalization has drawn scholarly attention towards multinational enterprises (MNEs) and their foreign expansion efforts. Firms are increasingly competing on a global scale to take advantage of business opportunities abroad and to secure consistent growth (Tihanyi et al., 2000). In line with this development, it has become a key endeavor in international business research to understand the drivers of firm internationalization. A large body of research has examined a variety of factors that are likely to affect firm multinationality (see Kirca et al. (2012) for an overview). Within this literature, one stream of research has highlighted the importance of executives’ characteristics as a major determinant of firm internationalization. By drawing from Hambrick and Mason's (1984) upper echelons theory, these studies suggest that the composition of the top management team (TMT) creates the foundation for managerial decisions and eventually firm behavior. However, as noted by Kaczmarek and Ruigrok (2013), the academic contributions concentrated primarily on the same set of TMT characteristics, including TMT size, age, tenure, and educational background (Carpenter et al., 2001; Herrmann & Datta, 2005; Rivas, 2012; Sambharya, 1996; Tihanyi et al., 2000).

(8)

as one dimension of TMT internationalization is comparably well researched by now (Athanassiou & Nigh, 2002; Sambharya, 1996; Tihanyi et al., 2000), only few studies have addressed the link between TMT nationality diversity and firm internationalization (Caliguri et al., 2004; Greve et al., 2009, 2015; Heijltjes et al., 2003). Most interestingly, however, little is known about how these two sources of TMT internationalization relate and compare to each other. Nielsen and Nielsen (2011) were among the first to give the impetus to compare and contrast both TMT internationalization dimensions. Their findings suggest that differently composed TMTs in terms of nationality and international work experience diversity show unique preferences and strategic choices in the context of foreign entry mode. However, as the entry mode choice is only one of the many decisions that have to be made in the process of firm internationalization (Beugelsdijk et al., 2018), scholars have to be cautious when drawing generalized conclusions about the two dimensions. Thus, as one important academic contribution, scholars could follow Nielsen and Nielsen’s (2011) example and further explore whether both dimensions lead to unique strategic preferences in other internationalization decisions as well (e.g., location choice, degree of ownership). This study, however, intends to go one step back and first get an impression of the big picture, namely the overall level of firm internationalization. Hence, the following research question is addressed:

Do TMT international work experience and TMT nationality diversity have a distinct influence on the overall level of firm internationalization?

(9)

information-processing capabilities and decision-making effectiveness of the TMT, which have been found to be critical determinants of the overall success of a firm’s internationalization efforts (Vermeulen & Berkema, 2002).

Furthermore, this study wants to contribute to TMT internationalization research by exploring how the influence of TMT international work experience and TMT nationality diversity on firm internationalization changes as TMT members continue to interact over time. Based on findings in prior research, the cognitive and social consequences of intra-group diversity may outweigh each other with increased social interaction (Barkema & Shvyrkov, 2007; Carroll & Harrison, 1998). Hence, while building on the first research question, this paper addresses the following sub-question:

Does continuous social interaction between TMT members change the influence of TMT international work experience and TMT nationality diversity on firm internationalization?

(10)

2. Literature Review

2.1 Firm Internationalization and the Role of the TMT

Over the past decades, firm internationalization has attracted considerable attention in international business research. The successful expansion of business activities across national borders is one of the major challenges MNEs encounter during their growth phase (Tihanyi et al., 2000). Scholars have long theorized about the potential drivers behind firm internationalization, but also the obstacles related to the process of foreign expansion (Kirca et al., 2012; Vermeulen & Barkema, 2002). When doing business abroad, MNEs face significant challenges and risks that go beyond different consumer tastes, new competitors and higher monitoring costs. The vast majority of complications and obstacles arise through the unfamiliarity with the institutional, cultural, and political environment, also known as liability of foreignness (LoF) (Zaheer, 1995). On the one hand, the foreign firm is perceived as an outsider in the local setting and thus likely to be excluded from domestic networks. However, foreign firms also face a lack of location-specific knowledge, hampering the communication and collaboration with local shareholders, such as customers, suppliers and regulators (Eden & Miller, 2004).

(11)

decision-making unit of MNEs, the top management team (TMT) (Beugelsdijk et al., 2018; Greve et al., 2009; Tihanyi et al., 2000). The foreign expansion process is considered a highly complex strategic decision process, characterized by incomplete information and insecurity regarding the actual outcomes (Nielsen & Nielsen, 2011). Furthermore, in an internationalization strategy, multiple critical, interconnected decisions have to be considered, such as choices regarding the location, form of entry and control structure (Beugelsdijk et al., 2018). Thus, the TMT is exposed to large, diverse sets of information flows, challenging the cognitive capabilities and information-processing capacities of its members (Carpenter et al., 2001; Tihanyi & Thomas, 2005). Furthermore, since internationalization is an integral part of the firm’s overall strategic course, the decisions most likely require the consensus of most or all TMT members (Barkema & Shvyrkov, 2007).

(12)

Mason's (1984) upper echelons theory. At its core, the theory implies that the decision-making process of top executives is greatly influenced by their values, beliefs and experiences. Especially under extreme conditions, characterized by high managerial complexity and uncertainty as in firm internationalization, the background and experience of these top executives is likely to have an essential impact on their strategic choices (Hambrick & Mason, 1984). Fundamental values and cognitions are used as lenses to scan, identify and analyze information, which, in turn, build the foundation for decision making (Shaw, 1990).

2.2 TMT Internationalization and Firm Internationalization

Although the idea of “matching managers and strategies” (Gupta & Govindarajan, 1984; Szilagyi & Schweiger, 1984) is well known in upper echelons research, scholars have only recently started to introduce and more thoroughly explore the concept of TMT

internationalization and its potential relationship with firm internationalization. By doing

so, scholars used a number of different terms, such as TMT internationalization (Nielsen, 2010; Pisani et al., 2018), TMT international orientation (Azam et al., 2018; Nielsen & Nielsen, 2011), TMT international profile (Greve et al., 2015), and TMT international exposure (Lee & Park, 2008).

(13)

(14)

effect of TMT nationality diversity on the abovementioned processes have remained largely unexplored in research (Nielsen & Nielsen, 2011). Therefore, the theoretical model developed in the following section is largely based on insights of international management, team dynamics and managerial cognition literature.

2.2.1 TMT International Work Experience and Firm Internationalization

(15)

knowledge and general cognitive competencies than working in fewer countries, enriching the individual’s cognitive base by more diverse knowledge. The person’s overall tolerance for ambiguity, multiple worldviews, and flexibility is enhanced (Caligiuri & Tarique, 2012). In this context, literature also talks about the development of a global mindset, which is reflected in a higher cultural self-awareness, openness to other cultures and a thorough understanding of foreign values and practices (Levy et al., 2007; Tung & Miller, 1990).

(16)

literature suggests that executives’ international work experience is likely to play an important role for TMTs engaging in firm internationalization decisions (Greve et al., 2009; Maitland & Sammartino, 2015). Maitland and Sammartino (2015) for instance found that the TMT’s cognitive base and decision-making in regard to the firm’s foreign expansion is enriched by a larger number of different perspectives and interpretations due to the individual TMT member’s breadth, depth and diversity of prior international work experience. The authors conclude that this is likely to substantiate the rationality and confidence behind the firm internationalization strategy (Maitland & Sammartino, 2015). By being more aware of and open for international opportunities, internationally experienced managers in the TMT may also have a better feeling for promising business opportunities abroad. Furthermore, prior international work experience helps executives and consequently the TMT as a whole to better deal with the uncertainty and ambiguity of foreign expansion (Sambharya, 1996). It is likely to improve their understanding and management of unfamiliar cultural and institutional environments, customer needs and business practices (Herrmann & Datta, 2006; Tihanyi et al., 2000).

(17)

behind the firm internationalization strategy, eventually resulting in higher levels of firm internationalization.

H1: TMT international work experience will be positively related to firm internationalization.

2.2.2 TMT Nationality Diversity and Firm Internationalization

Based on findings in cross-cultural psychology research, nationality strongly influences an individual’s underlying values and personality (Hofstede, 1980; Triandis & Suh, 2002). These fundamental values and cognitions are used as lenses to scan, identify and analyze information in the environment. Consequently, nationality is likely to affect an individual’s behavior and represents the foundation for decision-making (Shaw, 1990). Over the past decades, scholars aimed to explore the impact of nationality diversity at the team level, in particular how diversity in team members’ nationality may influence team dynamics and group performance (Hambrick et al., 1998). It is, however, still rather unclear how intra-group nationality diversity impacts firms’ strategic decision-making processes in regard to foreign expansion (Nielsen & Nielsen, 2011). Despite this lack of knowledge, some recent studies have observed an overall increase in the number of foreigners in top management positions in MNEs over the past decade (Ruigrok & Greve, 2008; Staples, 2007; Van Veen & Marsman, 2008).

(18)

collective cognitive base of the TMT (Nielsen & Nielsen, 2011) and reduce, for example, group think (Nederveen et al., 2013). Moreover, Luo (2005) suggests that foreign executives show a unique kind of internationally deployable expertise, that goes beyond specific knowledge of their home country market. Most importantly, scholars also argue that foreign managers tend to have superior information-processing capabilities related to foreign markets (Greve et al., 2009). Thus, in the process of foreign expansion, firms may be able to meet the more complex information-processing requirements with the appointment of foreign executives.

(19)

the subgroups, the TMT may not be able to take full advantage of the potentially valuable cognitive resources and strong information-processing capabilities of its individual members. As a consequence, it is less likely that novel ideas and strategic options are proposed and appreciated by the entire TMT (Finkelstein and Hambrick, 1996). Furthermore, research specifically focusing on intra-group communication found that communication in nationally diverse groups tends to be slower, more difficult and more confused. This is likely to severely hinder the decision-making process, contributing little to the task itself (Frijns et al., 2016; Miller et al., 1998, Thatcher et al., 2003).

Another stream of research within group dynamic literature highlights the higher likelihood of relationship or affective conflict in nationally diverse teams (Frijns et al., 2016; Nederveen et al., 2013; Pelled et al., 1999). Relationship and affective conflict may involve contradictory views within a team in regard to differences in norms and values, but also personality differences. It has been concluded that these types of conflict have a detrimental effect on group commitment and group decision quality (De Wit et al., 2012). Scholars also focused on the negative impact of nationality diversity on intragroup trust. Bjørnskov (2008), for instance, shows that there is less trust in more heterogeneous groups, most likely due to the differences in world views, ideas and opinions. In addition, this can further deepen relationship conflict, as argued by Simons and Peterson (2000).

(20)

firm’s internationalization process most likely requires the full consensus and commitment of all TMT members (Hambrick et al., 1996), it is expected that TMT nationality diversity will therefore result in less firm internationalization.

H2: TMT nationality diversity will be negatively related to firm internationalization.

2.3 The Moderating Role of Overlapping Team Tenure

Over the past decades, research aimed to explore how the effects of team characteristics on group outcomes change with time (Carroll & Harrison, 1998). As demonstrated on the example of TMT nationality diversity and TMT international work experience, diversity can lead to both positive (cognitive) as well as negative (affective) consequences (Milliken & Martins, 1996). It is crucial to understand under what conditions these consequences may change, in particular when, or if, the benefits can outweigh the costs of diversity. In order to see whether the time a group has spent together as a team has an influence on these dynamics, a number of scholars included team members’ tenure time in their research (Kirca et al., 2012; Nielsen & Nielsen, 2013).

2.3.1 TMT International Work Experience and Overlapping Team Tenure

(21)

process. Research however found that, over time, continued interaction between the same team members is likely to reduce the cognitive diversity in TMTs (Barkema & Shvyrkov, 2007). Positive outcomes of cognitive diversity, such as constructive debate and strategic innovation, diminish the longer a group works together, since the flow of new ideas and diversity of preferences decreases (Finkelstein & Hambrick, 1996; Katz, 1982). Over time, the members of the TMT increasingly resemble one another in their knowledge and views. Consequently, the group’s knowledge base, which is the source of the search for alternatives, becomes smaller and less diverse (Finkelstein & Hambrick, 1996). Furthermore, TMT members may get better in resolving or avoiding conflict on the long run. This will, however, decrease the chance to learn from the knowledge asymmetries, which are the foundation of productive task conflict (Gruenfeld et al., 1996). This tendency towards consensus building is likely to suppress any form of creativity and novel initiatives, eventually leading to groupthink, rigidity, and strategic inertia (Katz, 1982).

Taken as a whole, it is expected that steady interaction of TMT members over a longer period of time, during which their tenure overlaps increases, reduces the positive, cognitive impact of TMT international work experience. Consequently, the positive relationship between TMT international work experience and firm-level internationalization will be weakened.

(22)

2.3.2 TMT nationality diversity and overlapping team tenure

In the case of TMT nationality diversity, this study expects that the negative consequences of nationality diversity, such as social categorization (Byrne, 1971) and behavioral disintegration (Li & Hambrick, 2005), will outweigh any positive cognitive effects, eventually resulting in less firm internationalization. Over time, however, as the members of a team become more familiar with each other, social categorization and stereotyping based on demographic attributes such as national origin may become less salient (Barkema & Shvyrkov, 2007; Chatman & Flynn, 2001). Hambrick et al. (1998) found that recently formed TMTs tend to be most vulnerable to the negative consequences of nationality diversity. After a while, team members increasingly accept and potentially even respect differences in behavior, values and beliefs. The boundaries of ‘in’ and ‘out’ fade and demographically diverse team members start to see each other as fellow in-group members. Hence, continuous interaction is likely to diminish relationship and affective conflict within teams as well (Hambrick et al., 1998). Barriers to intra-group communication and cooperation are likely to slowly disappear and the exchange of ideas and strategic innovation is fostered (Chatman & Flynn, 2001). According to Eisenhardt and Schoonhoven (1990), common team tenure positively influences a team’s ability to concentrate on issue-related differences without getting distracted by personal disagreements.

(23)

more likely to be reached. Hence, this paper proposes that overlapping team tenure is able to diminish the negative effects of TMT nationality diversity on firm internationalization.

H3b: Overlapping team tenure positively moderates the relationship between TMT nationality diversity and firm internationalization.

2.4 Conceptual Model

The following conceptual model (Figure 1) summarizes and visualizes the developed hypotheses:

(24)

3. Methodology

This section provides a detailed overview of the research methodology of this paper. In a first step, the data collection process is outlined. Then, the sample is described and the variables for both the main analysis and the robustness tests are introduced. Lastly, the empirical data analysis is explained and the preliminary requirements for linear regression are tested.

3.1 Data Collection

This paper applies a quantitative approach to examine the relationship between the variables. Based on the Research Onion model, developed by Saunders et al. (2007), the study follows the philosophy of positivism. By using a deductive approach, the proposed hypotheses are built upon existing theory. The underlying concepts are converted into variables, after which the data collection can be carried (Jonker & Pennink, 2010). The study is relying on secondary data as a source since archival data is utilized to test the proposed hypotheses.

(25)

ratio of foreign sales to total sales. Moreover, one requirement of using financial measures is that the sampled firms employ the same accounting policies, which cannot simply be assumed in a study covering multiple countries and industries without further investigation. ORBIS was also the main information source for the two control variables Firm Size and

Industry. The database provides detailed information on the latest total number of

employees worldwide, as well as the industry the firm is operating in, based on the Standard Industrial Classification (SIC) codes.

(26)

database has shown, that the data on executives’ international experience is largely outdated and incomplete. Hence, BoardEx is only used to complement the newly compiled database. Only those companies for which complete information on nationalities, international career backgrounds and tenure times was available for all TMT members were included in the analysis. This requirement, as well as the overall very time-consuming and complex data collection process, is one justification why it was not possible to base this research on a larger sample size.

This study theorizes that both independent variables TMT Nationality Diversity and

TMT International Work Experience precede Firm Internationalization. However, it

(27)

3.2 Sample

(28)

Following the example of Greve et al. (2015), TMT is used as a generic term for the highest level of management in the companies, while relying on the firms’ self-reported TMT definitions. In cases where two or more levels of top executives were reported, the Chief Executive Officer (CEO) and the immediate subordinates were defined as the core TMT. Eventually, this study compiled a new database comprising complete information on the international careers, nationalities and tenure times of 744 TMT members in 100 MNEs and these firms’ level of internationalization at year-end 2017. The individual-level TMT member data was used to compute the team-level variables required for the analysis.

3.3 Variable Measurement

3.3.1 Dependent Variable

(29)

3.3.2 Independent Variables

TMT International Work Experience

The independent variable TMT International Work Experience is operationalized as the average number of foreign countries the TMT has experienced, based on the individual experiences of its members. By following this approach, this study attempts to develop and apply a more sophisticated measure of the overall level of international work experience in the TMT. Instead of simply calculating the proportion of internationally experienced members in the TMT (e.g., Carpenter et al., 2001; Hermann & Datta, 2005; Nielsen & Nielsen, 2011), this measure also takes the diversity of international experience of the TMT into account (Azam et al., 2018). Although this operationalization is more time-consuming and difficult to implement, it is hoped that it will result in a more refined picture of the level of TMT international work experience. With this measurement approach, this study follows the example of a stream of researchers (e.g., Le & Kroll, 2017; Maitland & Sammartino, 2015) who also included the number of countries of international work experience in their studies. For each executive, it was analyzed in how many different foreign countries he or she has worked in the course of the career. In a second step, the mean number of foreign countries per TMT was calculated. Taking the mean over the median is in this case an appropriate way to operationalize the variable, because no extreme outliers that may influence the results were detected during the data collection and analysis.

TMT Nationality Diversity

(30)

prior research has applied a number of methods to calculate nationality diversity within groups, Blau’s index of diversity (Blau, 1977) is described as one of the more sophisticated options (Van Veen & Marsman, 2008). Hence, in this paper TMT Nationality Diversity is measured with Blau’s index to ensure that a comprehensive picture of the nationality diversity within the TMTs is created. Over the years, Blau’s index has become a commonly used measure of group heterogeneity in TMT research (Finkelstein & Hambrick, 1996). One main reason for its popularity is the fact that the index specifically captures the dispersion of team members across all possible categories of a certain dimension (Greve et al., 2009; Nielsen & Nielsen, 2011). In regard to this study that means that the number of

different nationalities and their frequency within the TMT is also considered. Blau’s index

is calculated as follows:

(31)

3.3.3 Moderating Variable

The moderating variable Overlapping Team Tenure is based on a measure which is so far rather less well-known in TMT research. Traditionally, studies have included TMT members’ median or mean tenure time to capture the interaction between team members over a period of time (Kirca et al., 2012; Nielsen & Nielsen, 2013; Wiersema & Bantel, 1992). However, whether these measures represent the most precise operationalizations for social interaction may be at least called into question here, since they do not include a comparison of the overlapping tenure times of the individual team members. On the contrary, the measure of common historical experience (TLAP) intends to capture this interaction among team members by specifically focusing on the pair-wise overlap in tenure times for all possible pairs in the TMT (Harris et al., 2012). TLAP was originally proposed by Carroll and Harrison (1998), and subsequently used by a small number of researchers (Barkema & Shvyrkov, 2007; Buyl et al., 2011). Especially in sport research, TLAP is more commonly used to examine changing team dynamics over time (Harris et al., 2012).

In a first step, data on each executive’s TMT tenure time was collected. Then, the tenure times of every possible pair of individuals in the TMT were compared and the length of the overlap was calculated. Lastly, the overlap times for all pairs were averaged. The formula for TLAP looks as follows:

(32)

TLAP is used to capture the diminishing effects of interactions between the members of a team over time (Gruenfeld et al., 1996).

3.3.4 Control Variables

Due to the small sample size, only two control variables are included in this study to avoid overfitting the model. It is acknowledged that larger studies observing TMT characteristics and firm internationalization typically control for a number of firm-, industry- and country level factors (see for example Rivas, 2012; Tihanyi et al., 2000). Scholars for instance argue that it is imperative to include TMT size as a control variable when examining the effects of TMT heterogeneity, because heterogeneity is likely to be positively associated with the size of the studied group (Allison, 1978; Carpenter et al., 2014). In this study, however, the size of the TMT is already incorporated in the measurement of the independent variables, which makes any further inclusion redundant.

Firm Size

(33)

nationality diversity (Heijltjes et al., 2003). Taken together, these findings stress the importance of controlling the model for firm size. Following existing literature, the total number of employees is used as a proxy for firm size and transformed with a logarithmic function to account for skewness in the data (Greve et al., 2009; Nielsen & Nielsen, 2008; Pisani et al., 2018).

Industry

In their study, Grøgaard et al. (2013) conclude that industry characteristics may play a key role when explaining different levels of internationalization among firms. The authors find that firms operating in the service industry tend to show lower internationalization levels than manufacturing firms. Pisani et al. (2018) also claim that firms from different industries may have specific reasons to expand internationally which might in turn substantially affect their level of diversification. Hence, most researchers control their models with industry dummies (e.g., Nielsen, 2010; Pisani et al., 2018; Rivas, 2012). However, the sample size of this paper is too small to introduce several industry categories. Following the approach of Ali et al. (2014), this study is therefore using a dummy variable to account at least for the two major industry categories: The first category is

Manufacturing (coded as 1) and the second one is Services (coded as 0) (Appendix A).

3.3.5 Robustness Tests

(34)

equivalence of the variables, determined by their correlation. Hence, only in case of a high correlation between the main and modified variable, the robustness check is valid (Acock, 2014). The correlation matrix (Appendix B) proves that this requirement is fulfilled for the subsequently mentioned variable modifications.

Modification of the Independent Variables

In TMT diversity research, some scholars (e.g., Pisani et al., 2018) first measure nationality diversity with a simple ratio of foreign executives and subsequently repeat the analysis with Blau’s diversity index (Blau, 1977) to check the robustness of their findings. Following this example, a second, modified variable representing nationality diversity is introduced: The proportion of foreign executives in the TMT. “Foreign executive” is coded as a dummy that is equal to one if the executive’s primary nationality differs from the firm’s home country, and zero otherwise. A ratio is then calculated by dividing the number of foreign executives by the total number of executives in the TMT (e.g., Piaskowska & Trojanowski, 2014; Pisani et al., 2018).

The same measurement approach is applied for the modification of TMT International

Work Experience. Instead of including the number of countries of foreign work experience,

(35)

Modification of the Moderating Variable

To ensure the robustness of the findings of the moderation analysis, the regression will be repeated with a modified version of Overlapping Team Tenure. Instead of using the measure of common historical experience (TLAP), the mean TMT tenure time is chosen for the robustness check. For each firm, the TMT tenure times of the individual executives are collected, summed up and averaged over all TMT members in the firm. By conducting this robustness test, it is also acknowledged that TLAP is a so far rarely established measure in TMT research. Thus, not much is known about the measure’s validity. By cross-checking the results with a more commonly used operationalization, such as the mean tenure time, this study may be able to contribute to the further understanding and reliability of TLAP.

3.4 Empirical Data Analysis

(36)

linear regression have to be tested and met to ensure the reliability of the regression results (Kohler & Kreuter, 2012).

3.4.1 Assumptions of Multiple Regression Analysis

In a first step, the data is tested for potential outliers. These are observations with extreme values that can heavily influence the results of a regression model if not detected and adjusted when necessary (Acock, 2014). As suggested by Kohler and Kreuter (2012), in the case of a model with multiple independent variables, Cook’s D should be applied. The test statistic creates a new variable in the dataset with the values of Cook’s D. If the values are greater than 1.00, the presence of an outlier can be assumed (Kohler & Kreuter, 2012). However, all values score clearly below 1.00, hence the dataset seems to contain no outliers.

Linearity between the predictor and outcome variables has been observed by creating

scatterplots. The key requirement for linear regression is that the dependent variable can be described as a linear function of the independent variables. As suggested by Kohler and Kreuter (2012), scatterplots are especially useful when the sample size is rather small as in the case of this study. There is no violation of the linearity assumption in this dataset.

Multicollinearity between the independent variables can be a problem in multiple

(37)

al. (1998) argue that one has to be already cautious with a VIF of 5. All values score below 5, hence, multicollinearity seems to be no issue in the model (Appendix C, Section 1).

Another important assumption of regression is the normality of the residuals (Acock, 2014). This assumption is checked both visually with a Kernel density estimate (with superimposed normal curve) and empirically with the Shapiro-Wilk test. The test statistic depicts a p-value of 0.21634 and the Kernel density estimates shows a rather normal curve (Appendix C, Section 2). Hence, it can be concluded that the residuals follow a normal distribution.

The assumption independence of errors is evaluated with the Durbin-Watson test statistic, which shows whether an autocorrelation exists between the error terms (Kohler & Kreuter, 2012). The result of the test statistic is in the accepted range between 1.5 and 2.5 (Field, 2009), as it can be seen in the appendix (Appendix C, Section 3). Hence, the assumption of independence of errors is met.

(38)

4. Results

After having tested the preliminary requirements of linear regression, the main analysis can be carried out. In a first step, the descriptive statistics and the correlation matrix will be analyzed. Afterwards, the results of the regression model will be described.

4.1 Descriptive Statistics

In Table 1, the descriptive statistics of the dataset are presented. The 100 sampled firms originate from 12 European countries. Most of them are headquartered in Germany (36), the United Kingdom (22), France (20), and the Netherlands (9) (Appendix D). 51 firms are operating in the service industry and 49 firms in manufacturing (Appendix A).

Variable Mean Min Max SD

Proportion Foreign Subsidiaries 0.6810166 0.0344828 0.9820896 0.2201687

Number Foreign Countries 49.66 1 150 30.91867

TMT Nationality Diversity 0.4079827 0 0.8148 0.2489274

TMT Int. Work Experience 1.438725 0 3.75 0.8042578

Overlapping Team Tenure 49.48318 22.02778 116.2857 19.34654

Firm Size 118746 1316 642300 119918.2

Table 1: Descriptive statistics; Variables are presented in original form and not mean-centered or log-transformed

(39)

the mean value of 68.10% is rather high. On average, about two third of the firms’ subsidiaries are located outside their home country. A similar picture can be drawn from the second firm internationalization variable. The number of foreign countries the firm is operating in ranges from 1 to 150, with a mean of 49.66 and a standard deviation of 30.92. Thus, on average, the firms are present in 50 countries. Taken as a whole, it can be concluded that the firms in this sample vary strongly in regard to their internationalization level. However, on average, the overall level tends to be rather high, which is of course also not unexpected since the sample included the 100 largest European MNEs.

The next two variables in Table 1 are the independent variables. The values of Blau’s index for calculating TMT Nationality Diversity are ranging from 0 to 0.8148. Hence, the sample includes both totally homogenous TMTs in which all members share the same nationality, but also very heterogenous teams. 16 out of 100 firms do not have any foreigner at all on their TMT. Interestingly, eight out of these 16 firms are based in Germany, and another five in the United Kingdom. TMT Nationality Diversity has a mean of 0.41with a standard deviation of 0.25. Thus, it can be concluded that the TMTs in this sample are fairly diverse in terms of nationality diversity.

(40)

Overlapping Team Tenure is measured in months. On average, the executives in the

TMT have been working together for 49.48 months (SD=19.35), ranging from 22.02 months up until an extreme value of 116.29 months. The average Firm Size in the sample is 118746 employees (SD=119918). Again, this high value is no surprise since the sample consists of large MNEs.

4.2 Correlations

Table 2 contains the pairwise correlations between the variables. The asterisks indicate whether the correlation is significant at the 0.01, 0.05, or 0.1 level.

Table 2: Correlation Matrix; ***p≤0.01, **p≤0.05, *p≤0.1

As it can be seen in the correlation matrix, the independent and control variables correlate significantly with the outcome variable Firm Internationalization. Furthermore, the two independent variables TMT Nationality Diversity and TMT International Work

Experience appear to significantly correlate with one another (0.5082***). Hence, in order

Firm Internat. TMT Nationality Diversity TMT Int. Work Experience Overlapping Team Tenure

(41)

separately and simultaneously in the following regression analysis, as suggested by Acock (2014). Particularly notable is the fact that the moderating variable Overlapping Team

Tenure shows an insignificant effect on all variables. This already indicates a potentially

non-existing moderation effect in the main regression model. Firm Size seems to significantly correlate with Firm Internationalization, but not with the two independent variables, whereas Industry correlates with all three of them.

4.3 Regression Results

4.3.1 Main Regression Model

The results of the hierarchical regression analysis are presented in Table 3. The overall analysis consists of six models. First, the effect of the two control variables on Firm

Internationalization is tested in Model 1. As it can be seen in Table 3, Firm Size (B=0.555,

p=0.001) and Industry (B=0.949, p=0.005) have significant explanatory power on Firm

Internationalization. Computing the standardized estimated regression coefficients (also

called beta coefficients) would further allow us to compare the effect sizes of the control variables, despite their different measuring units. However, since we cannot compute beta coefficients for binary variables, in this case our industry dummy, a direct comparison is not possible (Kohler & Kreuter, 2012). The beta coefficient of Firm Size indicates that a one-standard-deviation increase in Firm Size leads to an increase in Firm

Internationalization of about 0.307 standard deviation. In regard to the interpretation of the

dummy variable Industry, it can be concluded that MNEs in the manufacturing industry are on average 0.949 units more internationalized than MNEs operating in the service

(42)

Table 3: Model Summary Regression Analysis

Firm Internationalization

Model 1 Model 2 Model 3 Model 4 Model 5 Model 6

Independent Variables

TMT Int. Work Experience 0.732 (0.001)*** 0.572 (0.017)** 0.716 (0.002)***

TMT Nationality Diversity 1.891 (0.005)*** 1.082 (0.140) 1.895 (0.006)***

Interaction

TMT Int. Work Experience X

Overlapping Team Tenure 0.306 (0.599)

TMT Nationality Diversity X

Overlapping Team Tenure -0.162 (0.921)

Control Variables Firm Size 0.555 (0.001)*** 0.519 (0.001)*** 0.549 (0.001)*** 0.524 (0.001)*** 0.518 (0.002)*** 0.550 (0.001)*** Industry 0.949 (0.005)*** 0.505 (0.137) 0.700 (0.036)** 0.460 (0.174) 0.514 (0.134) 0.701 (0.038)** F-stat. 9.83 11.32 9.74 9.15 6.73 5.72 Prob > F 0.0001*** 0.0000*** 0.0000*** 0.0000*** 0.0000*** 0.0001*** R Square 0.1685 0.2614 0.2333 0.2781 0.2637 0.2334 Adjust. R Square 0.1514 0.2383 0.2093 0.2478 0.2245 0.1926

R Square Change to Model 1 0.093 (0.001)*** 0.065 (0.005)*** 0.110 (0.001)***

Note: N=100 MNEs; p-values in parentheses; ***p≤0.01, **p≤0.05, *p≤0.1 Standardized estimated regression coefficient (beta coefficients)

Model 1: Firm Size (beta=0.307)

(43)

explained by the regression model (Acock, 2014). In this case, the regression model with the two control variables explains 16.85% of the variance in Firm Internationalization, which could be considered as moderate according to Acock (2014).

In model 2, the independent variable TMT International Work Experience is separately added to the two control variables to test Hypothesis 1. A highly significant, positive effect (B=0.732, p=0.001) can be detected, which allows us to accept Hypothesis 1. Thus, higher levels of international work experience within TMTs results in more firm

internationalization. Furthermore, R2 _{increases by 0.093 in comparison to Model 1,}

reaching a value of 26.14%. One has to put this improvement in perspective, however,

since R2_{tends to go up if variables are added to the model (Kohler & Kreuter, 2012). Acock}

(2014) also notes that R2_{may end up being big just by chance when having a small sample}

with many predictors. In order to offset this bias, the adjusted R2 _{is also considered in this}

study. Since the adjusted R2 _{in model 2 increases by more than 0.08 points as well, it can}

be concluded that TMT International Work Experience indeed strengthens the explanatory power of the regression model.

In Model 3, Hypothesis 2 is tested by adding TMT Nationality Diversity to the two control variables. TMT Nationality Diversity shows a significant positive effect on Firm

Internationalization (B=1.891, p=0.005) and the adjusted R2 _{increases by 0.0579 points in}

comparison to Model 1. It can be concluded that 23.33% of the variance in Firm

Internationalization is explained by using this set of predictors. Following the rule of

(44)

shows a significant positive effect on Firm Internationalization. Hence, an increase in nationality diversity within the TMT results in higher levels of firm internationalization, holding all other variables constant.

In Model 4, the two independent variables are added simultaneously into one regression model. TMT International Work Experience shows a significant, positive effect (B=0.572, p=0.017) on Firm Internationalization, which confirms the findings of Model 2 and allows us to fully accept Hypothesis 1. TMT Nationality Diversity on the contrary still depicts a positive, but now insignificant effect on the outcome variable (B=1.082, p=0.140). The reason behind the changes in the significance and effect size may be grounded in the high collinearity between the two independent variables TMT International Work Experience and TMT Nationality Diversity (0.5082), as illustrated in Table 2. There is a high likelihood that the strong linear relationship between both variables has influenced the regression results (Kohler & Kreuter, 2012). Thus, Model 2 and 3 are the more appropriate and meaningful choice when testing the hypotheses. As a concluding remark for this analysis, it has to be pointed out that all models are highly significant (Prob > F is smaller than p < 0.001).

In regard to the implications of this study, we further want to know which of the two predicting variable has the stronger effect on Firm Internationalization. When comparing the standardized estimated regression coefficients of both independent variables, one can see that the beta coefficient for TMT International Work Experience (Beta=0.2579) is higher than for TMT Nationality Diversity (Beta=0.1511) (Table 3). Acock (2014) also

suggests observing the changes in R2_{when entering specific variables last as another}

(45)

increases R2_{by 0.093 points, whereas the change in R}2_{after adding TMT Nationality} Diversity is with 0.065 points slightly smaller. Hence, both approaches suggest that TMT international work experience seems to have a stronger impact on Firm Internationalization than TMT Nationality Diversity.

4.3.2 Moderation Analysis

Fig. 2a: Interaction Effect on Int. Work Experience Fig. 2b: Interaction Effect on Nationality Diversity

Before empirically testing Overlapping Team Tenure as a moderating variable, the interaction effect is visualized in Figure 2a and 2b to get a first impression on the overall direction of the interaction and its influence on subgroups within the TMT. By simply comparing the slopes of the lines in Figure 2a and 2b, one can already predict that the effect of Overlapping Team Tenure on Firm Internationalization seems to be stronger for TMT

International Work Experience (Figure 2a) than for TMT Nationality Diversity (Figure 2b).

In Figure 2a, the effect of Overlapping Team Tenure on the relationship between TMT

International Work Experience and Firm Internationalization is plotted. To get a more

thorough understanding of the interaction, the independent variable TMT International

Work Experience is split into three subgroups: Very internationally unexperienced TMTs,

(46)

medium experienced TMTs and highly internationally experienced TMTs. The interaction effect is almost 0 for TMTs with no international work experience at all. However, as the TMT gets more internationally experienced, the effect of Overlapping Team Tenure on

Firm Internationalization becomes more and more positive. Thus, a steady interaction of

team members over a longer period of time may in particular help TMTs with very high levels of international work experience to increase the firm’s level of internationalization. This graphical analysis, however, already indicates that Hypothesis 3a cannot be accepted, since an overall negative effect of Overlapping Team Tenure on the relationship between

TMT International Work Experience and Firm Internationalization has been predicted.

In Figure 2b, the effect of Overlapping Team Tenure on the relationship between TMT

Nationality Diversity and Firm Internationalization is visualized. Again, the independent

variable TMT Nationality Diversity is divided into subgroups to examine whether the effect of Overlapping Team Tenure differs among more or less nationally diverse TMTs. Three subgroups are created: Very nationally homogeneous TMTs, medium diverse TMTs and very nationally heterogeneous TMTs. The graph in Figure 2b illustrates that the effect of

Overlapping Team Tenure on Firm Internationalization varies only marginally among the

(47)

Finally, in Table 3 (Model 5 and 6), the results of the interaction regression analysis are presented. As already observed in the graphical analysis, Overlapping Team Tenure appears to have a positive moderating effect (B=0.306) on the relationship between TMT

International Work Experience and Firm Internationalization, but a negative effect

(B=-0.162) on the relationship between TMT Nationality Diversity and Firm

Internationalization. However, the p-values of both models are highly insignificant. With

a number of additional statistical tests, it has been analyzed whether the results may be significant for certain subgroups within the TMT. The overall findings, however, remain insignificant across all subgroups. Therefore, it has to be concluded that Hypothesis 3a and b need to be rejected.

4.3.3 Robustness Tests

Modification of the Independent Variables

To check the robustness of the findings, the main regression analysis is repeated with a modification of the independent variables (Appendix E). The robustness tests using the proportion of foreigners and internationally experienced executives in the TMT validate the overall findings of the main regression model, although the coefficients are slightly changed. The Proportion of Internationally Experienced Executives depicts a significant, positive effect (B=2.116, p=0.001), which supports Hypothesis 1. The effect of the

Proportion of Foreign Executives on Firm Internationalization is also significant and

(48)

insignificant results for the modified independent variables, thus Hypothesis 3a and 3b cannot be supported.

Modification of the Interaction Effect

The moderation analysis is in a subsequent step repeated with a modification of the interaction term (Appendix E). Instead of using the pairwise overlap of team tenures within the TMT, the mean TMT tenure time is included in the regression analysis. TMT mean tenure time shows no significant interaction effect on the relationship between TMT

International Work Experience and TMT Nationality Diversity on Firm Internationalization. Hence, a modification of the interaction variable does not change the

outcome of the main model, Hypothesis 3a and 3b still have to be rejected.

5. Conclusion

5.1 Discussion and Conclusion

(49)

drawn for the overall level of firm internationalization. By grounding the theoretical framework on insights from team dynamics and international management literature, this study proposes that TMT international work experience has a positive impact on the level of firm internationalization, whereas TMT nationality diversity shows the contrary effect. This study also aimed to further extend diversity research by exploring how social interaction may impact TMT diversity over time. It is expected that the influence of TMT international work experience and TMT nationality diversity on firm internationalization will change due to the continued interaction between the TMT members that are involved in the internationalization decision.

(50)

combination of both, what eventually leads to more firm internationalization can only be guessed at this point.

Contradictory to this study’s expectation, no support is found for a negative relationship between nationality diversity within TMTs and the level of firm internationalization. Based on findings in group dynamic literature, it was expected that the decision-making process in nationally diverse TMTs may be complicated, for instance due to the formation of subgroups and behavioral disintegration (Byrne, 1971; Li & Hambrick, 2005). However, the significant positive relationship between the variables suggests that a higher level of nationality diversity in TMTs actually leads to more firm internationalization. This result indicates that the cognitive benefits of intra-group nationality diversity outweigh any negative social implications. Hence, this study seems to confirm the argumentation of scholars that highlight the potential benefits of appointing foreign executives to TMTs, such as their increased information-processing capabilities and their unique knowledge about foreign markets (Greve et al., 2009; Luo, 2005; Nielsen & Nielsen, 2011). Again, the result holds true independent of which operationalization or measurement method is used. It is however worth mentioning that the explanatory powers of the main variable and its modified version slightly differ. TMT Nationality Diversity (main regression), captured with Blau’s diversity index, has a higher explanatory power than the Proportion of Foreign

Executives in the TMT (robustness test). By considering the number of different

(51)

foreigners in their top executive ranks, but also make sure that a diverse set of nationalities is represented.

Although this study does not find support for an opposing effect of TMT international work experience and TMT nationality diversity on firm internationalization, we can still derive some interesting insights which highlight the uniqueness of the dimensions, as

claimed by Nielsen and Nielsen (2011). As the beta values and R2_{indicate, TMT}

(52)

marketplace. On the contrary, foreign executives tend to only have a very specific, in-depth knowledge of one particular country (Greve et al., 2009). This might subsequently result in international work experienced TMTs choosing the target countries with less hesitation, potentially explaining the firm’s higher geographic dispersion. Hence, although TMT international work experience and TMT nationality diversity both have a positive influence on firm internationalization, the underlying process and strategic choices may be still very unique. In essence, however, this study seems to support the idea that firms match strategies and managers in the context of firm internationalization.

(53)

Hofstede, 1980; Levy et al., 2007). The sampled TMTs in this study, however, have on average an overlapping team tenure of only four years. Within this short period of time, it is not likely that an individual’s mindset and behavior changes so drastically, especially when it is strongly shaped by prior international experiences or nationality. A global mindset, for instance, cannot simply be erased within a few years (Levy et al., 2007). Similarly, a manager’s cultural identity is not likely to be lost over time, nor can it easily be overshadowed by other experiences (Nielsen & Nielsen, 2011). In order to confirm this possible explanation, research should apply a longitudinal study design. To further explore the impact of social interaction on team diversity, scholars could also choose TMT characteristics that are less likely to have such a strong, long-lasting effect on an individual’s mindset and behavior, for instance a person’s educational background.

Lastly, the control variables do not show any unexpected effects, the findings are largely in line with prior research. Firm size has a significant positive effect on firm internationalization. This supports the idea that larger firms are more likely to possess the necessary resources, capabilities and expertise to facilitate firm internationalization (Barkema & Shvyrkov, 2007; Pisani et al., 2018). Furthermore, firms operating in the manufacturing industry show a higher level of internationalization than service industry firms, which confirms the findings of Grøgaard et al. (2013).

5.2 Theoretical Implications

(54)

only few scholars, in particular Nielsen and Nielsen (2011), explored the concept more thoroughly and emphasized the uniqueness of both TMT internationalization dimensions. This study builds on that by also highlighting the distinctiveness of TMT international work experience and TMT nationality diversity in the context of firm-level international diversification. Hence, this paper supports the suggestion of Nielsen and Nielsen (2011) to treat the dimensions separately and not combine them in one index, as it has been done multiple times in research (Azam et al., 2018; Lee & Park, 2008; Nielsen, 2010).

Second, this study contributes to TMT literature by looking more closely into TMT nationality diversity. Although scholars have noted an increase in the overall number of foreign executives in TMT positions over the past decade (Ruigrok & Greve, 2008; Staples, 2007), we still lack a good understanding of nationality diversity and its consequences on strategic decision-making processes. Unlike other TMT characteristics, the stream of research on TMT nationality diversity has been only slowly emerging over the past years (Greve et al., 2009; Nielsen & Nielsen, 2013; Van Veen & Marsman, 2008). The findings add to existing upper echelons literature by showing that nationality diversity is one of the TMT characteristics which are likely to promote foreign expansion.

(55)

precise manner. Hence, this study adds to literature by raising awareness of this measurement method and encouraging researchers to further check the robustness of this measure.

Finally, this study supports the general conclusion of upper echelons research that TMT characteristics alone only have a limited explanatory power on firm internationalization. As Kirca et al. (2012) show in their meta-analysis, firm multinationality is a complex construct with a large number of known and unknown drivers. Although both TMT international work experience and TMT nationality diversity have a significant impact on firm internationalization, they only explain a small part of the whole construct.

5.3 Practical Implications

(56)

international postings. Especially multiple foreign work assignments in different countries will help broaden managers’ international knowledge and strengthen their cognitive competencies (Godart et al., 2015). Hence, firms could develop an international work rotation program, especially tailored for future executive candidates.

5.4 Limitations and Future Research

This study has several limitations, which also serve as possibilities for future research. Following the upper echelons perspective, this study assumed that observable characteristics of TMTs, namely national origin and prior international work experience, can be used as proxies for psychological constructs, team dynamics and information processing capabilities. Like many other studies in the upper echelon tradition (Acar, 2016; Barkema & Shvyrkov, 2007; Nielsen & Nielsen, 2013; Pisani et al., 2018) the mediating mechanisms (e.g., cognitive diversity and strategic choice) are not directly measured, the focus is exclusively on secondary outcomes such as firm internationalization. Future research needs to probe into this black box and closely examine the intervening mechanisms through which TMT composition influences firm outcomes, which will help validate the findings of this research. However, this would require the application of alternative research designs and data-collection methods, for instance the use of primary survey data.

(57)

et al., 2015; Le & Kroll, 2017; Maitland & Sammartino, 2015). However, the limited data availability of this study did not allow for such a refined measure. Future research could further develop this construct by considering other potential sources of international experience, such as international education, language capabilities, and international network contacts (Athanassiou & Nigh, 2002). Moreover, scholars could examine whether the national background and the international experiences of the firm’s CEO plays a special role, as suggested by Le and Kroll (2017). Future research may also benefit from studies exploring the possible interaction effects between TMT international work experience and TMT nationality diversity. This study did not consider that both measures may strengthen or weaken each other’s effect on group dynamics and team functioning. For instance, the negative implications of nationality diversity may be less pronounced if the level of international experience of the TMT members is high. Secondly, it is acknowledged that firm internationalization could have been measured in a more sophisticated way, for instance by adopting different transnationality indices (Doerrenbaecher, 2000). However, largely due to the limited time and scope of this study, firm internationalization was only operationalized with two structural indicators of foreign activity.

(58)

study does not directly control for prior firm internationalization. In this regard, it is acknowledged that firm size is only a vague proxy for prior firm internationalization.

(59)

References

Acar, F. (2016). The effects of top management team composition on SME export performance: An upper echelons perspective. Central European Journal of Operations

Research, 24(4), 833-852.

Acock, A. (2014). A gentle introduction to Stata (4th ed.). College Station, TX: Stata Press. Ali, M., Ng, Y. L., & Kulik, C. T. (2014). Board age and gender diversity: A test of competing linear and curvilinear predictions. Journal of Business Ethics, 125, 497-512. Allison, P. (1978). Measures of inequality. American Sociological Review, 43, 865–880. Ashby, W. R. (1956). An introduction to cybernetics. Chapman & Hall, London.

Athanassiou, N., & Nigh, D. (2002). The impact of the top management team′s international business experience on the firm′s internationalization: Social networks at work. Management International Review, 42(2), 157–181.

Azam, A., Boari, C., & Bertolotti, F. (2018). Top management team international experience and strategic decision-making. Multinational Business Review, 26(1), 50-70. Barkema, H., & Shvyrkov, O. (2007). Does top management team diversity promote or hamper foreign expansion? Strategic Management Journal, 28(7), 663-680.

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

Bjørnskov, C. (2008). Social Trust and Fractionalization: A Possible Reinterpretation.

European Sociological Review, 24(3), 271-283.

Blau, P. M. (1977). Inequality and heterogeneity: A primitive theory of social structure. New York: Free Press.

Bonett, D., & Wright, T. (2011). Sample size requirements for multiple regression interval estimation. Journal of Organizational Behavior, 32(6), 822-830.

Buyl, T., Boone, C., Hendriks, W., & Matthyssens, P. (2011). Top Management Team Functional Diversity and Firm Performance: The Moderating Role of CEO Characteristics.

Journal of Management Studies, 48(1), 151-177.

(60)

Caliguri, 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.

Caligiuri, P. & Tarique, I. (2012). Dynamic cross-cultural competencies and global leadership effectiveness. Journal of World Business, 47(4), 612–622.

Carpenter, M., Sanders, W., & Gregersen, H. (2001). Bundling human capital with organizational context: The impact of international experience on multinational firm performance and CEO pay. Academy of Management Journal, 44(3), 493-511.

Carroll, G., & Harrison, J. (1998). Organizational Demography and Culture: Insights from a Formal Model and Simulation. Administrative Science Quarterly, 43(3), 637-667. 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, 396-414.

Chatman J.A., & Flynn F.J. (2001). The influence of demographic heterogeneity on the emergence and consequences of cooperative norms in work team. Academy of

Management Journal 44, 956–974.

Clark, D., Li, D., & Shepherd, D. (2018). Country familiarity in the initial stage of foreign market selection. Journal of International Business Studies, 49(4), 442-472.

Cui, Lin, Li, Yi, Meyer, K., & Li, Zijie. (2015). Leadership Experience Meets Ownership Structure: Returnee Managers and Internationalization of Emerging Economy Firms.

Management International Review, 55(3), 355-387.

Cyert, R. M., & March, J. A. (1963). A behavioural theory of the firm. Englewood Cliffs: Prentice-Hall.

De Wit, F., Greer, L., Jehn, K., & Kozlowski, Steve W. J. (2012). The Paradox of Intragroup Conflict: A Meta-Analysis. Journal of Applied Psychology, 97(2), 360-390. Doerrenbaecher, C. (2000). Measuring corporate internationalisation. Intereconomics,

35(3), 119-126.

Does a firm’s strategy match its managers?

Master Thesis