Upper echelons nationality diversity
and firm internationalization
Lorenzo Marcotullio 11087099 June 2016
MSc Business Administration: International Management Track University of Amsterdam
Final Version Master Thesis Supervisor: Dr. Niccolò Pisani
Statement of Originality
This document is written by Lorenzo Marcotullio who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
ABSTRACT
Extensive research has been conducted to investigate the relationship between corporate governance characteristics and firms’ level of internationalization. One stream of research examined the impact of increased internationalization on firms’ choices regarding the structure and the composition of their BoDs and TMTs. Another stream of research followed the tenets of upper echelons theory and argued that corporate elites’ background and demographic characteristics influence corporate strategic outcomes, including the level of internationalization. The research in support of upper echelons theory has comparably paid more attention to managers’ and directors’ international work experiences than to their nationality, although nationality has a more pervasive and extensive impact on mindsets, cultural schemas and personality traits. In order to shed more light on the role of nationality, this thesis develops a comprehensive conceptual model that simultaneously considers the effects of CEO non-nativity and shareholders’, BoD and TMT nationality diversity on firms’ level of internationalization. Besides, this thesis examines the moderating effects of industry globalization and home country economic and institutional openness on the main relationships. The hypotheses were tested on the Fortune Global 500 companies listed in 2014. The findings suggest that nationally diverse shareholders and TMTs tend to push for the internationalization of their firms and that the degree of industry globalization positively moderates the relationship between shareholders’ nationality diversity and firm internationalization. Contrary to our expectations, our empirical findings also indicate that the degree of home country economic and institutional openness negatively moderates the effects of nationally diverse BoDs, non-native CEOs and nationally diverse TMTs on firms’ level of internationalization. Finally, in this study, both the degree of industry globalization and the degree of home country economic and institutional openness had a positive and direct impact on firms’ level of internationalization.
Keywords: upper echelons theory, shareholders’ nationality diversity, BoD nationality diversity, TMT nationality diversity, CEO non-nativity, level of internationalization, industry globalization, home country economic and institutional openness.
TABLE OF CONTENTS
1. INTRODUCTION ... 6
2. LITERATURE REVIEW ... 9
2.1 Internationalization and Corporate Governance ... 9
2.2 Internationalization as the independent variable ... 10
2.3 Internationalization as the dependent variable ... 11
2.3.1 Upper-Echelons theory……….. 12
2.3.2 Empirical support for Upper-Echelons theory……….. 13
2.3.3 The role of nationality within the Upper-Echelons theory……… 15
2.4 Research gap ... 16
3. THEORETICAL FRAMEWORK ... 18
3.1 Nationality diversity ... 18
3.2 Firm internationalization ... 19
3.3 Conceptual model ... 20
3.3.1 Nationality diversity of shareholders………. 21
3.3.2 Nationality diversity of the board of directors………... 21
3.3.3 CEO nationality………. 22
3.3.4 Nationality diversity of the top management team……… 23
3.3.5 Industry globalization……… 25
3.3.6 Home country economic and institutional environment……… 27
4. METHOD ... 29
4.1 Sample and data collection ... 29
4.2 Measures ... 31
4.2.1 Independent variable………. 31
4.2.2 Dependent variable……… 32
4.2.3 Moderating variables………. 32
4.2.4 Control variables………... 33
4.3 Statistical analyses and results ... 34
5. DISCUSSION ... 44
5.1 Academic relevance ... 40
5.2 Managerial implications ... 43
5.3 Limitations and suggestions for future research ... 44
6. CONCLUSION……… 50
ACKNOWLEDGEMENT ………. 53
REFERENCES ... 53
1. INTRODUCTION
In recent decades, cross-border product and factor market integration have dramatically increased, even though the world economy remains semi-globalized (Ghemawat, 2013). Especially for firms facing small domestic markets, internationalization has become necessary in order to tap foreign customers, labor and capital (Oxelheim, Gregoric, Randøy & Thomsen, 2013). In certain industries, crossing geographic borders rapidly has become an imperative, because firms with global production and distribution networks might gain a competitive advantage over less internationalized companies (Chang & Rhee, 2011). That said, increased levels of internationalization also create new challenges and risks for firms.
Multinational Companies (MNCs) are generally defined as firms that generate sales in multiple countries, own assets outside of their home country, run foreign operations and employ foreign labor (Sullivan, 1994). Selling products or services overseas requires firms to make a choice between national responsiveness and global integration (Prahalad & Doz, 1987). Ownership of foreign assets exposes MNCs to the risk of expropriation (Feinberg & Gupta, 2009). Foreign subsidiaries might engage in rent seeking behavior and thus destroy firm value (Mudambi & Navarra, 2004). Finally, managing foreign employees might result in cultural conflicts (Hofstede, 1994). In short, internationalization, while often necessary in today’s increasingly integrated world economy, heightens the complexity that firms face when running operations (Luo, 2005; Sanders & Carpenter, 1998).
Two streams of research have emerged relative to the relationship between internationalization and corporate governance. Some scholars argued that MNCs adopt new corporate governance arrangements in order to enhance their executive ranks’ information processing capacity and cognitive ability and thus enable them to face the uncertainties and
complexities of international operations (Nielsen, 2009; Sanders & Carpenter, 1998). In other words, these scholars investigated the direct effect of internationalization level on corporate governance arrangements.
Another stream of research took the opposite path to explain the relationship between corporate governance arrangements and firm internationalization and thus considered how corporate governance influences the internationalization process of firms. This line of research has its roots in Hambrick and Mason’s (1984) upper echelons theory. According to this theory, a firm’s upper echelons characteristics directly affect organizational outcomes, such as the level of international expansion. Specifically, psychological and demographic characteristics of a firm’s top managers and directors shape their interpretation of events and influence their strategic decisions (Hambrick & Mason, 1984). Therefore, the empirical studies that tested Hambrick and Mason’s (1984) propositions took various corporate governance arrangements as the independent variables and analyzed their impact on the degree of internationalization. Several scholars investigated the impact of age, firm tenure, international work experience, functional background and education level on the strategic choices of a firm’s top management (Athanassiou & Nigh, 2002; Carpenter & Fredrickson, 2001; Herrmann & Datta, 2005; Rivas, 2012a; Sambharya, 1996; Sherman, Kashlak & Joshi, 1998; Tihanyi, Ellstrand, Daily & Dalton, 2000). Besides, previous works also studied how diversity at a firm’s upper echelons affects organizational outcomes. These studies demonstrated that heterogeneity within Top Management Teams (TMTs) and Boards of Directors (BoDs) with respect to education, firm tenure, functional background and international experience has a positive impact on a firm’s level of internationalization (Carpenter & Fredrickson, 2001; Rivas, 2012b; Sambharya, 1996; Tihanyi et al., 2000).
Despite the fact that the role of foreigners in MNCs has increased in recent decades (Greve, Biemann & Ruigrok, 2015; Kaczmarek & Ruigrok, 2013; Nielsen & Nielsen, 2010),
the literature that empirically tested Hambrick and Mason’s (1984) propositions relatively neglected the relationship between nationality diversity and firm internationalization. Besides, to the best of our knowledge, nobody has ever extended upper echelons theory to a firm’s shareholders. Thus, this thesis seeks to fill these gaps in the literature by simultaneously analyzing the impact of foreign Chief Executive Officers (CEOs) and nationally diverse shareholders, BoDs and TMTs on firms’ degree of internationalization.
The conceptual model developed in this study brings several additions to the one developed by Kirca et al. (2012) in their meta-analysis of the drivers of firm multinationality. First, at the individual level, we added CEO nationality as a predictor of firm internationalization. Second, we decomposed at an even finer-grained level the group-level variable ‘TMT diversity’ by focusing on TMT nationality diversity. Third, we simultaneously examined the role of nationality diversity at all the upper-level sub-echelons, including shareholders, whose impact on corporate strategic outcomes has received comparably less attention from scholars. Finally, we included industry globalization and home country economic and institutional openness as moderating variables in our conceptual model.
This study produced several meaningful results. First, shareholders’ and TMT nationality diversity were found to be positively associated with firms’ level of internationalization, but the predicted effects of BoD nationality diversity and CEO non-nativity on firm-level internationalization were not confirmed. Second, the degree of industry globalization had a positive moderating effect only on the relationship between shareholders’ nationality diversity and firm-level internationalization. Third, contrary to our expectations, the degree of home country economic and institutional openness negatively moderated the relationships between BoD nationality diversity, CEO non-nativity and TMT nationality diversity and firms’ level of internationalization. Finally, both the degree of industry
globalization and the degree of home country economic and institutional openness had a positive direct effect on firms’ level of internationalization.
The rest of this thesis is structured as follows. The next section will review the relevant literature regarding the relationship between firm internationalization and corporate governance arrangements. After that, we will present the theoretical framework and develop the hypotheses. The following section will outline the processes of data collection and data analysis and describe the results. Next, we will discuss the theoretical and managerial implications stemming from the empirical findings, identify possible limitations to our study and suggest avenues for future research. Finally, some concluding remarks will recap our work.
2. LITERATURE REVIEW
2.1 Internationalization and Corporate Governance
The construct of firm internationalization has received substantial attention in the academic literature. According to an extensively cited article by Sullivan (1994), the degree of firm internationalization has three main attributes. First, international performance refers to the extent to which foreign sales make up a great proportion of a firm’s total revenues. Second, international structure indicates the degree to which a firm’s assets and subsidiaries are spread all over the world as compared to domestically concentrated (Caligiuri, Lazarova & Zehetbauer, 2004). Third, international attitude alludes to the international orientation of a firm’s top management (Sullivan, 1994).
Whereas a considerable number of scholars have studied the relationship between the degree of internationalization and firm performance (Hennart, 2007), relatively few studies have focused on the relationship between the degree of internationalization and corporate governance arrangements. A firm’s corporate governance structure describes the system of
control, monitoring and incentive mechanisms that determine the division of decision-making power and the distribution of rights, duties and responsibilities among selected bodies at the upper echelons of a firm (Luo, 2005; Sanders & Carpenter, 1998). Two opposite approaches have emerged in the literature with respect to the direction of the relationship between internationalization and corporate governance.
2.2 Internationalization as the independent variable
Some scholars took a firm’s degree of internationalization as the independent variable and studied its effects on an array of corporate governance arrangements. The underlying assumption is that foreign expansion creates a new set of challenges for firms. Internationalized companies have to operate in many different countries with heterogeneous cultures, institutions, competitors, customers and regulations. Besides, internationalization requires a firm to coordinate geographically dispersed activities, create synergies among worldwide operations and subsidiaries (Sanders & Carpenter, 1998) and find a balance between the conflicting demands of national adaptation and global integration strategies (Rugman & Verbeke, 1992). In short, international diversification increases the volume and the diversity of the information firms have to analyze in order to manage operations successfully. Consequently, information processing costs rise and new agency demands emerge (Luo, 2005; Sanders & Carpenter, 1998). Because of this heightened complexity, firms adopt new governance arrangements, with respect to CEO appointment, TMT and BoD size and composition (Nielsen, 2009; Sanders & Carpenter, 1998).
Specifically, Sanders and Carpenter (1998) found that, in response to a higher degree of internationalization, firms increase the size of both the TMT and the BoD, and add more outsiders to the BoD. Similarly, Luo (2005) argued that MNCs with a global scale of operations have larger BoDs with a higher number of specialized committees and a higher proportion of independent directors than domestic firms.
Furthermore, Nielsen (2009) showed that highly internationalized firms are more likely to select new TMT members that differ from existing ones with respect to nationality and international experience. These findings corroborate the argument that managerial diversity plays an important role in solving the complexity associated with international strategic decisions, because of the different network contacts, expertise, capabilities and knowledge that foreign or internationally experienced managers bring to the TMT (Nielsen, 2009). Recently, Kaczmarek and Ruigrok (2013) followed this line of thought and demonstrated that firm internationalization is positively associated with TMT nationality diversity. According to them, foreign executives not only extend the cognitive capacity of the TMT, but also ‘align the cognitive map of the TMT members with the geographic map of an MNC’s international operations’ (Kaczmarek & Ruigrok, 2013: 517), thus matching a firm’s management with its strategy (Greve, Nielsen & Ruigrok, 2009).
Similarly, Greve et al. (2015) found evidence of a positive association between a firm’s degree of internationalization and the probability that it will appoint a foreigner on the TMT. They argued that hiring foreign executives increases the information processing and cognitive capacities of the TMT, expands the network of knowledge, contacts and resources of the TMT and helps establish legitimacy and build credibility in overseas markets. Therefore, foreign TMT members can solve, at least partially, the increased complexity stemming from international expansion (Greve et al., 2015). Finally, at the CEO level, Magnusson and Boggs (2006) found that broadly internationalized firms seek CEOs with a high level of international experience.
2.3 Internationalization as the dependent variable
Another stream of research took the level of internationalization as the dependent variable and studied how upper echelons characteristics, such as TMT size and composition, affect a firm’s strategic choices, such as the decision to expand abroad. As Kirca et al. (2012)
noted in their review of previous research, scholars have studied the antecedents of firm internationalization at five main levels of analysis: individual, group, firm, industry and country. For the purpose of this thesis, we are mainly interested in those studies that examined the impact of individual and group level factors on the degree of firm internationalization. These scholars based their arguments on the upper echelons theory developed by Hambrick and Mason (1984).
2.3.1 Upper Echelons theory
According to Hambrick and Mason (1984), a firm’s strategic choices are a reflection of its top management’s demographic and background characteristics. In other words, upper echelons attributes determine organizational outcomes and, as a result, firm performance. Hambrick and Mason (1984) considered two types of upper echelons characteristics: psychological and observable. Psychological characteristics consist of managers’ cognitive base and values, which are deemed to directly affect their perceptions and, consequently, their strategic choices. Instead, observable characteristics include managers’ age, functional background, career experiences, education, socioeconomic roots and financial position. Besides, in their model, Hambrick and Mason (1984) regarded managerial group characteristics, such as group heterogeneity, as an observable determinant of a firm’s strategic decisions.
Although Hambrick and Mason (1984) recognized that upper echelons characteristics can sometimes be a reflection of the environment a firm operates in and the situation it faces, upper echelons theory clearly predicts a causal relationship between managerial backgrounds – the independent variable – and organizational outcomes – the dependent variable. However, most of the empirical studies that tested Hambrick and Mason’s (1984) propositions failed to establish causality between senior executives’ attributes and organizational outcomes and
called for the need to carry out longitudinal studies (Herrmann & Datta, 2005; Sambharya, 1996).
2.3.2 Empirical support for Upper Echelons theory
Hambrick and Mason’s (1984) propositions received ample empirical support (Magnusson & Boggs, 2006). The empirical research based on Hambrick and Mason’s (1984) seminal work can be divided into two distinct streams. The first stream of research studied the impact of psychological and observable upper echelons characteristics on organizational strategic outcomes, such as the degree of internationalization. Because of difficulties in measuring managers’ principles, assumptions and beliefs (i.e., their values and cognitive base) most scholars used observable managerial characteristics – such as demographics and career experiences – as proxies for the psychological constructs that affect managers’ decision-making process (Magnusson & Boggs, 2006). The second stream of research analyzed the effect of group heterogeneity within TMTs and BoDs on organizational strategic outcomes, such as the level of internationalization. Since Hambrick and Mason’s (1984) model did not include international expansion in the list of strategic choices that are affected by senior managers’ characteristics, both of the above-mentioned streams of research expanded the original model by taking the degree of internationalization as the dependent variable (Kaczmarek & Ruigrok, 2013).
As mentioned above, several scholars examined the impact of various CEO, TMT and BoD characteristics on the internationalization extent of a firm. At the CEO-level, Athanassiou and Nigh (2002), Kirca et al. (2012) and Magnusson and Boggs (2006) found a positive relation between CEOs’ international work experience and their firms’ level of internationalization. Moreover, Jaw and Lin (2009) found an inverted U-shaped relationship between CEO tenure and the degree of firm internationalization. Most of the academic papers, though, focused on TMT and BoD characteristics. TMT members’ level of
international work experience (Athanassiou & Nigh, 2002; Carpenter & Fredrickson, 2001; Herrmann & Datta, 2005; Kirca et al., 2012; Rivas, 2012a; Sambharya, 1996; Tihanyi et al., 2000), previous experience as CEO of an MNC (Rivas, 2012a), average firm tenure (Sherman et al., 1998; Tihanyi et al., 2000), average education level (Herrmann & Datta, 2005; Kirca et al., 2012;Tihanyi et al., 2000) and TMT size (Kirca et al. 2012; Sherman et al., 1998) were all found to be positively associated with the degree of firm internationalization. Instead, average age of TMT members (Herrmann & Datta, 2005; Tihanyi et al., 2000) was found to be negatively associated with the level of firm internationalization. Finally, Rivas (2012a) found evidence of a positive relation between BoD international experience and firm internationalization, whereas Sherman et al. (1998) failed to find support for their hypotheses that the proportion of non-executive directors in the BoD and the size of the BoD are positively related to the extent of firm internationalization.
Some scholars, instead, studied the relationship between diversity at the upper echelons of a firm and various organizational outcomes. Within the field of upper echelons theory, a considerable amount of research studied how diversity, including nationality diversity, within firms’ executive ranks affects corporate performance. Instead, other organizational outcomes, such as the level of internationalization, received relatively less attention from scholars (Schuster & Bader, 2015). There is no consensus in the literature concerning the relationship between TMT diversity and firm internationalization. On the one hand, heterogeneity within the TMT can enhance the quality of its decisions because of higher information processing capacity and lower probability of group think. On the other hand, TMT diversity can hamper the implementation of decisions and therefore slow down the internationalization process, because group diversity creates the potential for communication problems, conflicts and subgroup formation (Kirca et al., 2012). However, Kirca et al. (2012) failed to establish the existence of either a positive or a negative
association between TMT diversity and firm internationalization. Only those studies that took a more narrow approach to diversity managed to obtain significant findings. Specifically, TMT’s educational heterogeneity (Carpenter & Fredrickson, 2001), firm tenure heterogeneity (Carpenter & Fredrickson, 2001; Tihanyi et al., 2000), functional background diversity (Rivas, 2012b) and heterogeneity of international experience (Sambharya, 1996) were all found to be positively correlated to the level of firm internationalization. With respect to BoDs, only functional background diversity (Rivas, 2012b) was proved to be positively associated with the degree of firm internationalization.
2.3.3 The role of nationality within the Upper Echelons theory
As outlined above, most of the literature within the scope of upper echelons theory explored the impact of age, firm tenure, international work experience, functional background and education level on TMTs’ strategic choices. However, nationality and nationality diversity have received relatively less attention from scholars (Kaczmarek & Ruigrok, 2013). Nielsen and Nielsen (2010) and Ghemawat and Vantrappen (2015) found evidence of a positive association between the odds of a firm having foreign TMT members and its level of internationalization. In addition, Caligiuri et al. (2004) found support for their hypotheses that a positive relation exists between both TMT and BoD nationality diversity and the degree of firm internationalization. However, their study is subject to serious limitations because nationality diversity was operationalized as the ratio of foreign TMT (or BoD) members to total TMT (or BoD) members. This ratio does not truly reflect diversity, because it ignores both the numbers of different nationalities represented on the TMT and the cultural distance between the represented nationalities (Caligiuri et al., 2004).
Kaczmarek and Ruigrok (2013) argued that both domestic managers with international work experience and foreign executives are potential sources of TMT internationalization. They also claimed that the impact of international work experiences on
managers’ mindsets, cultural schemas and personality traits is less extensive and pervasive than that of nationality. Therefore, nationality arguably plays a fundamental role in forming the values and the cognitions that, according to upper echelons theory, affect managers’ perceptions and strategic decisions (Hambrick & Mason, 1984). However, despite the intuitive appeal of the connection between TMT nationality and firm internationalization, the academic literature has comparably paid more attention to the construct of international work experience (Caligiuri et al., 2004; Kaczmarek & Ruigrok, 2013; Nielsen & Nielsen, 2010). Nationality thus remains an under-studied characteristic of a firm’s upper echelons.
Nonetheless, the relevance of TMT nationality is proved by the fact that over the past 20 years the number of foreigners appointed to the TMTs and BoDs of large MNCs has increased substantially (Greve et al., 2015; Kaczmarek & Ruigrok, 2013; Nielsen & Nielsen, 2010). Foreign TMT members can play a crucial role in alleviating the liability of foreignness that MNCs face in host countries (Ghemawat & Vantrappen, 2015), because foreign executives possess valuable knowledge about the culture and the institutions of their country of origin. Nationality is thus a source of precious insights about the norms, the customs and the rules of the game of culturally and institutionally distant economies (Nielsen & Nielsen, 2010). Besides, nationality diversity within the TMT is likely to produce a heterogeneous set of cultural attitudes, opinions, values, preferences, information resources and skills sets. This wide diversity will enhance the international orientation of the TMT (Caligiuri et al., 2004) and will arguably be reflected in a higher propensity to expand abroad.
2.4 Research gap
Although the study by Caligiuri et al. (2004) represented a first attempt in this direction, the relationship between nationality diversity and firm internationalization remains relatively unexplored, especially using upper echelons theory. Besides, the academic literature has neglected the relationship between shareholders’ nationality diversity and the
level of firm internationalization, although the existence of this relationship could be inferred by combining the studies by Oxelheim et al. (2013) and Caligiuri et al. (2004). The former discovered a positive association between foreign strategic ownership and the proportion of foreign directors, while the latter demonstrated that BoD nationality diversity is positively correlated with firm-level internationalization. In other words, while many scholars have investigated the antecedents of foreign ownership of equity, the effects of foreign ownership on organizational outcomes are still quite unknown. For example, at the firm level, Miletkov et al. (2014) showed that increasing board independence attracts foreign investors and, at the country level, Leuz et al. (2010) found that underdeveloped legal institutions, poor disclosure norms and weak protection of investors’ rights reduce foreigners’ investments in a country’s firms. Instead, Fernández and Nieto (2006) uncovered a relation between foreign ownership and firm-level internationalization, but their study only focused on Small and Medium Enterprises (SMEs). Hence, we argue that the model by Hambrick and Mason (1984) could be extended to include the impact of shareholders’ characteristics, such as nationality diversity, on organizational outcomes, such as the level of internationalization.
To the best of our knowledge, no study has ever simultaneously analyzed the impact of non-native1 CEOs and nationally diverse shareholders, BoDs and TMTs on the strategic decision to expand business operations beyond domestic borders. This thesis seeks to fill this gap in the literature and advance the current knowledge of how the presence of foreigners in a firm’s upper echelons affects strategic outcomes, such as the decision to expand overseas. In other words, this thesis aims to test whether CEOs who originate from a country other than the one where the MNC is headquartered, as well as nationally diverse shareholders, BoDs and TMTs lead their firms to higher levels of internationalization.
3. THEORETICAL FRAMEWORK
According to the original theoretical model by Hambrick and Mason (1984), top managers’ cognitions and values affect their perception and interpretation of reality and therefore determine their strategic choices. In addition to psychological factors, top managers’ demographic characteristics – age, education, career experiences, firm tenure, functional background, socioeconomic roots and financial situation – impact their strategic decisions. Examples of such strategic decisions are product innovation, related and unrelated diversification, backward and forward integration and financial leverage. Moreover, group heterogeneity influences the group’s decision making process, because it affects the level of conflict and the generation of strategic alternatives.
Although it is widely recognized that someone’s nationality strongly influences his or her values and cognitions (Hofstede, 1994), the original framework by Hambrick and Mason (1984) did not include nationality among the relevant demographic characteristics of top managers. Besides, the initial model did not consider the decision to expand a company’s operations across national borders as one of the strategic choices determined by upper echelons characteristics. However, while upper echelons theory was soon extended to include firm internationalization as a dependent variable, the study of managers’ nationality is a more recent enhancement to the original framework (Kaczmarek & Ruigrok, 2013).
3.1 Nationality diversity
Nationality diversity refers to the extent to which a firm’s top managers, directors and shareholders differ in terms of their country of origin. Despite its intuitive appeal, nationality is an understudied construct, because most scholars used international business experience as a proxy for managers’ and directors’ international orientation. However, nationality is an important determinant of the level of internationalization of a firm’s upper echelons, because
it has a direct effect on the cultural values, attitudes, preferences, communication styles and interaction patterns of top managers, directors and shareholders (Caligiuri et al., 2004; Kaczmarek & Ruigrok, 2013). In other words, nationality is associated with a given national culture, which in turns determines the cultural beliefs, cognitive schemas and some personality traits of the members of that national group (Greve et al., 2009). Besides, nationality is a critical dimension upon which social identification occurs (Nielsen, 2009).
Nationality diversity has both positive and negative effects on a group’s decision making process. On the one hand, cross-national groups are, on average, more creative and less likely to experience groupthink. They also tend to consider more varied perspectives and viewpoints. On the other hand, these groups are more prone to miscommunication, conflict and subgroup formation, which lengthen and complicate the decision making process (Kirca et al., 2012; Punnett & Clemens, 1999). Moreover, a high level of demographic dissimilarity within a group can lessen the strength of the relationships between group members, slow down the integration process and increase the probability of exit (Greve et al., 2015). Since nationality diversity has a significant impact on groups’ decision making process, upper echelons theory should be expanded to include nationality diversity among the upper echelons characteristics that determine strategic choices and other corporate outcomes. Therefore, we follow the second stream of research outlined above, namely the one taking corporate governance characteristics as the independent variable and firm internationalization as the dependent variable.
3.2 Firm internationalization
The degree of internationalization refers to the extent to which a firm undertakes value-adding activities beyond the borders of its home country. Many labels, such as multinationality, geographic diversification and international expansion have been adopted in the academic literature to refer to the same construct, namely the level of involvement of a
company in foreign markets (Hitt, Tihanyi, Miller & Connelly 2006; Kirca et al. 2012). The higher a firm’s degree of internationalization, the greater the scope of its operations beyond the domestic market (Wiersema & Bowen, 2008). In other words, highly internationalized firms are characterized by a geographic dispersion of sales, assets, employees and subsidiaries across national borders (Athanassiou & Nigh, 2002; Kaczmarek & Ruigrok, 2013).
Thanks to the opening of emerging markets and because of increasing globalization and cross-border competition, international expansion and geographic diversification have become remarkable strategic options for companies seeking to sustain their competitive advantage (Hitt et al., 2006; Nielsen, 2009). Therefore, the original upper echelons model should be expanded to include firm internationalization as one of the strategic outcomes that reflect upper echelons characteristics.
3.3 Conceptual model
At the highest level of the corporate hierarchy there are shareholders, who collectively own the company and its assets. Shareholders elect the members of the board of directors, who, in turn, appoint the chief executive officer. The CEO then nominates the members of the top management team. As a result of this progressive selection process, there are clear complementarities in terms of nationality diversity across the four main upper sub-echelons, meaning that nationality diversity within one sub-echelon tends to lead to enhanced nationality diversity in other sub-echelons (Ghemawat & Vantrappen, 2015; Oxelheim et al., 2013). We propose that nationality diversity at each upper sub-echelon individually affects the level of firm internationalization. Furthermore, we argue that two contingencies affect these relationships. Specifically, we expect that the level of industry globalization and the degree of economic and institutional openness of the home country positively moderate the relationship between upper echelons nationality diversity and firm internationalization.
3.3.1 Nationality diversity of shareholders
In their review of the research on international diversification, Hitt et al. (2006) included ownership as one of the antecedents of international diversification. Beyond the BoD and the TMT, owners have the ability to influence the company’s decision to expand and diversify internationally (Hitt et al., 2006). More specifically, Fernández and Nieto (2006) discovered a positive association between the presence of foreign blockholders in SMEs and both the export propensity and the export intensity of SMEs. They argued that foreign blockholders affect the strategy of SMEs by providing knowledge of foreign markets. This effect of foreign shareholders on SMEs’ level of internationalization can be expected to hold for larger companies as well. In other words, we claim that foreign shareholders facilitate the internationalization of their firms by supplying knowledge of foreign markets and by providing access to foreign capital and foreign business networks.
Hypothesis 1a. Shareholders’ nationality diversity is positively related to the firm’s level of internationalization.
3.3.2 Nationality diversity of the board of directors
Boards of directors play three main roles in corporations. The first one is the monitoring role, which means that directors promote, nominate, evaluate and, if necessary, dismiss top executives. The second one is the advisory role, which refers to the fact that directors participate in the development of the main guidelines of the firm’s strategy. Finally, the resource provision role implies that directors can provide access to key resources, such as contact networks and information channels, and help the company obtain institutional legitimacy (Oxelheim et al., 2013). Nationally diverse BoDs can directly affect the level of firm internationalization through the advisory and the resource provision roles. First, when performing their advisory role, nationally diverse BoDs are less likely to be constrained by domestic myopia, because of the variety of cultural backgrounds, opinions, information
sources and skill sets they have (Caligiuri et al., 2004). Second, given the complexity of the strategic issues faced by multinational firms, foreign directors can represent a valuable resource during the internationalization process. Oxelheim et al. (2013) argued that foreign directors are more likely to possess valuable network contacts and precious knowledge of foreign markets’ dynamics (i.e., customers, suppliers, competitors and key resource providers). Besides, the presence of foreign directors can signal a company’s commitment to foreign markets and resources, which can ease the company’s entry into those foreign markets (Oxelheim et al., 2013). Therefore, through the advisory and resource provision roles, nationally diverse BoDs facilitate and encourage the internationalization of their firms.
Hypothesis 1b. BoD nationality diversity is positively related to the firm’s level of internationalization.
3.3.3 CEO nationality
Chief executive officers affect the strategic direction of their companies, both indirectly through the appointment of TMT members and directly through their strategic decisions, the latter being a reflection of their work experiences and demographic characteristics (Athanassiou & Nigh, 2002). Indeed, Ghemwat and Vantrappen (2015) found evidence of a positive correlation between the presence of foreign CEOs and the share of non-native TMT members. Moreover, foreign CEOs might be better equipped to face the complexity of internationalization demands, because of their knowledge of foreign markets and their access to foreign network contacts (Kirca et al., 2012). Therefore, CEOs that are not native of the country in which the corporation is headquartered are expected to promote the internationalization of the corporation, both through their tendency to nominate foreign executives to the TMT (Ghemawat & Vantrappen, 2015) and through their ability to deal with the difficulties of conducting business operations in different and often quite diverse foreign markets (Kirca et al., 2012).
Hypothesis 1c. CEO non-nativity is positively related to the firm’s level of internationalization.
3.3.4 Nationality diversity of the top management team
Although most previous studies in the upper echelons literature took managers’ years of foreign work experience as a proxy for their international orientation, nationality also directly affects managers’ international attitude. Therefore, executives’ nationality is an essential, albeit understudied, determinant of firm internationalization (Caligiuri et al., 2004; Kaczmarek & Ruigrok, 2013; Nielsen & Nielsen, 2010), which deserves more academic research, especially because the number of foreigners appointed to top managerial positions has increased substantially over the past 20 years (Greve et al., 2015).
When firms internationalize their operations, they usually face the liability of foreignness (Zaheer, 1995) and incur some additional costs of doing business abroad (Hymer, 1976). These costs arise because of the cultural, institutional, geographic and economic distance between the home and the host country. In other words, when entering into a foreign market, MNCs have to cope with the demands of local customers, suppliers, competitors, legal systems, institutions and a multitude of other constituencies with whom they are unfamiliar (Greve et al., 2009). Besides, MNCs have to establish legitimacy in the host country to minimize the costs associated with negative attitudes towards foreign companies (Greve et al., 2015). Together, all these costs raise the uncertainty of conducting business operations in a foreign market. In turn, this uncertainty enhances the information-processing demands and the complexity of the strategic decisions facing MNCs’ TMTs (Greve et al., 2015, Nielsen & Nielsen, 2010).
The academic literature on the antecedents and the consequences of TMT composition has suggested several reasons why nationality diversity can help the TMT
reduce the uncertainty and solve the complexity associated with foreign business operations. First, foreign executives extend the TMT’s information-processing capacity because of their superior knowledge of the dynamics of foreign markets and institutions (Greve et al., 2015) as well as greater understanding of culturally different norms of behavior, rules of conduct and business practices (Nielsen & Nielsen, 2010). Second, foreign TMT members can increase the TMT’s resource-seeking capacity because of their wider contact networks abroad, especially in those countries where the MNC wishes to enter (Greve et al., 2015, Nielsen, 2009). In addition, foreign TMT members are more likely than domestic TMT members to be aware of foreign market opportunities (Kirca et al., 2012). Third, the appointment of foreign executives to the TMT sends a powerful signal to foreign stakeholders about the firm’s commitment to foreign operations and helps the firm build legitimacy and credibility in foreign countries (Ghemawat & Vantrappen, 2015; Greve et al., 2015).
Therefore, a higher level of nationality diversity at TMT-level helps the MNC attenuate the liability of foreignness and gain legitimacy in foreign markets (Greve et al., 2009), especially if foreign TMT members come from target countries, i.e. those countries in which the firm operates or wishes to start operations (Ghemawat & Vantrappen, 2015). Besides, nationally diverse TMTs are not constrained by domestic myopia and consider a wider range of strategic alternatives than nationally homogeneous TMTs (Kirca et al., 2012). We thus argue that nationality diversity fosters the international orientation of the TMT and, as a result, enhances the TMT’s propensity to choose in favor of foreign expansion plans. Indeed, Caligiuri et al. (2004), Kaczmarek and Ruigrok (2013) and Ghemawat and Vantrappen (2015) found a positive association between TMT nationality diversity and the level of firm internationalization. Hence, since TMT members play a key role in the decision
to internationalize the scope of a firm’s operations (Hitt et al., 2006), we expect nationality diversity at TMT-level to be positively correlated to the level of firm internationalization.
Hypothesis 1d. TMT nationality diversity is positively related to the firm’s level of internationalization.
3.3.5 Industry globalization
According to the theoretical framework by Hitt et al. (2006), which summarizes the outstanding empirical work on international diversification, the competitive environment of a firm’s industry affects the firm’s level of internationalization both directly and by moderating the relationship between upper echelons conditions – such as TMT characteristics, BoD composition and ownership structure – and the level of firm internationalization. More specifically, both the overall degree of competition and the extent of foreign competition in a company’s domestic market positively affect the level of firm internationalization (Kirca et al., 2012; Wiersema & Bowen, 2008). On the one hand, industry rivalry shrinks profit margins and encourages companies to seek foreign market opportunities (Kirca et al., 2012). On the other hand, the presence of foreign competitors in the domestic market creates incentives for domestic firms to expand their operations internationally for two main reasons. First, since foreign rivals may introduce new technologies and capabilities to the domestic market, domestic firms are forced to increase their competitiveness, productivity and efficiency in order to remain profitable. Second, those domestic companies that manage to face up to the challenges of foreign competition also prove their ability to compete against foreign rivals at a global level (Wiersema & Bowen, 2008).
Furthermore, the degree of globalization of an industry also plays a key role in determining the level of firm internationalization (Wiersema & Bowen, 2008). According to Porter (1986) there are two main types of industries: multidomestic and global. In
multidomestic industries competition takes place on a country-by-country basis. This means that although MNCs may benefit from the one-time transfer of capabilities and know-how from one country to another, their competitive advantage is largely country-specific. Instead, in global industries firms’ competitive position in one country is tightly connected to its position in another country and competition occurs on a global scale (Porter, 1986). Moreover, in global industries the conduct of global-scale operations represents a source of competitive advantage and internationalization becomes a strategic priority, because the pressure to internationalize operations is such that a slow international expansion poses greater risks than a rapid international expansion (Chang & Rhee, 2011; Wiersema & Bowen, 2008). Indeed, Wiersema and Bowen (2008) proved the existence of a positive relationship between the degree of industry globalization and the level of firm internationalization.
In their study of the Fortune Global 500 companies, Ghemawat and Vantrappen (2015) found that the presence of non-natives in companies’ upper echelons is positively correlated to the sector in which a company operates and to the level of firm internationalization. Certain sectors, such as health care and pharmaceuticals, consumer retail and business services, have historically exhibited both a higher share of non-native managers and a higher level of firm internationalization (Ghemawat & Vantrappen, 2015). Besides, industry was often included as a control variable in studies linking corporate governance characteristics to the level of firm internationalization (Rivas, 2012a; Rivas, 2012b). For example, Rivas (2012b) found that membership to the NAICS industries ‘Trade & Transportation’ and ‘Other Services’ had a negative impact on firms’ level of internationalization.
The studies mentioned above proved that industry membership has a direct effect on firms’ level of internationalization, because certain industries are more globalized than others. In other words, the firms operating in more globalized industries are under greater
pressure to internationalize their operations because competition occurs on a truly global basis in these industries, and foreign expansion may be necessary to remain competitive vis-à-vis foreign competitors (Porter, 1986; Wiersema & Bowen, 2008). We argue that the level of industry globalization positively moderates the relationship between upper echelons nationality diversity and the degree of firm internationalization. Specifically, we contend that in global industries, nationally diverse corporate elites face an even greater pressure to internationalize their firms’ operations than in multidomestic industries. Instead, in multidomestic industries, the efforts of foreign shareholders, directors, managers and CEOs to diversify their companies’ operations internationally may be constrained by the concerns of their domestic counterparts that the costs of foreign expansion may exceed the benefits. Therefore, we claim that the greater is the level of industry globalization, the greater is the incentive of nationally diverse shareholders, directors, managers and CEOs to expand their companies’ operations abroad.
Hypothesis 2. The degree of industry globalization positively moderates the relationships hypothesized in H1a, H1b, H1c and H1d.
3.3.6 Home country economic and institutional environment
According to Hitt et al. (2006), home and host country resources and institutional environments both moderate the relationship between upper echelons characteristics and the level of firm internationalization and directly affect the latter. However, early academic research in this area focused on host country resource endowments and institutions, suggesting that MNCs prefer to enter countries with peculiar country-specific advantages, low bureaucratic costs and high protection of property rights (Hitt et al., 2006). More recently, Kirca et al. (2012) found a positive association between home country munificence and the level of firm internationalization and concluded that the home country environment influences the degree of firm internationalization because of the unique resources that each
country’s institutions provide to domestic firms. Besides, Ghemawat and Vantrappen (2015) discovered a positive correlation between the presence of non-natives in a firm’s upper echelons and the country in which the firm is headquartered. Specifically, they found that the levels of development and globalization of the home country are positively correlated with the incidence of non-native executives in the upper echelons of the country’s largest firms, while the size of the home country’s economy is negatively correlated with it. Besides, Ghemawat and Vantrappen (2015) found that firms headquartered in English-speaking countries and in those countries whose citizens speak multiple languages tend to hire more foreign managers.
According to the studies cited above, the development and the openness of the home country’s economy are positively associated with domestic firms’ level of internationalization, because wealthy and globalized countries facilitate the growth and the internationalization of their firms. We argue that an environment characterized by a low degree of economic and institutional openness constraints the efforts of foreign shareholders, directors, managers and CEOs to internationalize their companies’ operations. Instead, we contend that an economically and institutionally open home country provides an additional inducement for nationally diverse corporate elites to push for the international expansion of their companies. In other words, we expect the openness of the home country’s economic and institutional environment to positively moderate the relationship between upper echelons nationality diversity and firms’ level of internationalization.
Hypothesis 3. The degree of economic and institutional openness of the home country positively moderates the relationships hypothesized in H1a, H1b, H1c and H1d.
4. METHOD
4.1 Sample and data collection
This study employed a cross-sectional research design to analyze how the nationality diversity of shareholders, managers and directors affects firms’ level of internationalization, as well as how the home country’s economic and institutional environment and the globalization of the firm’s industry moderate the main relationships. The sample was taken from the Fortune Global 500 as of 2014, which lists, in descending order, the world’s largest companies according to their revenues in 2013. The vast majority of the Fortune Global 500 companies are multinational enterprises, meaning that ‘they produce and/or distribute products and/or services across national borders’ (Rugman & Verbeke, 2004: 6). Since MNCs are allegedly more likely to be cross-listed and to hire foreigners at their upper
echelons than single-country companies, this sample was deemed appropriate to investigate the relationship between the presence of foreigners at firms’ upper echelons and the level of internationalization of those firms.
The Fortune Global 500 companies operate in any type of industry. Hence, this sample is free of any biases related to the characteristics of particular industries. Moreover, these companies are based in 37 different countries. Thus, this sample fits the purpose of studying the moderating effect of firms’ home country institutions on the relationship between nationality diversity at firms’ upper echelons and their level of internationalization. Most of these companies are incorporated in the United States (128 or 25.6%), followed by China (95 or 19%), Japan (57 or 11.4%), France (31 or 6.2%), Germany (28 or 5.6%) and Great Britain (28 or 5.6%).
For each company in the sample, firm-level data was collected from the Orbis database and from the firm’s annual report in 2013. The Bureau van Dijk’s Orbis database was deemed to be suitable for this study, because it is a comprehensive database, which contains specific information about most public and private organizations in virtually any country (De Jong & Van Houten, 2014). However, for most state-owned Chinese companies, data was not available on Orbis and the annual report was not accessible online. Therefore, the sample does not contain sufficient information about Chinese companies to obtain meaningful insights about them.
The afore-described data collection method enabled us to obtain detailed information about large and geographically dispersed MNCs operating in any industries. Therefore, our results will be free of country- and industry-based biases. However, this sample is not representative of the entire population of companies, because it only includes very large firms whose annual revenues range from USD 476.294 million (Walmart) to USD 23.706 million
(Raytheon). Therefore, the results of this study are only applicable to large-sized MNCs. Besides, since data about the Fortune Global 500 firms was collected at a point in time, meaning that this study is cross-sectional, it was not possible to prove causality between nationality diversity and firm internationalization.
4.2 Measures
4.2.1 Independent variable
The independent variable in this study is nationality diversity. As mentioned above, Caligiuri et al. (2004) operationalized nationality diversity by means of a ratio of foreign to total team members. However, such a simple ratio merely measures a team’s degree of internationalization, and not its nationality diversity. More recently, Kaczmarek and Ruigrok (2013) applied Blau’s (1977) index2
in order to truly capture TMT nationality diversity. The index ranges from zero to one and the closer it gets to unity, the larger the number of nationalities in the team. Besides, Caligiuri et al. (2004) called for the need to take into account the cultural distance between team members’ countries of origin. Magnusson and Boggs (2006) provided a first attempt in this direction. They measured the cultural diversity of CEOs’ international experiences by creating a new cultural distance construct based on Ronen and Shenkar’s (1985) cultural clusters. Ronen and Shenkar (1985) identified eight country clusters (Arab, Near Eastern, Nordic, Germanic, Anglo, Latin European, Latin American and Far Eastern) and four independent countries (Brazil, Japan, India and Israel). During the data collection process, we grouped managers’ and directors’ nationalities into six regions: North America (US and Canada), Central and South America, Europe, Asia, Oceania and other regions. Although these macro-regions do not exactly match Ronen and Shenkar’s (1985) cultural clusters, they still represent broad cultural groups. Therefore, in order to assess heterogeneity in terms of both nationality and culture, we operationalized TMT and
2 (1 - ∑ 𝑝
BoD nationality diversity with Blau’s (1977) index, computed using the afore-mentioned macro-regions, rather than specific nationalities.
Since shareholders are very large groups compared with TMTs and BoDs, it may be less relevant to measure the scope of their internationalization. Instead, we argue that the scale of their internationalization provides a sufficiently meaningful gauge of their nationality diversity. Therefore, shareholders’ nationality diversity was operationalized with the ratio of foreign to total shareholders. Finally, CEOs were coded as 1 if they were native of the country in which their company is headquartered, as 2 otherwise.
4.2.2 Dependent variable
The dependent variable is firm internationalization. Athanassiou and Nigh (2002) and, more recently, Ghemawat and Vantrappen (2015) operationalized firm internationalization with a composite measure based on the average of the ratios of foreign to total assets, sales and employees, which are highly correlated. Unfortunately, our data set lacked information about the percentage of companies’ foreign assets. Therefore, we operationalized the level of internationalization using the percentage of foreign sales, and then used the ratio of foreign to total employees as a robustness check. The ratio of foreign to total employees refers to the percentage of a company’s employees who work outside the company’s home country.
4.2.3 Moderating variables
There are two moderators in this study, namely the degree of economic and institutional openness of a firm’s home country and the degree of globalization of the industry in which the company operates. In order to account for the degree of economic and institutional openness of each company’s home country, we used a revisited and updated version of the Maastricht Globalization Index (MGI) by Figge and Martens (2014). The MGI is a composite index, which aggregates information about a country’s political, economic,
social and cultural, technological and environmental domains. The MGI can be deemed to be a reliable measure because it is based on data from the World Bank’s World Development Indicators. Unfortunately, the latest version of the MGI only covers 117 medium-to-large countries (Figge & Martens, 2014). Therefore, five companies from Taiwan, two from Singapore and one from Luxembourg were removed from the sample when we tested hypothesis 3.
Wiersema and Bowen (2008) captured the degree of an industry’s globalization by looking at its international trade linkages. More specifically, they operationalized industry globalization with the ratio of world industry imports to world industry sales. Similarly, we retrieved data about trade levels per industry from the World Bank’s World Integrated Trade Solution, and then ranked the industries in the sample on a scale from 1 to 4, where industries scoring 4 are the most globalized.
4.2.4 Control variables
In this study, we controlled for company size, performance and age. Company size is a frequently used control variable (Rivas, 2012b), because of its relation to performance and internationalization (Athanassiou & Nigh, 2002; Carpenter & Fredrickson, 2001). Large firms usually have a large number of employees and other abundant resources to deal with the difficulties of entering foreign markets and operating abroad (Kaczmarek, & Ruigrok, 2013; Rivas, 2012a). In other words, larger firms have greater information-processing capabilities, which enable them to solve the complexities associated with transcending national borders (Athanassiou & Nigh, 2002). At the same time, though, bigger firms may experience more internal resistance against substantial strategic changes (Rivas, 2012a). In order to account for the potential impact of company size on firms’ level of internationalization, we included it as a control variable in our analysis. Similarly to most
prior studies in this field of research, we measured company size as the logarithm of annual sales in the most recent fiscal year (Nielsen & Nielsen, 2010).
The second control variable is company performance. Sanders and Carpenter (1998) argued that a firm’s performance is related to its level of internationalization. On the one hand, poor performance may encourage firms to seek higher returns by changing the corporate strategy and entering into new markets. On the other hand, poor performance may shrink the amount of resources available to the firm and thus limit its ability to expand abroad (Tihanyi et al., 2000). Therefore, because of its ability to influence a firm’s internationalization decisions, we included company performance as a control variable. In this study, company performance was measured as the Return On Assets (ROA), which is defined as the ratio of net earnings to total assets.
The third control variable is company age, measured as the difference between 2013 and the year of incorporation. Although company age is not a common control variable in this field of research, we included it in the analyses because previous research has found that firm age affects corporate performance both in the short and in the long term (de Jong & van Houten, 2014). Besides, according to the Uppsala model, firms internationalize only after a lengthy development period in the home country (Johanson & Vahlne, 1977), although the relatively recent rise of international new ventures might undermine the reliability of the internationalization process model (Oviatt & McDougall, 2005).
4.3 Statistical analyses and results
Table 1 presents some descriptive statistics regarding the mean, standard deviation and correlation coefficients of the dependent, independent and control variables. As expected, the two measures of the level of internationalization – the ratio of foreign to total sales and the ratio of foreign to total employees – are highly correlated, r(142) = 0.631, p < 0.01.
However, this does not hamper the reliability of the regression analysis, because these measures were tested in separate models. Anyway, it is necessary to look at the correlations between the predictor variables, namely the independent and the control variables, in order to check for the existence of multicollinearity. Indeed, high correlation coefficients may signal multicollinearity (Field, 2009).
Shareholders’, BoD and TMT nationality diversity and CEO non-nativity all correlate significantly (p < 0.01) with each other and with the two measures of the level of internationalization, although no correlation coefficient exceeds 0.7, the threshold beyond which the correlation is deemed to be very high. Company size does not correlate with any other predictor variable, while company age correlates significantly with BoD nationality diversity (p < 0.05), CEO non-nativity (p < 0.05) and TMT nationality diversity (p < 0.01), and company performance correlates significantly with BoD nationality diversity (p < 0.01) and CEO non-nativity (p < 0.05). These results evidenced the need to work out additional collinearity statistics in order to check for the existence of multicollinearity between predictor variables. Therefore, we computed the tolerance levels and the Variance Inflation Factors (VIF). These measures indicate the portion of variance of a predictor variable that is not explained by other predictor variables (O’brien, 2007; Pallant, 2011). Most researchers regard a VIF of 10 and a tolerance level below 0.2 as signs of severe multicollinearity (Field, 2009). In our multiple regression analyses, the tolerance level was above 0.2 and the VIF below 10 for all the variables. Therefore, the analyses were not affected by multicollinearity.
Table 1. Descriptive statistics: means, standard deviations and correlation coefficients
Statistical significance: *p < 0.5; **p < 0.01
Variable Mean Std. Dev. 1 2 3 4 5 6 7 8 9
1. Foreign to total sales 0.3515 0.33473 1 2. Foreign to total employees 0.4169 0.29027 0.631** 1 3. Shareholders’ nationality diversity (foreign/total) 0.4981 0.29078 0.330** 0.441** 1 4. BoD nationality diversity (Blau’s index) 0.1521 0.18868 0.207** 0.290** 0.183** 1 5. CEO non-nativity (1-No, 2-Yes) 1.1411 0.34855 0.151** 0.293** 0.228** 0.512** 1 6. TMT nationality diversity (Blau’s index) 0.1305 0.19548 0.274** 0.359** 0.144** 0.664** 0.461** 1 7. Company age (years) 75.1256 61.41631 0.068 0.145 -0.018 0.101* 0.099* 0.223** 1
8. Company size (log sales)
4.7002 0.25382 -0.013 -0.053 0.064 -0.017 -0.063 -0.033 -0.075 1
9. Performance (ROA)
The descriptive statistics indicate that the Fortune Global 500 companies of 2014 had an average age of 75.13 years and average sales of USD 62.12 million. As mentioned earlier, in this study the level of firms’ internationalization was operationalized as the ratio of foreign to total sales and as the ratio of foreign to total employees. Data about foreign sales was available for 320 companies. On average, these companies earned 35.15% of their revenues abroad. Data about the percentage of employees working abroad was available for 146 companies. On average, these companies employed 41.70% of their workforce outside their home country. Finally, out of the 404 companies for which data was available, only 57 had a foreign CEO.
Correlation analysis provided initial evidence of the existence of a correlation between the independent variables and both measures of the level of firm internationalization. However, we carried out hierarchical multiple regression to prove the significance of our conceptual model. Tables 2 and 3 show the results of these analyses.
Hypotheses 1a, 1b, 1c and 1d predicted that nationally diverse shareholders, BoDs and TMTs and non-native CEOs lead their firms towards higher levels of internationalization. Therefore, we performed hierarchical multiple regression to investigate the ability of upper echelons nationality diversity and CEO non-nativity to predict firms’ level of internationalization, after controlling for company size, age and performance. At first, we took the percentage of foreign sales as the dependent variable.
In the first step of the hierarchical multiple regression, we entered the control variables – company size (log of total sales), age (years) and performance (ROA) – as predictors. The model was marginally significant, F (3, 327) = 2.555, p < 0.1, and explained 2.3% of the variance in foreign sales. After entering shareholders’ nationality diversity (foreign/total), BoD nationality diversity (Blau’s index), CEO non-nativity (1=No, 2=Yes)
