TOP MANAGEMENT TEAM NATIONALITY
DIVERSITY AND FIRM MULTINATIONALITY
By: Paula Roxana Bogățan Student number: 10602909
Date: 30 June 2014
MSc Business Studies: International Management Final Version Master Thesis
Supervisor: Dr. Niccolò Pisani Second supervisor: Dr. Stephan von Delft
ABSTRACT
This research takes a closer look at the relationship between top management team (TMT) diversity and firm multinationality. Deriving from the upper echelons theory and decomposing the TMT diversity construct, the association is made between a specific demographic characteristic of the TMT – nationality diversity – and the level of expansion into foreign markets. By focusing on the largest multinationals as ranked in the Fortune Global 500 list of 2013 we find that the TMT nationality diversity positively relates to firm multinationality. Furthermore, we also find evidence that this positive association is weaker in case the focal firm is based in a region characterized by high policy integration.
Keywords: Upper Echelons Theory; Multinational Enterprise; Multinationality; Top Management Team; Nationality Diversity.
TABLE OF CONTENTS
INTRODUCTION ... 5
LITERATURE REVIEW ... 7
TMT diversity and firm multinationality ... 9
TMT diversity and firm performance ... 10
TMT diversity in complex environments ... 11
TMT nationality diversity ... 12
THEORY AND HYPOTHESES ... 14
TMT nationality diversity and firm multinationality ... 14
Moderating factors ... 16
The moderating role of firm’s global versus home region focus ... 16
The moderating role of the institutional diversity of MNE's home region ... 18
The moderating role of the policy integration of MNE's home region ... 20
METHODOLOGY ... 21
Sample and data collection ... 21
Variables ... 22
Dependent variable ... 22
Independent variable ... 23
Moderating variables ... 24
Control variables ... 25
Statistical Analysis and Results ... 26
DISCUSSION ... 33
Academic relevance ... 33
Managerial implications ... 35
Limitations and suggestions for future research ... 36
CONCLUSIONS... 38
AKNOWLEDGEMENT ... 41
REFERENCES ... 42
LIST OF TABLES
Table 1. Descriptive statistics: means, standard deviations and correlations ... 30 Table 2. Results of OLS regression ... 31 Table 3. Results of OLS regression - continuation ... 32
INTRODUCTION
Today’s global business environment has brought upon several complexities and surfaced the need to go beyond the generic advantages firms use in order to move ahead. Top management teams (TMT) are the ones making strategic decisions regarding the international environment firms have to face and therefore need to possess the gamut of capabilities that are a prerequisite for operating in such complex environments.
In the past several decades, international business scholars have explored the relationships between top management team demographic diversity and performance with demur, obtaining contradictory results (Hambrick, Davison, Snell and Snow, 1998; Kaczmarek and Ruigrok, 2013; Nielsen and Nielsen, 2013). The question of when the benefits of TMT diversity surpass the costs arises and a closer look at the specific circumstances when this applies becomes apparent. Although the TMT diversity can result in improved solutions in complex situations and cross-cultural competencies and learning, it can also lead to tensions and a disruption in dynamics.
Hence it is necessary to decompose TMT diversity and analyze the effects of specific TMT attributes on the strategic decision-making process or performance, as they will vary in magnitude, significance and direction. Moreover, the context in which TMTs operate is of great significance as it can influence the salience of certain attributes and enhance or diminish their effect. Although there has been substantial research done on the relationship between different characteristics of top management teams, (e.g. age, tenure, and international experience) and firm multinationality, the literature has paid relatively little attention to TMT nationality diversity. In other words the following questions still need an answer: Does a nationally diverse TMT contribute to the level of multinationality of a company? If so, how and under which circumstances?
Constructing on Hambrick and Mason’s (1984) upper echelon theory that specifies that the characteristics of managers influence the strategic decision-making process and firm performance, this study investigates the application of this theory on the relationship between a specific demographic characteristic of the TMT, i.e. nationality diversity, and the level of firm multinationality. A central study for this purpose is the meta-analysis of Kirca, Hult, Deligonul, Perryy and Cavusgil (2012) that aggregates the results of 145 studies and investigates multilevel factors, including executives’ characteristics, which affect firm multinationality. An important observation drawn from this meta-analysis is the absence of studies investigating a refined demographic characteristic of TMTs, such as the nationality diversity, and its salience regarding firm multinationality. While other studies have researched the effects of TMT international experience (Carpenter, Sanders and Gregersen, 2001; Carpenter, Pollock and Leary, 2003) or other TMT characteristic such as age, tenure, education level (Kirca et al., 2012; Tihanyi, Ellstrand, Daily and Dalton, 2000), only a few focused on TMT nationality diversity (Kaczmarek and Ruigrok, 2013; Nielsen and Nielsen, 2011 and 2013). In an increased globalized environment, TMT nationality diversity can trigger a higher degree of firm multinationality through a wider spectrum of distinct capabilities that can better tackle uncertain situations and the potential to lower intercultural barriers. Besides the academic relevance and the additions to the upper echelons literature, this study has also important managerial implications. Multinational enterprises (MNEs) could use the findings in order to develop recruitment strategies that would underline the overall firm strategy regarding the expansion into foreign markets.
This research is structured in the following way. First, we review the relevant literature and introduce the constructs that are of interest for the purposes of this study. We identify the TMT nationality diversity as an important deterministic attribute and, after the discussion of important contextual factors, develop our hypotheses. Subsequently, the methodology section
describes the data collection, the variables operationalized, and the method of analysis used to test the hypotheses. In the following chapters we present and discuss the findings that emerged in the analysis. Finally, we bring forward the academic and managerial implications of this study and mention the limitations and suggestions for future research.
LITERATURE REVIEW
The diversity of top management teams (TMT) has been in the spotlight of numerous studies in the past several decades, many of them drawing from the pioneering work of Hambrick and Mason (1984) on the upper echelons theory. Drawing from the behavioral theory of the firm, the upper echelons theory suggests that the decisions top executives make are bounded by certain limitations due to the nature of human beings and are not necessarily always the most rational decisions. Their bounded rationality, different aspirations and goals influence the strategic decision-making process and consequently firm performance. Hambrick and Mason (1984) use as a substitute for the behavioral and psychological attributes of the top executives their demographic characteristics, as these are easier to measure and evaluate. Decomposing the TMT diversity in different characteristics that have an effect on strategic decision making, firm performance or level of multinationality has led to a range of different approaches in the field. The results have been contradictory with regard to the effectiveness of heterogeneous teams and their relationship with firm performance, multinationality or other strategic decisions. Some empirical studies report mixed results relative to the significance of TMT heterogeneity on the aforementioned outcomes (Carpenter, 2002; Hambrick, Seung Cho and Chen, 1996), others find only marginal positive significance (Tihanyi et al., 2000) while a few report a stronger association (Bartel-Radic, 2006; Sambharya, 1996). The terms diversity (Kirca et al., 2012; Nielsen and Nielsen, 2013) and heterogeneity (Hambrick et al., 1996; Tihanyi et al., 2000) have been used to refer to the
range of distinct aspects and characteristics with regard to the TMT and will be used interchangeably in this study.
A pivotal study is the meta-analysis on the drivers of firm multinationality of Kirca et al.(2012) investigating different level factors such as individual, group and firm level. The findings bring cumulative evidence supporting the salience of executives’ characteristics in determining firm level of multinationality, thus reinforcing the upper echelon theory. They test two models that measure the effects of the TMT-level and firm-level factors. The group-level antecedents of firm multinationality that they measure are the TMT size, TMT international experience, TMT diversity and TMT tenure. Those at the firm-level are firm size, firm international experience, company age and human and relational capital, though the firm-level factors are not of interest for the present study. They arrive at the conclusion that not all characteristics of a TMT are significant in influencing the firm level of multinationality. The most important factor, international experience, seems to have the best explanatory power when it comes to the TMT capability to enhance firm multinationality through knowledge about complex and diverse environments, as obtained by Carpenter et al. (2001) and Carpenter et al. (2003). TMT size, diversity and tenure are less significant especially in the presence of international experience, as they cannot entirely explain their germane competences. Although they find limited explanatory power for the TMT diversity characteristics, this could be explained by the fact that they do not look at a specific measure of diversity but take a holistic view of the demographic characteristics. They do not specifically mention which diversity factors they are analyzing, whether it is nationality, age or others. This limitation suggests the need of a finer-grained study that analyses specific demographic characteristics of TMTs, such as the nationality diversity, as not all characteristics have the same effects or importance.
TMT diversity and firm multinationality
Tihanyi et al.’s (2000) study analyses a more specific demographic characteristic of TMTs, age, and shows that the composition and characteristics of the TMT have a significant influence on the firm’s level of multinationality as they affect the strategic decisions taken. Additionally to age, tenure, education and international experience are the characteristics they study and demonstrate that younger managers, greater tenure, elite education and greater international experience lead to higher levels of firm multinationality. However, when considering the heterogeneity of the TMT with regard to these characteristics, they find only marginal positive significance. This contradicts the previous findings of Sambharya (1996) who in light of the upper echelon theory shows a significant relationship between the TMT heterogeneity in international experience and the firm level of multinationality. Their contributions on the influence of the TMT with a certain composition regarding demographic factors on the strategic decisions and firm multinationality open up the door for further research on other TMT characteristics.
Furthermore, Rivas (2012) investigates TMT diversity and finds additional evidence that this relates to firm multinationality. Considering three categories of diversity, age, tenure and functional background, he finds support for the existence of a direct relationship but mixed findings in relation to the sign of these relationships when compared to previous research (Carpenter and Fredrickson, 2001; Tihany et al., 2000). TMT age diversity doesn’t influence the level of firm multinationality, but the diversity in tenure negatively relates to it, more homogenous teams proving to be more effective as they have been working together for longer. Functional background is the only characteristic that seems to have a positive impact as the members of the TMT can bring different perspectives and knowledge and can tackle issues in a more effective way. They also test for the age and tenure variable and confirm,
contrary to the expectations, Tihanyi et al.’s (2000) results concerning a younger age and longer tenure of TMT members for higher levels of firm multinationality.
TMT diversity and firm performance
In their study on airline companies, Hambrick, Seung Cho and Chen (1996) find that TMT heterogeneity has positive significance on firm competitive moves. By analyzing three characteristics, functional and educational background and tenure, they reveal a connection between the heterogeneity of the TMT with regard to these characteristics and the action or response propensity, magnitude and speed to competitive moves. Heterogeneous teams seem not to be so alert about competitive moves and have a slower response speed, but when they reply or take actions of their own, the magnitude of these actions is more significant than that of homogeneous teams. Overall, TMT functional, educational and tenure heterogeneity prove to improve performance and the competitive position of the firm.
In an attempt to reconcile the contradictory results of the effects of TMT heterogeneity on team performance, Carpenter (2002) uses firm multinationality as a proxy for the complexity of the business environment the TMT has to face. The results show that TMT heterogeneity is beneficial at low levels of complexity, but when the level of complexity is higher, TMT heterogeneity becomes harmful. Additionally, he finds evidence that the impact of TMT demographic heterogeneity is stronger in short-tenured teams, thus implying that the differences between TMT members, though important in the beginning, become less obvious over time.
Similar results concerning the influence of functional background diversity on firm performance can be observed in Cannella, Park and Lee’s (2008) study. They investigate the relationship between both intrapersonal functional diversity and dominant functional diversity and firm performance and find positive implications for the first one. Taking into
consideration the internal and external context, such as the TMT member co-location and the environmental uncertainty, they find empirical evidence suggesting that they positively
moderate the relationship between TMT functional diversity and firm performance. Co-locating the TMT improves team functioning and the environmental uncertainty seems to
be more salient when considering intrapersonal functional diversity. One of the contributions of this study is demonstrating that firm multinationality and globalization are not that easy to accomplish as dispersed TMTs have lower performance than co-located TMTs.
TMT diversity in complex environments
Extant research documents the need to consider the peculiar circumstances under which TMTs operate, the different configurations that could emerge, and the specific interrelationships resulting from such combinations. Hambrick, Davison, Snell and Snow (1998) investigate in a study the dynamics and possible benefits or pitfalls of multinational groups, including TMTs, and conclude that there are certain factors that highly influence their success. Group task and the dimensions of group heterogeneity, such as the diverse values, cognitions, demeanors and language, can guide the effectiveness of multiple nationality groups and prove to be superior to single-nationality groups only in situations where their varied characteristics can bring solutions to complex company issues.
Although not the focus of this paper, group theory presents some links to the benefits of intercultural teams. As already mentioned, the effects of cultural diversity have been shown mainly to be insignificant unless we take into account the context in which this diversity is considered. Cultural diversity in global teams leads to intercultural learning and acquisition of intercultural competences which are valuable when doing business in culturally diverse environments (Bartel-Radic, 2006). In TMTs this cultural diversity could generate conflicts when making strategic decisions, however such conflicts could also be beneficial in leading to better solutions.
While at the individual level new members in TMTs seem to have the same demographical characteristics as the rest of the team due to the similarity-attraction mechanism, the strategic complexities and the industry characteristics can overwrite this (Nielsen, 2009). Factors such as the firm’s strategic complexity and industry characteristics have a positive impact on the degree of heterogeneity of TMTs in terms of nationality diversity, international experience, educational background and industry experience. This is mostly due to the fact that a more diverse environment and highly complex firm operations require different configurations of TMTs that have superior information-processing capabilities. Similar to Bartel-Radic’s (2006) findings of building dynamic competencies and resources through the use of intercultural teams, they confirm that TMT cultural heterogeneity can be used as a strategic advantage. When operating in foreign countries, it can bring more diverse and innovative solutions when confronting unknown barriers.
TMT nationality diversity
The study of Nielsen and Nielsen (2013) focuses on a specific dimension of TMTs, the nationality diversity, and investigates the relationship between this construct and firm performance. They find that nationality diversity is one of the most important attributes that has a long lasting effect on performance, unlike functional or educational diversity or international experience which tend to diminish over time. They also study the moderating role of team tenure, firm internationalization and industry munificence on the relationship between TMT nationality diversity and firm performance and find significant positive results. However, when testing the impact of the moderators on the relationship between the other diversity factors and firm performance they find negative or insignificant effects.
Another study by Nielsen and Nielsen (2011) on the effects of TMT international experience and nationality diversity on the choice of entry modes shows as well how different attributes of TMTs have different consequences in international decision-making. While international
experience is found not to have a significant importance in the choice of entry modes, it is however negatively related to a choice for shared-control vs. full-control modes. Moreover, as nationality diversity hasn’t been studied in relation to this outcome, here it is shown to have significant and relevant impacts on the choice of entry modes, with an inclination towards shared-control rather than full-control modes. This study brings additional evidence in favor of the salience of analyzing different characteristics of TMT separately as they can have disparate effects on international decision-making.
Reinforcing the importance of the TMT nationality diversity construct, Kaczmarek and Ruigrok (2013) correlate this characteristic with firm multinationality and find empirical results suggesting that the relevance of such diversity becomes more obvious in challenging situations regarding internationalization. Only when there is a proper environment that demands a mix of cultural aware members of the TMT, such diversity can be justified otherwise coordination problems arise. Consequently they do not find a direct relationship between TMT nationality diversity and firm performance, as opposed to Nielsen and Nielsen (2013), but confirm the necessity to look at the strategic and social context in which the two constructs are considered.
Although there have been numerous studies done on the relationship between different TMT characteristics and either strategic decisions, such as entry modes, or firm performance or level of multinationality, the literature reviewed presents results that remain contradictory. There isn’t an exact appreciation of the most salient characteristics to consider and the degree to which their heterogeneity can positively increase their value remains relatively unclear. While Nielsen and Nielsen (2013) and Kaczmarek and Ruigrok (2013) concentrated on TMT nationality diversity they mostly studied its relationship with firm performance. Other studies, though researching the association between certain TMT characteristics such as age, tenure (Tihanyi et al., 2000; Rivas, 2012), international experience (Carpenter et al., 2001;
Carpenter et al., 2003; Sambharya. 1996) or diversity as a holistic measure (Kirca et al., 2012), and firm’s level of multinationality, have not focused explicitly on the aspect of TMT nationality diversity. This gap constitutes the main focus of this thesis.
As already mentioned and noticed by previous researchers, the relationship between various TMT characteristics and different organizational outcomes such as firm level of multinationality is not necessarily straightforward. This relationship can be influenced by several factors that could strengthen or weaken it. The moderating factors that this study will focus upon are the global versus home region orientation of the firm, the institutional diversity, and the policy integration faced by the focal firm in the home region.
THEORY AND HYPOTHESES
TMT nationality diversity and firm multinationality
The nationality of the members of a group draws its deterministic nature from the country and institutional environment in which the individuals spent the majority of their formative years and has implications regarding the group functioning. Certain characteristics of individuals such as their values, cognitive schemas, demeanor and language are correlated to their nationality and to the way they perform different tasks or respond to risks (Hambrick et al., 1998). The specific values attributed to the members of a certain nationality are considered to be deeply rooted and established since childhood, matching a system of collectively held beliefs and values (Hofstede, 1980). These differences in values that originate from the diversity in national cultures affect behavior in task groups and allow for individual-level differences to be used as a predictor of ethnic diverse group behavior (Cox, Lobel and McLeod, 1991). The cognitive schemas refer to the diverse knowledge structures and assumptions that characterize the group members of different nationalities. These schemas are underpinned by the knowledge and experience accumulated, guiding their
decisions and shaping their information processing models (March and Simon, 1958; Dearborn and Simon, 1958, in Hambrick et al., 1998). Diversity in demeanor, such as body language, outward behavior, and the degree to which the working language is shared to a high or low extent by the members of the group are showed to affect group functioning only when certain tasks are to be completed (e.g. coordinative tasks).
TMTs in MNEs have to take decisions pertaining to international matters which incorporate dealing with various social, political, legal and economic environments with different levels of complexity and uncertainty. Diverse teams can tap into their various information networks and enlarge the pool of knowledge available for problem solving and enhancing group effectiveness (Maznevski and Athanassiou, 2006). Therefore, the tacit knowledge of individual members can be transferred to the group and the organization and, for instance, support expansion into foreign markets (Winter, 1987, in Athanassiou and Nigh, 2002). Some managers are even appointed due to their different nationality and the knowledge they hold about the specific environment in their home country or region, thus bringing additional value in the TMT which is seen as “a self-contained international business advice network” (Athanassiou and Nigh, 2002: 161).
Scholars have been interested in a variety of TMT characteristics and team’s heterogeneity effects on different organizational outcomes, such as the level of firm multinationality. Characteristics such as age, tenure, education, international experience, functional background and the TMT’s heterogeneity regarding them have been vastly studied (Carpenter, 2002; Hambrick et al., 1996; Rivas, 2012; Sambharya, 1996; Tihanyi et al., 2000) and linked to firm multinationality.
When talking about the heterogeneity of culturally diverse groups, the tenure of the group can influence its performance as teams that have been working together for longer periods had
time to adjust to the nationality differences and coordinate their specific skills and competences better. They are better able in the long-run to overcome the initial limitations and develop solutions to the identified problem perspectives (Watson et al., 1993).
Group theory also presents insights relative to the intercultural learning processes that occur in culturally diverse teams and the benefits that can be accrued when doing business in complex environments (Bartel-Radic, 2006). Similar results were found by Nielsen (2009) which sees TMT cultural heterogeneity as a strategic advantage that can help overcome barriers and move past some of the cultural impediments that would inhibit a company’s ability to perform well in different countries. As the level of multinationality of a firm increases, the number of distinct international markets that the firm will be exposed to increases as well. Therefore the challenge posed by the increasing complexity of the firm’s environment must be faced by the managers (Kirca et al., 2012). Knowing that complexity and uncertainty are two defining characteristics of strategic decision making, TMTs need to be equipped with the right combination of diverse competences, skills and values, offered by the diversity in nationality, in order to be able to maximize the quality of their decision making power. Consequently this will affect the level of firm multinationality, leading to the first hypothesis:
Hypothesis 1: TMT nationality diversity is positively related to the level of multinationality of the firm.
Moderating factors
The moderating role of firm’s global versus home region focus
The benefits of TMT nationality diversity will vary according to the complexity of the environment it is facing and the challenging decisions and situations it needs to tackle. Under certain circumstances the diversity of the TMT regarding the nationality of its members can
be justified when facing uncertainty, risk and unusual situations (Hambrick et al., 1998; Kaczmarek et al., 2013). In their study of the regional and global strategies of Japanese firms, Delios and Beamish (2005) find that although a global strategy proves to be more efficient and improves performance of firms more than for those firms pursuing a home-oriented strategy, the later one is still prevailing. One of their assumptions is that managers are more reluctant to go outside of their home region where they don’t find as much resistance. This impediment could be overcome by having a TMT that is more prone to taking risks and better equipped for dealing with an environment outside their home-region.
Though some scholars find the need of a gradual internationalization for the “strategies, structures, processes and management capabilities (to) realize a good match so that high managerial effectiveness can be achieved” (Tushman and Romanelli, 1985, in Li, 2005: 41). They argue that at a certain level of internationalization, the managerial effectiveness will be compromised also due to the high liability of internationalization and costs such as lack of familiarity with the local culture, local governments and relevant information about the foreign environment. In a similar vein, Banalieva & Santoro (2009: 347) reiterated that at high levels of multinationality “managers must obtain and update different types of knowledge about many more diverse global environments resulting in information overload and added costs”. What they don’t take into account is the diversity in the composition of the managerial team, which could actually be more prolific in such a complex and demanding environment that requires distinct capabilities and comprehensive knowledge.
Countries that are a part of the same region are more similar to each other and have a stronger cross-border integration if they belong to the same regional trading block, as a result of common currency, geographic proximity, similar institutional environments (Frankel and Rose, 2002, in Asmussen, 2009). Thus operating in different regions would call for different strategies, structures and managerial capabilities (Rugman and Verbeke, 2008). This would
entail different configurations for TMTs of companies operating only in their home region compared to those operating globally. Therefore we expect that a global orientation would implicate a higher impact of a nationality diverse TMT on the level of firm multinationality:
Hypothesis 2: Ceteris paribus, firm's global (versus home region) focus positively moderates the relationship between TMT nationality diversity and firm's level of multinationality.
The moderating role of the institutional diversity of MNE's home region
The institutional profiles of the countries in which MNE’s operate can vary substantially (Kostova and Roth, 2002). Previous research suggests that managers are aware of the diverse institutional-cultural environments and take them into consideration when expanding abroad, however they also need to have “a triad consisting of a global mindset, knowledge and skills […] to be globally competitive” (Kedia and Mukherji, 1999: 234). Extant research confirms that MNEs “prefer to locate foreign operations in host countries that are more ‘proximate/similar’ to their home country” (Flores & Aguilera, 2007: 1193) in order to decrease uncertainty and risk. DiMaggio and Powell (1983) talk about isomorphism as a force that drives homogenization of organizations in a certain environment and molds the organizational characteristics according to the environmental diversity. Though for homogenization to take place the individuals that manage those organizations must have the necessary capabilities to understand the differences between environments and the need to adapt.
As the institutional environment of MNE’s home region becomes less harmonized and more complex, the pressures on the MNE become more cumbersome and requirements for restructuring the organization arise if they want to maintain their legitimacy abroad (Banalieva et al., 2012). The diverse political systems drive up the costs of operating abroad and increase uncertainty in the relationship between MNEs and governments due to a poor
prediction in the reactions of local governments and/or other organizations (Dow and Karunaratna, 2006).
Knowing that MNEs most often make decisions not only on a country-by-country basis, but have a wider perspective, it is appropriate to take a regional approach in this study. “Institutional diversity at the regional level is the variation in the institutional environments across the countries within the home region.” (Banalieva and Dhanaraj, 2013: 96). A differentiation can be made between the multiple domains of the institutional environment, such as the regulatory, the cognitive and the normative domain drawn from Scott’s (1995) three pillars of institutional environments, and the multiplicity of institutional environments faced by MNEs that vary across national environments (Kostova and Zaheer, 1999). This complexity of the diverse institutional environments increases the costs of processing information and coordinating activities across boarders and presents more risks when making strategic decisions (Banalieva and Dhanaraj, 2013; Kostova and Zaheer, 1999).
Furthermore, when MNEs decide to expand in foreign countries they have to go through a “process of search and deliberation” (Rangan, 2000: 206, in Banalieva and Dhanaraj, 2013: 96) which entails two types of costs. Although expanding in the home region would reduce the search costs of possible partners due to spatial proximity, institutional diversity would increase the deliberation costs of assessing these partners as there would be a lack of common characteristics between them (Rangan, 2000, in Banalieva and Dhanaraj, 2013).
Although MNEs could use the knowledge they learned in one country and apply it to other similar countries in the home region, thus decreasing costs and building competitive advantages, this rationale is not valid anymore in situations when the institutional environment in the home region varies substantially. This will impede MNEs from using their current knowledge and would therefore steer them towards global markets (Banalieva and
Dhanaraj, 2013). By having a competent nationality diverse TMT that is better able to deal with global markets, MNEs would be more prone to increasing their level of multinationality outside of the home region when this is characterized by a high institutional diversity. Accordingly:
Hypothesis 3: Ceteris paribus, the institutional diversity of MNE's home region positively moderates the relationship between TMT nationality diversity and firm's level of multinationality.
The moderating role of the policy integration of MNE's home region
Regional integration is an essential factor to be taken into consideration due to the spread of regional trade agreements (RTAs) that usually match the home regions of many MNEs. RTAs are “intergovernmental treaties through which signatory countries agree to more advantageous conditions in the conduct of their mutual trade and investment relationships than those conditions applied to other, non-signatory partners” (Cuervo-Cazurra and Un 2007: 227, in Banalieva et al., 2012: 494) and lead to harmonization of policies across member countries.
Some RTAs have higher levels of policy integration, such as the European Union (EU) and succeeded in reaching a common ground in some key areas such as “product standards, market regulations, trade, fiscal, and monetary policies” (Banalieva et al., 2012: 501). This allows for reducing costs, due to a standardization of product offerings and a use of similar distribution channels, and receiving preferential treatment in comparison with the non-member countries. These policies are designed in such a way that they would appeal to a higher number of RTA members in order to promote trade diversion and trade creation inside the RTA (Freund and Ornelas, 2010).
On the other hand, having less harmonized policies could in fact harm the member countries by increasing the uncertainty and risk they have to face and instigating disagreements between the RTA members (Anson, Cadot, Estevadeordal, De Melo, Suwa-Eisenmann and Tumurchudur, 2005). As an illustration we can consider the rules of origin, i.e. the regulations that ensure the inner-RTA provenance of products and the applicability of duty free treatment. If the rules of origin differ among member countries of an RTA, this could lead to an increase in the costs of production, administration, and bookkeeping (Holden, 2003, in Banalieva et al., 2012).
If policy integration is not achieved inside an RTA, then MNEs fail in proving their legitimacy in the home region due to the uncertain complex institutional environment (DiMaggio and Powell 1983). However, if harmonization of policies is realized within the home region, MNEs would be more willing to expand their operations mostly there, and thus avoid expanding in less familiar markets. Hence, we hypothesize:
Hypothesis 4: Ceteris paribus, the policy integration of MNE's home region negatively moderates the relationship between TMT nationality diversity and firm's level of multinationality.
METHODOLOGY
Sample and data collection
The sample consists of MNEs from the Fortune Global 500 2013 list which ranks corporations worldwide on the basis of their revenue. This is the most recent published list and uses the revenues from the 2012 financial year as numerous companies have not rolled out their 2013 annual report. Additionally, this study does not focus on firms from specific countries unlike previous research on TMT nationality diversity that concentrated on Swiss
firms or companies from the Netherlands and United Kingdom (Kaczmarek and Ruigrok, 2013; Nielsen and Nielsen, 2013). This will allow for a better generalizability of the results on companies worldwide as no issues of country specific bias will influence the data. Also most of these companies are operating internationally which allows for the measurement of the extent of firm multinationality.
The firm specific data was obtained from ORBIS, a comprehensive database with information on companies worldwide provided by Bureau van Dijk, from companies’ annual report and information generated from internet searches. Some companies from China or Japan, for instance, don’t have annual reports written in English so this limits the sample to those companies with available information. The companies for which enough information couldn’t be gathered were discarded, arriving at a final sample consisting of 362 MNEs.
The categorization of countries according to specific regions was accomplished through the use of the United Nations country classifications as done in prior research (Banalieva and Dhanaraj, 2013) and presented in Appendix A. Data from the Fraser Index of Economic Freedom of the World published by the Fraser Institute was used for measuring the regional institutional diversity (Banalieva and Dhanaraj, 2013). The components of the index are described in Appendix B.
Variables
Dependent variable
In order to capture the level of firm multinationality we rely on the measure of foreign sales divided by total sales (FSTS) (Carpenter, 2003). Robustness checks for other measures of firm multinationality are also conducted in order to further ensure the validity of our results. In particular, the two other proxies that are used are the ratios of foreign assets to total assets (FATA) and of the number of foreign equity affiliates to the total number of equity affiliates.
The FSTS measures the degree to which the company is dependent on foreign markets, whereas FATA measures the degree of dependence on foreign production. The last ratio proxies the geographic dispersion of companies and thus denotes a scope measure. Prior studies have used these three ratios and also integrated them in a composite measure derived from Sullivan’s (1994) index (Carpenter et al., 2001; Kaczmarek and Ruigrok, 2013; Rivas, 2012). Other measures used in previous research focus on regional sales (Hitt et al., 1997), employee data (Kim, Hwang, and Burgers, 1989; 1993) or country scope (Tallman and Li, 1996; Tihanyi et al, 2000).
Independent variable
The top management team (TMT) is delimited as the executive team consisting of the top-tier executives listed in the company’s annual report (Kaczmarek and Ruigrok, 2013). TMT nationality diversity is the independent variable that correlates with the level of firm multinationality, as assumed in this study. The executives’ nationality was determined by the information available in the annual reports or other secondary sources of information. The information on the nationality of the TMT members is not categorized by country but by region, therefore we measure a regional TMT diversity. This is still relevant as we posit that the greatest differences among managers occur in case they come from different regions rather than countries. This regional diversity is captured through a Blau index (Blau, 1977) by applying the formula: B = [1 - ∑(pi)2], where p is the percentage of members in the ith region.
The higher the value of B, the higher the diversity of the TMT. This methodology is the standard in several studies on TMT diversity (Carpenter, 2002; Finkelstein and Hambrick, 1996; Nielsen, 2010). Furthermore, the number of different nationalities represented in the TMT will also be used as a proxy for this variable, thus ensuring that results are confirmed at both levels of analysis.
Moderating variables
Going further into studying the relationship between the TMT nationality diversity and the level of firm multinationality, we identified three moderating variables that influence this relationship in different ways.
The firm’s global versus home region orientation is the first moderating variable. The extent to which a company has a global focus, rather than concentrating on the home region, is captured through the percent of sales in foreign markets. A similar approach to Delios and Beamish (2005) is followed and a 20 percent and a 10 percent criteria is used for defining whether a firm has a global (tri-regional) focus or not. Six regions were used to divide the world (North America, Europe, Asia, South America, Oceania and Other regions) instead of the three triad regions (Rugman and Verbeke, 2003). The 10 percent cut-off point was used due to this division in six regions that yields a different even distribution of sales. For a firm to have a global focus it should have three regions of the world each account for at least 20 percent of its foreign sales.
Another moderating variable is the institutional diversity of the MNE’s home region, measured through the Fraser index of Economic Freedom of the World published by the Fraser Institute. The index is composed of several other sub-indexes: government, legal, economic and regulatory (Gwartney at al., 2013). First I divided the countries from the Fraser index into categories corresponding to their home regions as denoted by the United Nations (UN) geographic based country mappings (Banalieva and Dhanaraj, 2013). Afterwards the coefficient of variation of the Fraser index was computed for the home region of the firm by excluding the home country of the respective MNE. The coefficient of variation is calculated by dividing the standard deviation of the distribution to its mean (Banalieva and Dhanaraj,
2013). The higher the values of the coefficient, the higher the diversity in the institutional environment.
The regional policy integration of the MNE’s home region is captured by the inclusion of the MNE’s home country in a regional trade agreement (RTA) characterized by a specific level of coordination. We use a category variable, assigning 0 to Other, 1 to APEC, 2 to NAFTA and 3 to EU (Banalieva et al., 2012). With higher levels of policy coordination between the countries in a RTA, higher values of regional policy integration are derived.
Control variables
TMT size was controlled as the number of key executives that are members of the executive board (Carpenter 2003). This variable is controlled for frequently as the available managerial resources in the TMT can affect its performance and efficiency. As the number of members increases coordination and communication can become more cumbersome (Hutzschenreuter, 2013). The information on the size of the TMT was gathered from the firm’s annual report. Diversity in terms of gender was controlled for under the TMT gender variable. As differences in behavior cannot be explained merely by the nationality diversity, accounting for the number of males and females is essential. Gender is a good predictor of behavioral diversity and therefore we include it as a control variable (Cannella et al., 2008).
At the firm level, there are three factors that are controlled for. Firm size is a commonly used variable due to its impact on firm performance (Gomez-Meji and Palich, 1997). The size of the firm influences its behavior and response to environmental situations. Larger firms are more prone to international activity and better equipped to dealing with uncertainty. However, smaller firms can be more flexible and open to change (Tihanyi et al., 2000). Gross annual revenue in millions of US dollar is used to measure the size of the firm. In this study the logarithm of the variable is extracted in order to account for skewness in the data. The
second control variable at the firm level is the firm age. The relationship between certain characteristics of executives and firm outcomes has been shown to be influenced by this variable (Miller, 1991). The number of years since the firm was founded until the year of reference for this study, 2012, is used to measure the firm age. Firm multinationality has been argued to relate to firm performance as MNEs that are performing well have the necessary resources to expand internationally (Geringer, Beamish, and daCosta, 1989; Hitt et al, 1997; Kim et al, 1989). They are also more stable and more inclined to take risky decisions. Firm performance is operationalized as return on assets (ROA) at the end of 2012 financial year.
The last control variable is at the industry level as firms from different types of industries can have distinct reasons to extend internationally and can behave differently. The industry effects were captured with a categorical variable from 1 to 5 which was then coded into 4 dummy variables. The 5 industry categories were developed by aggregating the prominent industries from the Fortune Global 500 2013 list. The SIC codes were used to identify the pertaining industry segments of the MNEs and the final categories are: 1 – mining, utilities and construction, 2 - automotive and machinery, 3 – professional and information services, 4 – wholesale and retail, 5 – manufacturing and other services.
Statistical Analysis and Results
The descriptive statistics and correlations between variables are presented in Table 1. The point of interest was whether any of the correlations have a value above .7 which would indicate the presence of problematic constructs (Pallant, 2011). After testing for multicollinearity and evaluating the bivariate correlations, all variables were retained as they have values below .7.
The mean value of the firm multinationality variable is .45 which suggests that on average the MNEs surveyed are quite international. Hence the sample is suitable for this study. When
looking at the global focus of the firm we notice that 6% of the firms have a global orientation when using the 20% cut-off point. When using the 10% cut-off point the percentage of firms that have a global focus increases to 20%. The descriptive statistics of the TMT indicate that its nationality diversity has a mean of .14, although it is important to underline that this variable was measured at a regional rather than country level. The average number of different nationalities represented in the TMT is of 2.09 and the typical TMT has approximately 10.78 members. Of these, 90% are male executives. As the values of the regional policy integration variable range from 0 to 3, its mean is of 1.86.
For testing the hypothesis a hierarchical regression analysis was used in order to determine the successive relationships with the explanatory variables. As the dependent variable has continuous values and we assume a linear relationship between the dependent and the explanatory variables, the most suitable model is the Ordinary Least Squares (OLS). For testing the impact of the moderator variables, interaction terms were used by computing the product between the independent and each moderating variable after standardizing them to alleviate potential multicollinearity issues. The regression analysis was performed in two steps. First, the control variables were introduced in order to observe their effects on the dependent variable. Afterwards the independent variable and then the interaction terms were added one at a time with the purpose of observing their additional explanatory power. Accordingly, we run a total of nine models. In the first model only the control variables were used. Models 2 and 3 test the first two hypotheses. Models 4-8 assess the third hypothesis by using the Fraser composite measure and subsequently checking for the individual sub-components’ effect on the dependent variable. Finally, model 9 tests the last hypothesis. Table 2 and 3 present the results of the hierarchical regression analysis. When analyzing these results there are three values that need to be looked at. The significance indicates whether the results are reliable or not and thus if they can support the hypotheses. The Beta
standardized coefficient accounts for the change in the dependent variable according to the explanatory variables. Its sign describes the direction in which the change occurs. Furthermore, the R2 measures the goodness of fit of the model and the degree to which it
actually explains the variance.
The results of the hierarchical regression analysis indicate that not all of the included factors have additional explanatory power as their significance is not below the .05 threshold. Hypothesis 1 states that TMT nationality diversity is related to firm multinationality. The coefficient of TMT nationality diversity is significant (b=.257, p=.000) and positively related to firm multinationality as expected. Thus strong support for hypothesis 1 is found. The R2
improves from .181 in the initial model to .243 after including the independent variable, which shows that this model fits the data better. The results are confirmed when using the foreign production measure or the geographic dispersion measure for the firm multinationality variable. When testing for the first interaction term, model 3 does not bring any support for hypothesis 2. The global focus variable has a positive significant effect on firm multinationality (b=.141, p=.068), but the interaction term is insignificant. The results are again insignificant when testing with the different measures for firm multinationality or when using the 10% cut-off point for measuring global focus. Therefore no significant positive moderating effects are found for the global focus of the firm variable on the relationship between TMT nationality diversity and firm multinationality. Hypothesis 3 is not supported as insignificant results are found for the effects of the interaction term between TMT nationality diversity and institutional diversity of the home region. Models 5-8 replace the Fraser composite index with each of its sub-components. All models have consistent results with the previous analysis. Only significant direct effects of the institutional diversity variable and of its sub-components are found (p=.000 in all cases). With the exception of the government sub-component, all of them have a negative effect on firm multinationality. The
last model supports hypothesis 4. While the effect of policy integration on firm multinationality is highly significant (b=.332, p=.000), the interaction term between TMT nationality diversity and policy integration is significant at the 10% level (b=-.148, p=.076). Slightly stronger results are obtained when using the geographic dispersion measure for the firm multinationality construct with a significance of the interaction term of .018 and a coefficient with a -.217 value. Consequently, the results suggest that the policy integration in the home region negatively moderates the relationship between TMT nationality diversity and firm multinationality. All the results are confirmed when repeating the entire analysis using the number of different nationalities of the executives as a proxy for the TMT nationality diversity variable.
As for the control variables, strong support is found for the firm size (b=.128, p=.012) and firm age (b=.167, p=.001) variable. These results suggest that company characteristics and resources have an influence on the degree of international expansion a firm engages in. Two industry dummies seem to be significant. The automotive and machinery industry and the wholesale and retail industry are significant at a .003 and respectively .013 level in comparison to the mining, utilities and construction industry of reference. Therefore the belonging to these two groups of industries impacts the level of firm multinationality.
Table 1. Descriptive statistics: means, standard deviations and correlations Variable Mean S.D. 1 2 3 4 5 6 7 8 9 10 1. Firm multinationality 0.45 0.31 2. TMT nationality diversity 0.14 0.20 .31** 3. Global focus 0.06 0.23 .25** .16** 4. Institutional diversity 0.10 0.03 -.44** -.11* -.05 5. Policy integration 1.86 0.93 .30** .01 .06 -.66** 6. TMT size 10.78 5.59 -.05 .06 -.04 .20** -.12* 7. TMT gender 0.90 0.11 .11* -.08 -.04 -.04 -.15** -.14** 8. Firm size 4.66 0.32 .15** .03 .004 -.06 .07 .14** -.06 9. Firm age 65.42 52.57 .16** .18** .04 -.16** .17** .08 -.07 .03 10. Firm performance 5.27 6.73 .09 .03 .13** .07 -.08 .03 -.05 .14** -.06 11. Industry effects 3.16 1.38 -.02 .06 .002 .10 -.04 -.03 -.08 -.21** .02 .03 * p < .05, **p <.01, *** p < .001
Table 2. Results of OLS regression
Dependent variable: FSTS
Control variables Beta Sig. Beta Sig. Beta Sig. Beta Sig.
TMT size -.072 .145 -.084 .081† -.073 .124 .020 .647
TMT gender .086 .083† .104 .030* .113 .018* .090 .032*
Firm size .128 .012* .124 .012* .129 .009** .089 .038*
Firm age .167 .001** .124 .010* .122 .010* .045 .290
Firm performance .070 .168 .061 0.215 .042 .385 .097 .023*
2. Automotive and machinery .191 .003** .178 .005** .151 .017* .316 .000***
3. Professional and information services -.092 .212 -.105 0.138 -.095 .175 -.036 .568
4. Wholesale and retail -.156 .013* -.129 .035* -.126 .036* -.049 .362
5. Manufacturing and other services .140 .047* .106 .119 .097 .149 .216 .000***
Independent variable
TMT nationality diversity (Blau index) .257 .000*** .234 .000*** .192 .153
Moderator variables
Global focus (20% criterion) .141 .068†
Institutional diversity in the home region -.469 .000***
Policy integration in the home region
Interaction terms
TMT nationality diversity × Global focus (20% criterion) .013 .874
TMT nationality diversity × Institutional diversity (FRASER) .014 .914
Constant -.397 .158 -.446 .099 -.497 .065 .191 .429 R2 Adjusted R2 Change in R2 p†<0.10; *p<0.05; **p<0.01; ***p<0.001. .181 .061 .082 .248 .181 .243 .263 .430 .160 .221 .238 .410 Controls H1 H2 H2
Table 3. Results of OLS regression - continuation
Dependent variable: FSTS
Control variables Beta Sig. Beta Sig. Beta Sig. Beta Sig. Beta Sig.
TMT size .018 .675 -.034 .442 .017 .695 .016 .705 -.025 .578
TMT gender .077 .065† -.001 .975 .114 .007** .060 .156 .161 .000***
Firm size .083 .053† .117 .009** .086 .049* .097 .025* .097 .035*
Firm age .045 .287 .088 .047* .042 .328 .054 .198 .069 .131
Firm performance .101 .019* .126 .006** .086 .047* .109 .012* .092 .046*
2. Automotive and machinery .305 .000*** .258 .000*** .312 .000*** .318 .000*** .226 .000***
3. Professional and information services -.026 .679 -.040 .535 -.041 .510 -.029 .637 -.074 .264
4. Wholesale and retail -.041 .440 -.045 .421 -.058 .284 -.038 .480 -.087 .129
5. Manufacturing and other services .214 .000*** .191 .003** .209 .001** .225 .000*** .137 .032*
Independent variable
TMT nationality diversity (Blau index) .407 .116 .017 .945 .224 .089† .119 .565 .391 .000***
Moderator variables
Institutional diversity (FRASER-GOV) .466 .000***
Institutional diversity (FRASER-LEGAL) -.399 .000***
Institutional diversity (FRASER-ECON) -.457 .000***
Institutional diversity (FRASER-REGUL) -.468 .000***
Policy integration .332 .000***
Interaction terms
TMT nationality diversity × Institutional diversity (FRASER-GOV) -.199 .442
TMT nationality diversity × Institutional diversity (FRASER-LEGAL) .231 .337
TMT nationality diversity × Institutional diversity (FRASER-ECON) -.020 .877
TMT nationality diversity × Institutional diversity (FRASER-REGUL) .089 .666
TMT nationality diversity × Regional policy integration -.148 .076†
Constant -1.128 .000 .550 .042 .123 .612 .486 .052 -.672 .009 R2 Adjusted R2 Change in R2 p†<0.10; *p<0.05; **p<0.01; ***p<0.001. .410 .351 .400 .407 .318 .248 .191 .239 .245 .160
Model 5 Model 6 Model 7 Model 8 Model 9
.429 .373 .420 .426 .341
DISCUSSION
The empirical findings and conceptual framework add to the upper echelons theory and bring interesting implications regarding TMT’s role in influencing the level of firm multinationality. While previous research has concentrated and found support mainly for factors that pertain to the professional experience of the top executives, such as international experience, this study brings strong support for the salience of TMT nationality diversity in improving the level of firm multinationality. One factor pertaining to the home region of the MNE, specifically the policy integration, was shown to have a negative moderating effect on the relationship between TMT nationality diversity and firm multinationality. For the other two factors, global versus home region orientation and institutional diversity of the home region of the firm, no evidence of a significant moderating effect was found. However all variables have a significant direct effect. The global orientation of a firm was, as expected, positively associated to firm multinationality as firms that are already operating in foreign diverse markets are more likely to report a higher level of internationality. The institutional diversity in the home region suggests that companies would rather diversify globally than stay in their home region and this could open a window towards more international markets. The Government sub-component of the FRASER index measuring institutional diversity is the only one that has a positive effect on firm multinationality. This could be attributed to the fact that diversity in government expenditure and marginal tax rates, captured through this sub-component, can drive companies to seek more attractive host environments. The academic relevance, managerial implications, limitations and suggestions for future research will be discussed below.
Academic relevance
The present research broadens the upper echelons literature by taking a more in depth perspective and analyzing a more refined characteristic of top managers, respectively their
nationality diversity, in relation with firm multinationality. Despite the fact that leading MNEs have increasingly relied on non-national top tier executives, the upper echelons stream of research has not directed its attention towards this area. The implications nationality diversity has on the firm’s expansion into foreign markets were not investigated in the academic field. Therefore by turning the attention towards this demographic characteristic, the study has important academic relevance and managerial implications.
As shown in Kirca et al. (2012) study, diversity was not seen as a significant characteristic of the TMT that would influence firm multinationality. However, this was previously analyzed as a holistic construct that would not make a differentiation between specific demographic characteristics, but assumed they are all equal. Hence this study extended the literature and filled the identified gap by extracting nationality diversity as an explicit attribute of the TMT and bringing evidence of a positive relationship with firm multinationality.
Geletkanycz (1997) studies the role cultural values play in shaping top managers’ strategic and leadership mindsets. He shows that there is a significant relationship between the executives’ cultural values and the degree to which they are more prone to changing the company’s status quo. Compared to the professional experience effects, the embedded values of the managers proved to be of the same importance. Furthermore, even after socialization between members of the executive team, “the imprint of their cultural heritage is not suppressed”(Geletkanycz, 1997: 628) but remains a significant factor in determining strategy and appropriate leadership styles. The present study therefore reinforces Geletkanycz’s (1997) findings and adds nationality diversity as an important ingredient in the mixture of TMT attributes especially when pursuing an extensive multinational expansion.
Some of the inconsistent findings in the upper echelons literature regarding diversity can be attributed to the decontextualization of the TMT. This runs the risk of omitting the effects of
different level factors, such as firm, industry and country effects, on the salience of various TMT characteristics (Kirca et al., 2012). Knowing that diversity is a double-edged sword, it is important to look at the environmental factors that could influence the TMT nationality diversity-firm multinationality relationship. The strength and direction of the relationship can change in various environmental configurations. More specifically, the conditions in the firm’s home region regarding the policy integration can diminish the effects of the TMT nationality diversity on firm multinationality. Therefore the mixed previous findings regarding the benefits of TMT diversity on organizational outcomes can be attributed to a disregard of the environmental conditions.
Managerial implications
The present study conveys important managerial implications as well. First, multinational companies that are pursuing an expansion strategy into international markets can use the findings of this research in order to make more appropriate hiring decisions regarding the top tier executives. They can formulate a more efficient recruitment strategy that fits the overall firm strategy better. More specifically, companies can focus on attracting and hiring diverse non-national executives that possess different values and characteristics that can facilitate the foreign expansion of the company. Using their varied cognitive schemas and information networks can benefit the firm when having to deal with uncertainty and risky unknown environments outside their home country. On the other hand, if the firm is not pursuing an international strategy and wants to focus only on the domestic market, the results of the present study show that it makes most sense to hire executives of the same nationality in order to avoid disruption and conflict.
Second, when taking such decisions, companies need to observe the surrounding environment as well and weigh in their options. Our findings suggest that the policy integration of the home region diminishes the effect of the TMT nationality diversity on firm multinationality.
Therefore, firms should analyze the situation more in depth and observe whether diversifying their top executive team by hiring more international managers would add more value or not. If the policy integration is quite high in the firm’s home region, this fundamental decision should be taken with more precaution.
Third, companies can use the current findings to analyze the future moves of their competitors. By looking at the composition of the TMT of the other competing firms and at possible changes and additions of international members, they could anticipate whether their rivals are considering to further increase their degree of multinationality. This is vital in today’s highly competitive business arena where first mover advantages can place a company in the lead. With globalization tearing down barriers, companies can move more freely and extend into more foreign markets which make first mover advantages quite significant.
Limitations and suggestions for future research
This study has also limitations that need to be acknowledged. Although nationality is an important construct, there are other factors that influence individuals’ values and cognitive schemas as well. Some of these factors could be situations where an individual has dual nationalities or when the nationality does not match the country of his/her upbringing. Also, the specific environment in which the certain individual grew up can influence his/her behavior. For instance, whether the parents have different nationalities or behaviors can certainly impact managers’ behavior. Such a fine grained analysis was not possible in this study, but future research can go more in depth and study the exact factors that influence individuals in their formative years.
Second, though this study relates nationality diversity to firm multinationality, it doesn’t analyze what other factors outside the managers’ attributes influence or lead to firm multinationality. Building on the upper echelons theory that specifies a relationship between
managers’ characteristics and their strategic decision making, this study does not analyze the mediating role of the firm’s strategy, for instance. The strategic choices managers make should be in line with the overall firm strategy. Therefore, future research could study how this mediating factor could influence the aforementioned relationship.
Third, due to the nature of the secondary data gathered on the executives’ nationalities it wasn’t possible to compute a Blau index that would indicate the actual nationality diversity by country of the TMT. Instead a regional diversity was computed which is not able to capture as accurately the nationality diversity construct, but is still highly relevant for the current study. Though, by using a proxy measuring the number of different nationalities this limitation was certainly alleviated. There are differences even between countries in a certain region therefore future research could tackle the impediment encountered in this study and use measures that would reflect the nationality diversity by country of the TMT.
In addition to the findings and limitations of the present study, some opportunities for future research arise. First, as this is a cross-sectional study, the longitudinal dimension is not taken into account. Thus, there is the potential to analyze the proposed relationships over a period of several years and observe whether any changes occur. This could improve the reliability of the results.
Second, investigating other factors that affect the behavior and knowledge structures of top executives, like psychological attributes or other experiences from the early life of the executives on the TMT could give more insight into the decision making processes. Access to such information was not possible in this study, but future research could investigate the effects such factors could have on firm multinationality.
Finally, including other moderating factors in the relationship between TMT nationality diversity and firm multinationality could provide additional elements for discussion. Taking a
