THE IMPACT OF CEO CHARACTERISTICS ON INTERNATIONALIZATION PACE, SCOPE AND RHYTHM

(1)

THE IMPACT OF CEO CHARACTERISTICS ON

INTERNATIONALIZATION PACE, SCOPE AND RHYTHM

by

YOUSSEF HAMKA

S2310023

University of Groningen

Faculty of Economics and Business

MSc International Business & Management

Master’s Thesis

First Supervisor: Dr. R. W. de Vries

Co-assessor: Dr. D. H. M. Akkermans

(2)

ABSTRACT

This paper has assessed the influence of CEO characteristics on the internationalization process of a sample of Fortune 500 firms. Primarily based on arguments from the Uppsala internationalization model, I relate CEO age, tenure, and compensation to internationalization pace, scope, and rhythm. The multiple regression analysis based on data from the period 2008-2012 on Fortune 500 firm’s CEOs and their internationalization process shows a weak but significant negative impact of CEO tenure on internationalization pace and scope. In addition, the results did not find evidence for a significant impact of CEO age on internationalization pace and scope. Finally, the role of CEO compensation in internationalization pace, scope and rhythm was found to be insignificant.

Keywords: Upper Echelons, Internationalization, Chief Executive Officer, Uppsala

(3)

INTRODUCTION

The role of the Chief Executive Officer (CEO) within a corporation is without doubt of great significance. Carrying a wide range of responsibilities, having to satisfy both stakeholders and shareholders, it is obvious that not every individual is fit to occupy this demanding top position. Their formal, but also symbolic power, is believed to have a significant impact on the activities and consequently, the performance of the firm (Finkelstein, 1988; Gupta, 1984). As such, many views on which characteristics a successful CEO should carry and how those characteristics affect various aspects of a firm operations have emerged over time (e.g. Rajagopalan & Datta, 1996; Manner, 2010; Kaplan, Klebanov, & Sorensen, 2012). These type of discussions have been synthesized by Hambrick and Mason (1984) as the Upper Echelons theory, attributing organizational outcomes to the characteristics of top managers. Though the majority of literature in this field assumes the role of the CEO to be equal to those of other top management team (TMT) members, Carpenter and Sanders (2002) present a more nuanced perspective in which the status and power of the CEO is considered to be substantially different from other TMT members. Brady and Helmich (1982) as well as Vancil (1987), for instance, argue that the role of the CEO is the most powerful among the power centres with ultimate legal authority and responsibility, controlling and directing key organizational goal efforts. Hence, it is may be more appropriate to analyse the CEO in its own person, rather than as a part of the TMT for accurately assessing the CEO’s influence on an organization’s activities (Arendt, Priem, & Ndofor, 2005; Jaw & Lin, 2009; Minichilli, Corbetta, & MacMillan, 2010).

(5)

aspects of the firm’s internationalization process. This is also indicated by Bilkey and Tesar (1977), and Kedia and Chokar (1986), who have shown that internationalization decisions are strongly affected by the management’s perception of market size, costs, risks and benefits. The role of the CEO in the internationalization of the firm is relatively under-researched and has still many venues left to be explored. Existing research predominantly looks at the influence of the CEO on the direction of FDI flows and choice of foreign entry modes (e.g. Hermann & Datta, 2002, 2006; Jaw & Lin, 2009; Musteen, Datta, & Hermann, 2009). The internationalization process itself, however, remains largely untapped. This paper therefore strives to address this gap and assesses the influence of CEO characteristics on a firm’s internationalization process. Internationalization process literature primarily revolves around the Uppsala internationalization model which was originally created by Johanson and Vahlne in 1977. Though the model has received much criticism over the past 40 years, it is still often cited in recent literature. One of those criticism concerns the ignorance of individual level aspects in the internationalization process (Nielsen, 2010). Since the model takes a firm-level perspective and ignores individual level influences, approaching it from an individual point of view provides some interesting research opportunities. One of those opportunities constitutes the impact of the aforementioned CEO characteristics on the internationalization process, which this paper will address. This topic is relevant to research as it may help organizations to employ more effective recruitment strategies and better align their goals and interests concerning internationalization with the characteristics of executives.

For an operationalization of the internationalization process, Vermeulen’s and Barkema’s (2002) dimensions of expansion patterns are considered. Specifically, I assessed the influence of various CEO characteristics, including CEO age, tenure, and compensation on a firm’s internationalization pace, scope and rhythm. The research question is as follows: How do CEO

characteristics impact a firm’s internationalization pace, scope and rhythm? By using data on

the CEO characteristics and internationalization tendencies of 159 Fortune 500 firms during the period 2008-2012, this question was addressed.

(6)

are addressed. Finally, since existing research predominantly focuses on the outcomes of internationalization (e.g. Vermeulen & Barkema, 2002; Wagner, 2004; Chang & Rhee, 2011), the analysis of internationalization pace, scope and rhythm as outcome variables may yield interesting insights into the antecedents of the internationalization process.

(7)

LITERATURE REVIEW

Conceptualizing internationalization

Internationalization is a key factor in the expansion of firms and can take many different forms, varying from the establishment of an entirely new subsidiary to exporting or the acquisition of a minority share in a foreign firm. Similarly, motives range from resource and efficiency seeking to the exploration of new high potential markets (e.g. Dunning, 1988). Internationalization can house innovation and knowledge transfer which in turn may enhance performance (Bartlett & Ghoshal, 1989; Kogut & Zander, 1993). Being such as a broad concept, a variety of theoretical perspectives on internationalization exists. Generally, one can distinguish between economic and behavioural internationalization theories (Hermannsdottir, 2008). This section will briefly elaborate on a few common internationalization perspectives from both streams in order to better understand the yet to be discussed Uppsala internationalization model and the criticisms it has received over time, but also to understand the challenges and considerations that are typically being encountered during the process of internationalization.

Economic internationalization perspectives

Common economic internationalization perspectives include the Eclectic Paradigm and the transaction cost theory (also referred to as internalization theory). These kind of theories describe under which conditions a firm decides to internationalize and view the firm as a sensible entity that is able to rationally choose between alternatives.

(8)

component of the model states that the decision whether to outsource foreign production or not depends on the benefits to be grasped from doing it internally. Hence, a transaction can take place either through the market or through the organization itself depending on which channel is most cost-efficient.

In addition to these motives, a number of entry modes that suit each of the OLI advantages are presented. A firm possessing all three OLI advantages should engage in Foreign Direct Investment (FDI), while those only having ownership and internalization advantages should engage in exporting. Finally, firms that only have ownership advantages should license its production.

Another well-known theory, bearing close resemblance to the internalization component of the Eclectic Paradigm is the transaction cost theory (TCT) (Coase, 1937). It proposes that, under certain conditions, it is more efficient to for a firm produce, or engage in activity in general, through the firm itself rather than through the market. Hence, there are certain costs bound to market transactions. These transaction costs arise from the need to reorganize, carrying out and controlling the transaction as a result of market imperfections. High transactions costs will consequently make a firm internalize the transaction, whereas low transaction costs may lead to a market transaction which in some cases involves internationalization. In addition, the TCT proposes that the firm’s organizational structure developments and location choices should be driven by the goal to minimize transaction cost (Coviello & Martin, 1999).

(9)

Behavioural internationalization perspectives

Relative to the aforementioned economic views, the behavioural approaches of internationalization are more dynamic in nature and actually discuss the internationalization process itself instead of its precedents. Two well-known theories are the foreign investment process theory by Aharoni (1966) and the Uppsala internationalization process model by Johanson and Vahlne (1977). I will discuss the former below and dedicate a special section to the Uppsala model.

Aharoni’s behavioural approach attempts to identify the reasons behind foreign investment and how this investment activity is managed. It is viewed as a complicated social process that is influenced by social relationships from both inside and outside the firm. The FDI decision making process in turn constitutes various elements of individual and organizational behaviour. The firm’s first foreign investment is considered to be essential. In such situation, the firm has not accumulated any international experience yet and may therefore not have set out a plan or have prepared a specific course of action. Aharoni considers the experience gained in the firm’s first foreign venture therefore as an important source of knowledge for subsequent international activities.

A critical step in the theory is that there must be an initiating force that triggers a foreign investment decision making process. This also means that it is not necessarily a deliberate decision which is assumed to be the case under the economic models. After such a triggering event, the organization is compelled to investigate FDI alternatives after which reviewing bodies within the organization will bargain as to which alternative solution is most appropriate (Larimo, 1995). During these investigative activities, the organization unwarily makes new commitments in the form of time and resources spent on the investigation which may eventually make it more difficult to abandon a project.

Aharoni’s model is already much more dynamic than the aforementioned economic theories by recognizing that there are multiple individual and organizational forces at play. It remains, however, rather ambiguous about how exactly the internationalization process is influenced by those social forces which are internal and external to the firm. As such, I consider its practical applicability as limited.

The Uppsala internationalization process model evaluated

(10)

analyses given their dynamic nature. Yet, I discarded the foreign investment decision model by Aharoni for lacking in depth details on the factors that affect internationalization. This brings us to the Uppsala internationalization model, which is actually partially derived from Aharoni’s model, but is more specific regarding the course of the internationalization process.

The original model

The Uppsala model was introduced by Johanson and Vahlne (1977) in their paper on the internationalization process of the firm. It revolves around the idea that internationalization is a process in which firms gradually increase their international involvement. In this context, the lack of foreign knowledge is considered as an important obstacle to the establishment of foreign operations. Such knowledge can in turn be obtained by gaining international experience in nearby markets, hence expanding gradually. Those nearby markets are characterized by a low degree of psychic distance, which is defined as ‘factors preventing the flow of information from and to the market’ (Vahlne & Wiedersheim-Paul, 1973: 308). They therefore view internationalization as the product of a series of incremental adjustment decisions. In retrospect, this conceptualization of internationalization as a process of learning was rather modern for the 1970s.

The theory is based on a number of observations of Swedish firms in the 1970s of which the majority developed their international operations in small steps rather than making large foreign investments at once. Four stages were identified: (1) no regular exports; (2) exporting through an independent representative; (3) operating an overseas sales subsidiary; and (4) operating an overseas production facility. They explain this establishment chain by considering it as the consequence of a process of incremental adjustments to changing conditions of the firm and its environment. This process involves learning as well. These environmental and intra-firm changes and adjustments are believed to present new problems but also new opportunities.

(11)

current business activities (current activities). Hence, the state aspects are assumed to impact the change aspects, and vice versa. All components of the model (as shown in Figure 1) are discussed in detail below.

FIGURE 1

Internationalization process model (Johanson & Vahlne, 1977).

Starting with market commitment, Johanson and Vahlne believe that market commitment impacts the way the firm perceives risks and opportunities. It consists of the amount of resources committed and the degree of commitment. The first basically represents the size of the investment in the market whereas the latter refers to the difficulty of finding an alternative use for those resources, defined as asset specificity. This degree of commitment increases when the amount of resources integrated with multiple parts of the firm rises. Being vertically integrated and thus having resources tailor-made to the organization, it becomes much more difficult to redeploy those resources elsewhere. The same applies to resources that are specific to certain markets.

(12)

experiential knowledge is more helpful in unstructured and ambiguous activities. Overall, the consequences of those current business activities are usually not realized until the activities are repeated continuously due to the effect of learning.

Moving on with the change aspects, current business activities are considered as an important source of experience. Such experience consists of firm experience and market experience and can therefore not be obtained by solely hiring experienced personnel. It requires people who are able to work and interpret information from both inside and outside the firm. An important note here is that the consequences of current business activities are not noticeable until those activities are carried out multiple times. Hence, there is a lag between most current activities and their consequences. The longer this lag, the greater the firm commitment.

Finally, commit decisions are dependent on what decision alternatives are raised and how they are chosen. Consequently, they are made in response to perceived problems and possibilities in the market. The awareness of those problems and possibilities is again depending on accumulated firm and market experience. As such, both individuals internal and external to the firm may propose solutions to opportunities and problems.

Briefly summarizing the theory, firms are expected to gradually increase their involvement in foreign markets. They typically start with a low commitment entry mode such as exporting and gradually increase commitment as they learn. This process is shaped by state and change aspects as discussed above.

Critical evaluation

(13)

different from what Johanson and Vahlne observed back in the days. Similarly, firms from emerging markets also tend to deviate from incremental internationalization patterns. Instead of expanding gradually by learning in psychically close markets, they internationalize with the purpose of taking advantage of opportunities that are presented in foreign markets and to obtain the strategic assets that they are lacking (e.g. Govindarajan & Gupta, 2001; Lessard, Lucea, & Vives, 2012; Luo & Tung, 2007). Moreover, it is also becoming increasingly clear that small and medium sized enterprises (SMEs) may follow different internationalization patterns compared to larger enterprises (e.g. Reuber & Fischer, 1997). Finally, research has shown that internationalization patterns differ across industries as well (e.g. Sinkovics, Yamin, & Forsgren, 2007; Carneiro da Rocha, & da Silva, 2008).

This raises the question as to what exactly has changed for some firms to internationalize in much larger steps. One of the main answers can be found in the globalization of not only the international business environment, but also the world in general. Transportation costs have significantly decreased over the past few decades and formerly closed markets have been liberalised (e.g. Lessard et al, 2012; Luo & Tung, 2007). In addition, trade barriers are vanishing and technology brings the world closer together. As such, markets have become increasingly global, but also more complex and dynamic. This globalization process has therefore presented many opportunities of which today’s INVs and Born Globals are taking advantage.

Second, Forsgren (2002) criticises the model for including only one dimension of learning (learning through experience), while other forms, such as mimetic learning and knowledge based acquisitions are not considered. Especially these acquisitions may allow for faster internationalization, thereby possible contributing to explaining the accelerated internationalization of INVs, SMEs and emerging market firms as discussed above.

(14)

Revised Uppsala model

Part of what appears to be a trend among IB researchers, Johanson and Vahlne introduced a revised model in 2009 after having taken into account the developments in the field and the criticisms they received. The main alteration of the model can be found in the consideration of networks. Whereas the original model revolved around the concept of psychic distance, resulting in a liability of foreignness, Johanson and Vahlne’s new model recognizes the importance of a firm’s position in relevant business networks that foster successful internationalization. Instead of the liability of foreignness, the focus shifts towards the liability of outsidership, referring to the extent to which a firm is (not) integrated in such business networks. Figure 2 represents the altered model.

FIGURE 2

Revised internationalization process model (Johanson & Vahlne, 2009).

(15)

a strong commitment to network partners in order to access and accumulate this otherwise inaccessible knowledge.

(16)

Internationalization pace, scope and rhythm

Since internationalization is an ongoing process that may take many years to complete, if it completes at all, it should also be analysed from such long-term perspective. Vermeulen and Barkema (2002) provide a solution to this. Their study analyses firm performance as the product of the internationalization process. They distinguish between three dimensions of the internationalization process: pace, scope, and rhythm. Each of these dimensions is briefly discussed below.

Internationalization pace involves the number of foreign subsidiaries that a firm establishes within a specified unit of time. The internationalization pace of a firm is said to have serious consequences for its learning process. Fast expansion reduces the time a firm and its management have to evaluate their experiences abroad, thereby impeding its applicability in subsequent expansions. As such, the number of foreign subsidiaries established in a given time period influences the firm’s absorptive capacity and learning process.

Internationalization scope comprises two diversification dimensions, product scope and geographic scope. Product scope is defined by the number of ventures and their dispersion into different product markets. The introduction of a new venture requires new knowledge and different routines and practices. This, in turn, requires substantial amounts of time and dedication from managers. A foreign expansion process characterized by a high product scope, or high degree of diversification, requires managers to spread out their dedication over multiple projects simultaneously and involves the processing of great amounts of information, which may lead to suboptimal decisions and outcomes. Geographic scope, on the other hand, refers to the number of different countries into which a firm expands. Since individual countries differ in terms of culture and institutions, the firm has to get acquainted with a country’s unique national setting. The more countries involved in an expansion process, the more a firm has to learn and the more difficult it becomes to absorb all the different experiences during a given time frame.

(17)

cause the absorptive capacity to be either overloaded or underutilized, which may have negative performance implications.

A significant difference between Vermeulen and Barkema’s and my approach is that I view the internationalization process dimensions as an outcome of CEO characteristics whereas Vermeulen and Barkema use them as an input to explain performance. An example of an approach that also uses the internationalization process as a dependent variable is provided by Lin (2012) who examines the effects of family ownership on the internationalization process dimensions that were presented by Vermeulen and Barkema.

After having discussed the concept of internationalization and the internationalization process dimensions, I will now develop the hypotheses concerning the role of CEO characteristics in the internationalization process.

The CEO and internationalization

Being at the apex of the organizational hierarchy, the CEO is generally considered as the most powerful person in an organization (Norburn, 1989). With such great power comes great responsibility. Dalton and Kesner (1983), but also Farkas and Wetlaufer (1996) view the CEO as the person who is ultimately responsible and accountable for action on and reaction to an organization’s strategic change. Combining this with Johanson and Vahlne’s (2003) and Sousa and Bradley’s (2006) suggestion that the relationship between market entry and a concept such as psychic distance applies at the level of the individual decision maker and not the firm in itself, this strongly suggests that the CEO impacts internationalization decisions. Moreover, the Upper Echelon’s perspective indicates that the demographics of CEOs represent a powerful explanation for variations in strategic choices and organizational outcomes (Hambrick & Mason, 1984). I will therefore evaluate a number of demographic CEO factors in order attempt to find an explanation for variation in internationalization processes that may be attributed to CEOs. These demographic factors include CEO age and tenure1_{. In addition, CEO}

compensation will be considered as it was found to impact strategic decisions concerning internationalization as well (e.g. Raposo, 2005; Carpenter & Sanders, 2002). Please note, that for each of the CEO characteristics, I will not evaluate all internationalization dimensions. Internationalization rhythm, for instance, will only be related to CEO compensation while a significant relationship between CEO age and tenure on one hand, and internationalization

(18)

rhythm on the other, is ungrounded in theory. Due to data constraints, only geographic internationalization scope is incorporated in the analysis.

CEO age

The age of the CEO is a commonly discussed measure in existing research. It is often related to risk taking in multiple aspects of the firm’s operations.

Yim (2013) found that CEO age negatively impacts on acquisition behaviour. She shows that acquisitions significantly increase financial compensation. This creates a financial incentive for young CEOs as they are more likely to announce an acquisition than older CEOs given their longer career horizon. Based on the high failure rates of mergers and acquisitions (M&As), this may imply that younger CEOs tend to take on more risk. Herrmann and Datta (2006) conclude the same from their analysis of 380 foreign market entry events. In addition, they argue that CEO age can be considered as a good proxy for their level of experience.

Prendergast and Stole (1996) argue that younger CEOs tend to take on greater risk and invest more aggressively in order to prove themselves and to show their alleged talent, but also because they the lack the experience that older CEOs have. In addition, Serfling (2014) analysed the stock return volatility of 2356 firms over a period of 18 years in which a total number of 4493 CEOs were involved. He found that CEO age is negatively related to stock return volatility, implying that the risk-taking propensity of younger CEOs is higher. Barker and Mueller (2002) share this finding. As age increases, CEOs tend to put more emphasize on security and therefore avoid risky decisions. Child (1974) argues that older CEOs possess less physical and mental stamina, making them more risk-averse in general. Finally, Musteen et al (2006) propose that older CEOs possess a more conservative attitude towards change, which may be related to risk taking as well.

(19)

capacity is believed to be negatively associated with CEO age, impeding decision making efficiency (Taylor, 1975). According to Johanson and Vahlne (1990) and Lane (1995), internationalization involves the adaption to different institutional and cultural settings. As this requires the processing of significant amounts of information, older CEOs may be limited in their understanding of foreign cultures and institutions. Since older CEOs are likely to be aware of the fact that they experience difficulties in absorbing and evaluating these information flows, either through self-awareness or sub-optimal internationalization performance, this may slow down the firm’s internationalization pace.

However, when one includes the consideration of networks as presented in the revised Uppsala model by Johanson and Vahlne (2009), older CEOs may have accumulated far more experience and may therefore be embedded in more extensive networks which can be addressed in internationalization activities (Hermann & Datta, 2006). This may enable them to accelerate the internationalization process as they can utilize existing networks in order to identify and exploit new international opportunities. Younger CEOs, on the other hand, are likely to lack experience and may therefore not be embedded in as many and extensive networks as older CEOs do. This would require the establishment of new relationship commitments, slowing the down the identification of opportunities and hence impedes the internationalization pace of the firm.

Given the difficulty of weighting both arguments, it is complicated to predict how exactly CEO age relates to a firm’s internationalization pace. For practical reasons, I therefore hypothesize that there will be a negative relationship between CEO age and a firm’s internationalization pace.

Hypothesis 1a: The age of the CEO is negatively related to a firm’s internationalization pace.

(20)

addition, operating in a variety of national contexts presents more risks, which older CEO tend to shy away from.

On the other hand, again using the network argument, the experience of older CEOs accompanied with their embeddedness in a variety of business networks may very well offset or compensate for the arguments presented above. Having access to network contacts in geographically diverse markets can foster geographical diversification as it may reduce the risk that CEOs perceive in expanding to these markets. Since it is again difficult to weight both arguments, I hypothesize that there will be a negative relationship between CEO age and internationalization scope.

Hypothesis 1b: The age of the CEO is negatively related to a firm’s geographic

internationalization scope.

CEO tenure

CEO tenure represents the duration of a person’s function as CEO at a specific firm. The role of tenure in internationalization related matters is under researched.

Generally, CEOs tend to be more reluctant to change as their tenure increases (Grimm and Smith, 1991; Hambrick, Henderson, & Miller, 1999). In addition, Hambrick and Fukutomi (1991) shows that, as their tenure increases, CEOs become increasingly committed to their own ideas of how the organization should be led. In such paradigm, there is not much room for other views which is said to impede change. This goes accompanied with a loss of interest in organizational change as the novelty of the CEOs job decreases over the course of the CEO’s tenure. Eventually, a long-tenured CEO may fail to keep up with the environmental conditions in which the firm is embedded, thereby neglecting to make the changes and adaptations necessary for the successful continuity of the firm. Miller (1991) comes to the same conclusion. Miller argues that is of great importance that an organization’s structure and strategy matches the challenges presented by its environment and found that the likelihood of achieving such match decreases as CEO tenure increases. According to Meyer (1975), long-tenured CEOs have more power to resist external pressures. They have been able to accumulate this power by hiring and promoting those individuals who possess similar views to their own and divest those who do not, increasing the CEO’s autonomy (Mintzberg, 1983).

(21)

identifying and exploiting international opportunities relative to short-tenured CEOs. In that regards, a short-tenured CEO may be more dynamic and rigorous in expanding the firm’s operations across borders. As such, I expect CEO tenure to negatively impact both internationalization pace and scope.

Hypothesis 2a: CEO tenure is negatively related to a firm’s internationalization pace. Hypothesis 2b: CEO tenure is negatively related to a firm’s internationalization scope. CEO compensation

In addition to the aforementioned demographic traits, the financial aspects of the CEO’s position may play a significant role in organizational outcomes as well. Not only the absolute amount of compensation, but also its structure was found to play an important role (e.g. Carpenter & Sanders, 2002). Moreover, senior management compensation is believed to ultimately impact strategic goals, goal alignment and risk preference (Devers, Cannella, Reilly, & Yoder, 2007). In existing literature, CEO compensation is therefore mainly related to risk-taking propensity and the performance of the firm (e.g. Sun, Zhao, & Yang, 2010; Werner & Ward, 2004; Murphy, 1985). Finkelstein and Boyd (1998), for example, show from their analysis of Fortune 1000 firms that organizations that compensate their CEOs based on managerial discretion perform better. In addition, Adithipyangkul, Alon, and Zhang (2011) found that firms from an emerging market context tend to rely more on non-cash compensation such as perks. These perks granted to CEOs are found to positively impact firm performance. Moreover, Dow and Raposo (2005) propose that CEO compensation influences the kinds of strategies a firm adopts. Performance-related compensation is said to create an incentive for overly ambitious and hard to implement strategies, accompanied with an increase in risk-taking propensity. Moreover, variability in compensation is positively related to the level of risk taking (Larraza-Kintana, Wiseman, Gomez-Mejia, & Welbourne, 2007). Hence, existing literature indicates that CEOs may view their compensation as a motivational construct that impacts the decisions they make. This is likely to be reflected in internationalization decisions as well.

(22)

itself in internationalization activities. This paper therefore addresses this void. One of the few studies, if not the only study that analyses this relationship is presented by Lin and Cheng (2013:1). The hypotheses of their study on 528 publicly listed Taiwanese firms are based on motivational theory, predicting that substantive compensation packages can motivate CEOs to increasingly engage in international expansion, which is represented by a combination of three dimensions: foreign sales, foreign production, and geographical dispersion. Lin and Cheng predict that compensation affects the behaviour of the CEO by functioning as a buffer to cognitive dissonance and triggering the CEO to attract professional managers who have international experience. Hence, a high level of compensation motivates CEOs to deal with the complexity of foreign operations but also enables the firm to attract and retain experienced and valuable experts (Lin & Cheng, 2013:1). Adhering to this rationale, I predict that the level of CEO compensation is positively associated with a firm’s internationalization pace and scope.

Hypothesis 3a: The level of CEO compensation is positively associated with a firm’s

internationalization pace.

Hypothesis 3b: The level of CEO compensation is positively associated with a firm’s

internationalization scope.

(23)

maintaining a regular and thus successful internationalization pattern. This is reflected in the hypothesis stated below.

Hypothesis 3c: A higher level of CEO compensation will lead to a stable internationalization

rhythm.

Conceptual model

Figure 2 summarizes the abovementioned discussion in a conceptual model. The overview and

justification of the control variables can be found in the methodology section.

(24)

METHODOLOGY

Sample and data collection

For the analysis of the CEO characteristics, I will use a sample of Fortune 500 firms, which are the 500 largest U.S. firms from a variety of industries ranked by total annual revenues according to annual ranking by Forbes. The Wharton Research Data Services (WRDS) database was addressed for the data on the CEO characteristics of the Fortune 500 firms. The WRDS database has data available for CEO age, gender, tenure and multiple compensation dimensions for the years 2004 up to and including 2014. In order to measure internationalization rhythm, the sample only includes firms that uninterruptedly had a place in the Fortune 500 ranking during the selected time frame. Moreover, only firms that have an international presence were included in the sample. The Fortune 500 was chosen for its heterogeneity in terms of industries, allowing to control for industry type in the analysis. In addition, all Fortune 500 firms originate from the same national context, eliminating the need to take into account significant cultural and institutional differences in the home country environment.

Similar to Vermeulen and Barkema (2002), internationalization data is gathered by analysing annual reports. The annual reports of U.S. firms are very elaborate in nature as a result of strict U.S. Securities and Exchange Commission (SEC) regulations. Listed U.S. firms are required to publish a so-called Form 10-K which is an annual report based on SEC standards and requirements. In addition to this 10-K form, many companies provide a summary of their main activities and achievements during a given year in a report addressed to their shareholders. These two reports are often merged into one comprehensive publication. Some firms even go beyond these requirements and present additional overviews of important events. As a result, these annual reports constitute a rich data source for this study. The annual reports were also addressed for the control variables concerning firm size and profitability as will be discussed in the following section. Companies of which the annual reports were either inaccessible or ambiguous concerning their international activities were excluded from the sample.

(25)

2008-2012. Comparing the Fortune 500 rankings for the period 2008-2012 yielded 282 firms that continuously occupied a position in the ranking during this period. Data restrictions in the WRDS database concerning the CEO factors for some of the firms reduces this number to 237 firms Another reason for excluding the years 2004 till 2007 is the limited availability of annual reports during this period as most companies only archive their annual reports for a period of ten years or even less. After manually searching the annual reports of those 237 firms for the year 2008, 40 firms were excluded from the sample since their 2008 annual report was not available or accessible on either their website or elsewhere online. Another 37 companies were omitted for lacking an international presence. This reduces the sample size to 159 firms per year and thus a total of 795 observations over the entire time frame.

Variables

Independent variables

All CEO data was gathered from the WRDS database, made available by the Wharton University of Pennsylvania. CEO age is measured in the number of the years. For CEO tenure, the number of years that the CEO held his or her position since appointment is taken. CEO compensation covers the natural logarithm of the total dollar amount that a CEO receives per annum, including the base salary, the annual bonus, performance based bonuses and miscellaneous compensations. A one year lag was taken for CEO compensation to avoid causality issues. This means that the data for CEO compensation covers the period 2007-2011.

Dependent variables

(26)

When the internationalization rhythm of a firm is irregular and thus characterized by an uneven expansion pattern, the distribution will be relatively concentrated and will thus yield a high kurtosis. A constant rhythm on the other hand, comes with a relatively equal distribution and thus yields a low kurtosis.

Control variables

(27)

(28)

RESULTS

Test method and preliminary analyses.

The independent variables and control variables were grouped according to their relationship with the dependent variable. As such, three analyses were performed in which all relevant independent and control variables are related to each of the three dependent variables. A multiple linear regression analysis was used to test the hypotheses. This method is suitable when the dependent variable is continuous while the predictor variables are of a mixed nature: all independent and control variables are continuous except the dummy variables created for industry type and investment year which are binary. For each analysis, the added explanatory power of the dependent variables on top of the control variables is calculated. This means that I started with a model that only contains the control variables after which I added the independent variables in order to see how the explanatory power changes.

Before actually testing the hypotheses, some basic analyses were carried out to identify any potential problems concerning multicollinearity, autocorrelation, linearity, normality, homoscedasticity, and outliers, These assumptions need to be met in order to perform a reliable multiple regression analysis. In order to test for multicollinearity, a bivariate correlation analysis was performed. The results of this analysis are shown in Table 2. The most common and reliable threshold used for identifying multicollinearity is a correlation of 0.7 (Dormann et al, 2013). The correlation analysis indicates that there are no issues concerning multicollinearity among the independent and control variables as all values are far below 0.7. This even remains the case with a more restrictive threshold of 0.5. In addition, Variance Inflation Factors (VIFs) and tolerance values were calculated and analysed in order to identify multicollinearity issues that cannot be visually derived from a correlation matrix. The tolerance and VIF coefficients indicate to what extent the variability of a regression coefficient is explained by the independent variables. High tolerance coefficients (above 0.1) and low VIFs (below 3) are desirable in this case (e.g. O’Brien, 2007; Mansfield & Helms, 1982). All tolerance coefficients are well above 0.7 with most close to 1 while the VIFs range between 1 and 1.4. Hence, there are no significant multicollinearity issues present in this sample.

(29)

any meaning in the context of this research. I therefore created linear regression plots with regression standardized residuals on the y axis and the regression standardized predicted values on the x axis, in order to visually deduct autocorrelation problems. This did not indicate any problems concerning autocorrelation.

In order assess to what extent the relationship between the relevant variables is linear, scatter plots were created with the dependent variable on the y axis and the independent variable on the x axis. These plots all indicated linearity and did not shows any deviating relationships such as parabolic relationship.

For normality, it was found that the control variables concerning firm size were heavily skewed towards the left. This issue was resolved by a logarithmic transformation. This also fixed heteroscedasticity issues for both firm size variables. Firm profitability was exponentially transformed to increase normality to an acceptable level.

Finally, all variables were checked for outliers. The threshold was set for 2.2 times the interquartile range, defined as the difference between the first (Q1) and third (Q3) quartile. This multiplier was found to be more accurate relative to the standard multiplier of 1.5 which is used in SPSS (Hoaglin & Iglewicz, 1987). Those observations that were labelled as outliers were manually checked and verified, and it was concluded that none constituted an error. Testing the three models with variables of which the outliers were removed significantly reduced the explanatory power of each model. It was therefore decided to retain all outliers.

Descriptive statistics

(30)

usually receive other forms of compensation such as stock options, further analysis showed that this was not the case for this one observation2.

For internationalization pace, the average number of foreign expansions is 1.94. Internationalization scope has a slightly lower mean with 1.76 different countries per year. Standard deviations are 2.437 and 2.060 for respectively pace and scope. As shown in Table 1, the percentage of firms that expanded abroad during a year as part of the total sample stably ranges between 62% and 65%. The annual averages of pace and scope show a decreasing tendency over the years. This negative trend may be the result of the 2008-2009 financial crisis. Internationalization rhythm had an average of 0.33 with a standard deviation of 2.79. The minimum kurtosis values was -3.33 while the maximum equals 5. In general, negative kurtosis values indicate a flat distribution with slight tails. Positive kurtosis values on the other hand imply more heavy tails with sharper peaks (DeCarlo, 1997). A low kurtosis indicates a distribution with relatively flat peaks while high values indicate sharp peaks. The low overall average therefore points to a relatively constant expansion pattern in the sample.

TABLE 1

Internationalization statistics per year Year Number of

firms

Of which have

internationalized during the year

Average pace Average scope

2008 159 99 (62.3%) 2.51 2.09

2009 159 100 (62.9%) 2 1.87

2010 159 101 (63.5%) 1.89 1.75

2011 159 103 (64.8%) 1.67 1.55

2012 159 102 (64.2%) 1.65 1.55

Concerning the control variables, the average firm age is 65 years while the youngest and oldest firm in the sample were respectively incorporated 6 and 175 years ago. The standard deviation is 41.498 years. For industry type, 53.5% of the sample is classified as a manufacturing firm. Retail trade and the combination of finance, insurance and real estate, respectively take 11.3%

(31)

and 10.1% of the total. All other categories are well below 10% each. Since the vast majority is active in the manufacturing industry, this category will be used as a reference category to which all other industry groups are compared in the analysis. The first control variable regarding firm size measures each firm’s assets and shows, that on average, the total asset value is $79,676.77 million. The largest firm in terms of assets is valued at $2,264,909 million while the smallest firm owns $1,301.7 million in assets. The second performance control variable looks at annual revenues. This variable has a mean of $35,152.64 million with a very high standard deviation of $59,785.50 million. The highest total revenues are $606,166 million while the smallest firm in terms of revenues earned $853 million. Finally, the performance control variable which divides net income divided by total revenues, shows that, on average, net income is 6.09% of total revenues with a standard deviation of 33.37%. The worst performing firm had a net loss of 8.94 times its total revenues whereas the best performing organization retained 78.23% of total revenues in its net income.

Hypothesis testing and results

Internationalization pace

(32)

0.000 with F(18,776)=5.493, indicating that 11.3% of the variation in internationalization pace is explained by the model and that adding the independent variables only explains an additional 0.9%.

Internationalization scope

The second analysis for internationalization scope as shown in Table 4 provides comparable results relative to internationalization pace when only entering the control variables. Industry type was significant at 0.000 for Transportation & Utilities (0.000), Construction (0.002), and Finance, Insurance, and Real Estate (0.000). Concerning expansion years, only 2011 and 2012 are significant at 0.007 and 0.004. Firm age showed a weak positive relationship with internationalization with a B value of 0.007 at 0.000 significance. Firm size is only significant when measured in total assets (B=.271, p=.018). The model with the control variables explains 10.5% of the variance in internationalization scope as indicated by the R square value of .105. Entering CEO tenure in model 2 increases the R square value by 0.8%. Tenure shows a weak but significant negative relationship (B=-.034, p=.014) with internationalization scope, indicating that internationalization scope tends to decrease with increasing CEO tenure. Hypothesis 2a is therefore supported. No support was found for hypotheses 1b and 3b as CEO age and CEO compensation are highly insignificant. Model 2 explains 11.3% of the variance in internationalization scope and is significant at 0.000 with F(18,776)=5.385.

Internationalization rhythm

The analysis concerning internationalization rhythm as shown in Table 5 is significant at 0.000 when entering the industry type control variables into model 1. Specifically, Wholesale Trade, Services, and Transportation & Utilities are significant at respectively 0.01, 0.000 and 0.003. Firm age is weakly related to rhythm with a B value of 0.007 at 0.004 a significance level. Firm size and profitability are not significant. Model 1 has an R Square of 0.068, indicating that the control variables explain 6.8% of the variation in internationalization rhythm.

(33)

Means, Standard Deviations and Correlations Mean S.D. 1 2 3 4 5 6 7 8 9 10 11 1. Pace 1.94 2.44 1 .962** _.102 _-.191* _.045 _.125** _.008 _.130** _-.119** _.137** _-.018 2. Scope 1.76 2.06 .962*** ₁ _.099 _-.214** _.030 _.129** _.010 _.130** _-.097** _.126** _-.028 3. Rhythm .33 2.79 .102 .099 1 .095 -.069 -.034 -.060 .073 .c _.043 _-.015 4. Industry 2.75 2.43 -.191* _-.214** _.095 ₁ _.355** _-.056 _-.192* _.004 _.c _-.058 _.173* 5. Assets 79676.77 249742.34 .045 .030 -.069 .355** ₁ _.268** _-.109** _-.011 _.023 _.035 _.042 6. Revenues 35152.64 59785.50 .125*** _.129*** _-.034 _-.056 _.268** ₁ _.010 _-.100** _.016 _.175** _.135** 7. Profitability .061 .33 .008 .010 -.060 -.192* _-.109** _.010 ₁ _.062 _.068 _-.028 _-.048 8. Firm Age 65.00 41.50 .130*** _.130*** _.073 _.004 _-.011 _-.100** _.062 ₁ _.034 _.057 _.024 9. Expansion Year 2010.00 1.42 -.119*** _-.097** _.c _.c _.023 _.016 _.068 _.034 ₁ _-.367** _.154** 10. CEO Compensation 6138078.03 6292536.18 .137*** _.126*** _.043 _-.058 _.035 _.175*** _-.028 _.057 _-.367** ₁ _-.011 11. CEO Age 57.04 5.66 -.018 -.028 -.015 .173* _.042 _.135*** _-.048 _.024 _.154*** _-.011 ₁ 12. CEO Tenure 7.11 5.86 -.101** _-.113** _.045 _.179* _-.013 _.029 _.048 _-.084* _.004 _.069 _.403***

(34)

TABLE 3

Multiple linear regression analysis internationalization pace

N=795. ***P < 0.001, **P < 0.01, *P < 0.05. SE: Standard Error.

Model 1 Model 2

B SE B SE

Control Variables

Retail Trade Industry .127 .297 .231 .300

Wholesale Trade Industry -.003 .354 .021 .353

(35)

TABLE 4

Multiple linear regression analysis internationalization scope

B SE B SE

(36)

TABLE 5

Multiple linear regression analysis internationalization rhythm

B SE B SE

Wholesale Trade Industry -1.069* .411 -1.063* .411

(37)

DISCUSSION

Academic and practical implications

This research indicates that the role of the CEO in the firm’s internationalization process is of importance and that it requires further attention in future research. The results do not support hypotheses 1a and 1b, implying that CEO age is not significantly related to internationalization pace and scope. This may suggest that, even though the internationalization related behaviours of CEOs vary with age, the characteristics that are unique to older and younger CEOs cancel each other out in internationalization. This nullifies the effect of CEO age on internationalization pace and scope. In specific, this indicates that the weight of higher risk taking propensity (Prendergast & Stole, 1996; Barker & Mueller, 2002) and superior information processing capacity (Taylor, 1975) of younger CEOs, may be more or less equal to that of the superior network embeddedness and experience of older CEOs (Hermann & Datta, 2006). As such, neither older CEOs nor younger CEOs have the upper hand in internationalization tendency as their distinctive characteristics have an equal impact on the internationalization process. Hence, this implies that an organization would not benefit from recruiting executives based on age in order to steer the internationalization process. However, it should be noted that the average CEO age (57 years) in this study’s sample is relatively high, just like other studies that analyse CEO age (e.g. Yim, 2013; Barker & Mueller, 2002). In addition, the ‘youngest’ CEO in the sample is already 43 years old. The practical matter here is that it leaves some subjectivity as to when a CEO can be considered as young or old when comparing with other studies or samples. These results should therefore be interpreted carefully. This issue is further addressed in the section with limitations and future research suggestions.

(38)

from implementing incentive mechanisms for longer tenured CEOs in order to prevent them from losing interest in the organization and its environment over time and hence losing their openness to change. This may improve their ability and willingness to identify and exploit international opportunities which can increase the internationalization pace and scope. Alternatively, an organization may increase CEO turnover by limiting tenure terms alike presidential term limits in politics in order achieve the same results by timely hiring new motivated executives.

Concerning CEO compensation, the results oppose Lin and Cheng (2013:2) who found that the level of CEO compensation contributes to a regular expansion pattern. CEO compensation was namely not significantly related to a stable internationalization rhythm. This may indicate that raising compensation levels does not necessarily result in a regular expansion pattern and that the form rather than level of compensation may be more important in this context. This deviating finding may also stem from sample differences concerning national contexts and firm sizes but between this and Lin and Cheng’s study. CEO compensation did also not significantly impact internationalization pace and scope, showing that the level of compensation does not motivate a CEO to more rigorously expand the firm’s internationalization presence in this sample. Again, this may suggest that compensation structure is more relevant.

Limitations and suggestions for future research

(39)

capacity of younger CEOs cancels out the superior network embeddedness and experience of older CEOs in the context internationalization may therefore not hold with samples that are differently distributed. For instance, a young CEO in this study may very well be considered as middle aged in another. CEO age related research may therefore substantially benefit from an age categorization in order to improve the interpretability and comparability of the results concerning old and young CEOs. This issue also applies to CEO tenure. Fifth, I was unable to measure the impact of gender on internationalization as the result of very few female CEOs in the sample. This may imply that, even though the share of female top executives is on the rise, the absolute number is still very low. It may therefore be necessary to compose a handpicked sample where the number of male and female CEOs is more or less equal in order to assess the impact of gender on the internationalization process. Sixth, the applicability of the findings of this research is currently primarily limited to U.S. firms. By replicating this study with a different sample, and hence with firms that operate under different cultural, institutional and economic conditions, one could get more insights into the extent to which the results of this research are generalizable across different national settings. In a similar way, it is recommended to use a sample with small and/or medium sized firms. Finally, a period of 5 years may be insufficient to accurately assess a company’s internationalization rhythm. Vermeulen and Barkema (2002), for example, based their analysis on data spanning a period of 25 years, while Lin and Cheng (2013) took a period of 10 years for their analysis.

(40)

(41)

CONCLUSION

The Upper Echelons theory suggests that CEOs play a key role in an organization’s strategic decision making processes and outcomes (Hambrick & Mason, 1984). Existing research has indicated that internationalization is a pervasive source of organizational complexity which in turn places significant pressures on CEOs (Sanders & Carpenter, 1998). As such, internationalization decisions are strongly affected by the top management’s perception of market size, costs, risks, and benefits (Kedia & Chokar, 1986; Bilkey & Tesar, 1977). Even though the role of the CEO is frequently researched in a variety of internationalization related contexts such as entry modes and the direction of FDI flows, the internationalization process itself is ignored. This research has therefore filled this gap and has extended the Upper Echelons argumentation to the firm’s internationalization process by building upon the Uppsala internationalization process theory and Vermeulen and Barkema’s (2002) internationalization process dimensions. By addressing this gap, this paper aimed to assess the impact of CEO characteristics on the internationalization process. Specifically, the impact of CEO age, tenure and compensation on internationalization pace, scope and rhythm was assessed. The CEO is commonly viewed as the most powerful individual in an organization and is therefore likely to play a significant role in the internationalization process.

(42)

(43)

REFERENCES

Adithipyangkul, P., Alon, I., & Zhang, T. 2011. Executive perks: Compensation and corporate performance in China. Asia Pacific Journal of Management, 28 (2): 401 – 425. Agarwal, S., & Ramaswami, S. N. 1992. Choice of foreign market entry mode: Impact of ownership, location and internalization factors. Journal of International Business Studies, 23 (1): 1 – 27.

Aharoni, Y. 1966. The foreign investment decision process. Cambridge, MA: Harvard University Press.

Andersson, S., Gabrielsson, J., & Wictor, I. 2009. International activities in small firms: Examining factors influencing the internationalization and export growth of small firms.

Canadian Journal of Administrative Sciences, 12 (1): 22 – 34.

Apple Corporation, 2012. Proxy Statement.

Available: http://files.shareholder.com/downloads/AAPL/1661508868x0x531628/b6ec469d-aff8-4eef-9077-1defc2258f6b/2012_Proxy.pdf

Accessed: May 20, 2016.

Arendt, L, A., Priem, R. L., & Ndofor, H. A. 2005. A CEO-advisor model of strategic decision-making. Journal of Management, 31 (5): 680 – 699.

Barker, V.L., & Mueller, G. C. 2002. CEO characteristics and firm R&D spending.

Management Science, 48 (6): 782 – 801.

Bartlett, C., & Ghoshal, S. 1989. Managing across borders. Boston, MA: Harvard Business School Press.

Bilkey, W., & Tesar, G. 1977. Export behaviour of smaller-sized Wisconsin manufacturing firms. Journal of International Business Studies, 8 (1): 93 – 98.

Brady, G., & Helmich, D. 1982. The hospital administrator and organizational change: Do we recruit from the outside? Hospital and Health Services Administration, 27 (1): 53 – 62. Carneiro, J., da Rocha, A., & da Silva, J. F. 2008. Challenging the Uppsala

internationalization model: A contingent approach to the internationalization of services.

Brazilian Administration Review, 5 (2): 85 – 103.

Carpenter, M. A., & Sanders, W.G. 2002. Top management team compensation: The missing link between CEO pay and firm performance? Strategic Management Journal, 23 (4): 367 – 375.

Carpenter, M. A., & Sanders, W. G. 2004. The effects of top management team pay and firm internationalization on MNC performance. Journal of Management, 30 (4): 509 – 528. Chang, S. J., & Rhee, J. H. 2011. Rapid FDI expansion and firm performance. Journal of

International Business Studies, 42 (8): 979 – 994.

Child, J. 1974. Managerial and organizational factors associated with company performance.

(44)

Coase, R. H. 1938. The nature of the firm. Economica, 5: 386 – 405.

Coviello, N. E., & Martin, K. A. M. 1999. Internationalization of service SMEs: An integrated perspective from the engineering consulting sector. Journal of International

Marketing, 7 (4): 42 – 66.

Crick, D. 2009. The internationalisation of born global and international new venture SMEs.

International Marketing Review, 26 (4/5): 453 – 476.

Dalton, D. R., & Kesner, I. F. Inside/outside succession and organizational size: The

pragmatic of executive replacement. Academy of Management Journal, 26 (4): 736 – 742. DeCarlo, L. T. 1997. On the meaning and use of kurtosis. Psychological Methods, 2 (3): 292 – 307.

Devers, C. E., Cannella, Jr., A. A., Reilly, G. P., & Yoder, M. E. 2007. Executive

compensation: A multidisciplinary review of recent developments. Journal of Management, 33 (6): 1016 – 1072.

Dormann, C. F., Elith, J., Bacher, S., Buchmann, C., Carl, G., Carré G., Marquez, J. R. G., Gruber, B., Lafourcade, B., Leitão, P. J., Münkemüller, T., McClean, C., Osborne, P. E., Reineking, B., Schröder, B., Skidmore, A. K., Zurell, D., & Lautenbach, S. 2013.

Collinearity: A review of methods to deal with it and a simulation study evaluating their performance. Ecography, 36 (1): 27 – 46.

Dow, J., & Raposo, C. C. 2005. CEO compensation, change, and corporate strategy. The

Journal of Finance, 9 (6): 2701 – 2728.

Dunning, J. H. 1979. Toward an eclectic theory of international production: Some empirical tests. Journal of International Business Studies, 11 (1): 9 – 31.

Dunning, J. H. 1988. The eclectic paradigm of international production: A restatement and some possible extensions. Journal of International Business Studies, 19 (1): 1 – 31. Dunning, J. H. 2000. The eclectic paradigm as an envelope for economic and business theories of MNE activity. International Business Review, 9 (2): 163 – 190.

Dunning, J. H. 2001. The eclectic (OLI) paradigm of international production: Past, present and future. International Journal of the Economics of Business, 8 (2): 173 – 190.

Farkas, C. M., & Wetlaufer, S. 1996. The ways chief executive officers lead. Harvard

Business Review, 74 (3): 110 – 122.

Finkelstein, S. 1988. Managerial orientations and organizational outcomes: The

moderating roles of managerial discretion and power. Unpublished doctoral dissertation.

Colombia University.

Finkelstein, S., & Boyd, B. K. 1998. How much does the CEO matter? The role of managerial discretion in the setting of CEO compensation. Academy of Management

Journal, 41 (2): 179 – 199.

(45)

Govindarajan, V., & Gupta, A. K. 2001. The quest for global dominance: transforming

global presence into global competitive advantage. San Francisco: Wiley.

Greve, H. 2003. A behavioural theory of R&D expenditures and innovations: Evidence from shipbuilding. Academy of Management Journal, 46 (6): 685 – 702.

Grimm, C. M., & Smith, K. G. 1991. Management and organizational change: A note on the railroad industry. Strategic Management Journal, 12 (7): 557 – 562.

Gupta, A. K. 1984. Contingency linkages between strategy and general manager

characteristics: A conceptual examination. Academy of Management Review, 9 (3): 399 – 412.

Hambrick, D. C., & Fukutomi, G. D. S. 1991. The seasons of a CEO’s tenure. Academy of

Management Review, 16 (4): 719 – 742.

Hambrick, D. C., Henderson, A. D., & Miller, D. 1999. Temporal dynamics of CEO tenures. Paper presented at the 1999 Academy of Management Meeting, Chicago, IL.

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

Hammer, M. H. 2010. The impact of CEO characteristics on corporate social performance.

Journal of Business Ethics, 93 (1): 53 – 72.

Henderson, A. D., & Fredrickson, J. W. 1996. Information-processing demands as a determinant of CEO compensation. Academy of Management Journal, 39 (3): 575 – 606. Hermann, P., & Datta, D. K. 2002. CEO successor characteristics and the choice of foreign entry mode: An empirical study. Journal of International Business Studies, 33 (3): 551 – 569.

Hermann, P., & Datta, D. K. 2006. CEO experiences: Effects on the choice of FDI entry mode. Journal of Management Studies, 43 (4): 755 – 778.

Hermannsdottir, A. 2008. Theoretical underpinnings of the internationalization process. Working paper 08 (2). Institute of Business Research. University of Iceland, Reykjavik. Hoaglin, D. C., & Iglewicz, B. 1987. Fine-tuning some resistant rules for outlier labelling.

Journal of the American Statistical Association, 82 (400): 1147 – 1149

Hordes, M., Clancy, J., & Baddaley, J. 1995. A primer for global start-ups. Academy of

Management Executive, 9 (1): 7 – 11.

Hymer, S. H. 1960. The international operations of national firms: A study of direct

foreign investment. Cambridge, MA: MIT Press.

Javalgi, R. G., White, D. S., & Lee, O. 2000. Firm characteristics influencing export

propensity: An empirical investigation by industry type. Journal of Business Research, 47 (3): 217 – 228.

Jaw, Y. L., & Lin, W. T. 2009. Corporate elite characteristics and firm’s internationalization: CEO-level and TMT-level roles. The International Journal of Human Resource

THE IMPACT OF CEO CHARACTERISTICS ON INTERNATIONALIZATION PACE, SCOPE AND RHYTHM