Company leaders’ personal characteristics and
temporal focus: Culture, loyalty, education and age
MSc Business Administration: International Management Final version Master’s thesis
Author: Chadny Jeukens (10871233) Supervisor: Robert Kleinknecht MSc Second reader: Dr. Ilir Haxhi
ABSTRACT
This study investigates the relationship between temporal focus and company leaders’ personal characteristics such as culture, loyalty, education, and age. Personal characteristics have been subject of many studies, although mixed and contradictory findings appear in many studies as well. Scholars have also investigated the relationship between temporal focus and personal characteristics before, yet not in this setting. Where previous research measures temporal focus, for example, through a financial measure like research and development expenditure, this study uses content analysis. Qualitative data are transformed into quantitative data through linguistic analysis, capturing a measure of company leaders’ cognition. Through analyzing 332 letters to shareholders, extracted from fiscal year 2012 annual reports – based on the 2013 Fortune
Global 500 – a temporal focus measure has been computed through a ratio using word counts
from references to either the short or the long term. The findings suggest that culture is a personal characteristic influencing company leaders’ temporal foci. In contrast, loyalty, education and age appear to have no influence on temporal focus as displayed by company leaders in letters to shareholders.
ACKNOWLEDGEMENTS
I would like to take this opportunity to thank Robert Kleinknecht for the valuable cooperation during his time as my supervisor. His guidance, comments and feedback have been extremely helpful for the completion of this thesis.
STATEMENT OF ORIGINALITY
This document is written by Chadny Jeukens, who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
TABLE OF CONTENTS
1. Introduction………. 5
2. Literature review………. 8
2.1 Temporal myopia, short-termism and temporal focus.………... 9
2.2 Causes of short-termism………... 10
2.3 Company leaders’ characteristics………. 12
3. Theoretical framework……… 14 3.1 Culture……….. 15 3.2 Loyalty……….. 16 3.3 Education……….. 18 3.4 Age……… 19 4. Methodology……….. 21 4.1 Data set………. 21 4.2 Measures………... 23 4.2.1 Dependent variable……… 23
4.2.2 Independent and moderating variables……….. 25
4.2.3 Control variables……… 28
5. Statistical analyses and results………... 30
5.1 Statistical analyses……… 30
5.2 Results……….. 34
6. Discussion……….. 36
6.1 Discussion of results………. 37
6.2 Limitations and future research……… 41
6.3 Practical implications………..………. 43
7. Conclusion………... 44
References………... 47
Appendices……….. 53
A. 2013 company list Fortune Global 500……….. 53
B. Content analysis dictionary……….... 57
C. Descriptive statistics and correlations……… 58
1. INTRODUCTION
Over the last decades, especially in light of the financial crises, short-termism has been a topic which received significant attention (e.g. Porter, 1992; Brochet, Loumioti & Serafeim, 2012; Gonzalez & André, 2014). Short-termism has received significant critique and is assumed to have played a major role in these financial crises, as a result of an extreme focus on results in the short term and taking enormous risks (Gonzalez & André, 2014). In the general view of the capital market – including board members, managers, investors, and auditors – short-termism can be described as preferring strategies regarding investments that pay off early rather than investments opting for maximal payoff over the full duration of these investments (Jackson & Petraki, 2011). Looking at the intra-firm level, this phenomenon can be viewed as the pursuit of short-term benefits – such as the maximization of profits on a quarterly basis – at the cost of long-term objectives – such as research and development (R&D) investments. Following this interpretation of short-termism, it can be regarded as a characteristic of firms and the industries in which they operate (Laverty, 2004). However, other scholars consider short-termism specific for individuals rather than for firms as a whole. At the individual level, Shipp, Edwards and Lambert (2009) refer to ‘temporal focus’ rather than to ‘short-termism’ with respect to the extent to which people value the past, present and future in relation to each other.
Company leaders – presidents, chairmen and CEOs – may sometimes consciously opt for short-termist behavior in an attempt not to lose face and retain self-esteem in the present. Company leaders want to boost their reputation as early as possible and, therefore, also their wages since performance is measured through the rate of return in the relationship with shareholders as the principals and managers as the agents (Narayanan, 1985; Lazonick & O’Sullivan 2000). This form of gaming (Marginson, McAulay, Roush & Van Zijl, 2010) helps company leaders to achieve personal performance levels in the short-term at the expense of the company’s long-term performance (Brochet et al., 2012; Marginson et al., 2010). In short, the
source of this problem is that the most desirable state in the short term is not aligned with the most desirable state in the long run (Laverty, 1996), where the line between short and long term can be drawn at one year (e.g. Brochet et al., 2012).
Throughout the literature, short-termism and temporal focus have proven to be difficult phenomena to measure. Laverty (1996) noticed early on that research had been too narrowly focused, solely on the economic aspects. In this respect, Marginson et al. (2010) gathered a sample of senior managers operating in the telecommunications industry to test the effect of different types of performance measurements, either financial or non-financial, on short-termism among firms. They found that organizational short-short-termism may be associated with non-financial measures rather than financial ones, where it is the usage of the measures that affects short-termism and not the measures in itself (Marginson et al., 2010). Temporal focus, as part of managerial decision-making, can and should, however, be considered from multiple perspectives. Not only do economic or financial dimensions matter, individual and organizational dimensions are important factors too (Laverty, 1996; Marginson & McAulay, 2008). From the individual perspective of temporal focus, the personal characteristics of top level managers largely impact strategic decision-making (Shipp et al., 2009; Kwee, Van Den Bosch & Volberda, 2011).
In an attempt to gather a comparable measure of temporal focus which can be related to company leaders personally, this thesis uses content analysis. Content analysis can be regarded as a method with both qualitative and quantitative influences, which is regarded as a high potential method for topics like short-termism, specifically temporal focus, which are difficult to measure and study (Duriau, Reger & Pfarrer, 2007). Using communications through letters to shareholders in annual reports, this thesis analyzes the content with respect to references to both short and long time horizons. Based on a final sample of 332 companies of the Fortune
through signal words indicating either short or long time horizons (e.g. Brochet et al., 2012; DesJardine & Bansal, 2014).
Thus, literature to date comprises many factors, both contextual and personal – respectively at the level of firms and their industries, and at the level of company leaders – which have an influence on short-termism. What the current short-term literature has not looked into yet, is the relationship between temporal focus at the individual level as displayed in shareholder letters in annual reports, using content analysis, and personal characteristics of company leaders. The cause of firms being short-termist can be either the firm, which would classify the firm as a whole as short-termist, or the company leader, who may be personally characterized as focusing on the short term. Therefore, the research gap that this thesis will try to fill up is to see whether company leaders’ personal characteristics can be indicative for temporal focus as expressed in letters to shareholders. Thus, the results will show whether short-termism within firms can be attributed to company leaders personally rather than to firms in general. This leads to the following research question: ‘Can temporal focus be attributed to company leaders’ personal characteristics, such as culture, loyalty, education, and age?’ The added value of this way of measuring temporal focus, as opposed to using R&D expenditure or any other financial figure as a measure for short-termism (Barker & Mueller, 2002), is that content analysis of letters to shareholders focuses on the temporal focus of company leaders specifically and can therefore be attributed more directly.
The remainder of this thesis is structured as follows. First, in the next chapter, the literature regarding short-termism, temporal focus and company leaders’ personal characteristics will be reviewed. Hereafter, in the theoretical framework, the company leaders’ characteristics to be hypothesized will be theoretically reviewed, as well as the base for the dependent measure. In the fourth chapter, the methodology regarding the data collection and the different types of variables will be discussed. In the following chapter, the data will be used
for the statistical analyses from which the results will follow. In chapter six, the results will be discussed in light of the expectations following from the literature. In the final chapter, a conclusion will be drawn from this thesis.
2. LITERATURE REVIEW
In their study on corporate strategy, Lazonick and O’Sullivan (2000) found that, during the twentieth century, a transformation has taken place in the Anglo-Saxon economies from a strategy of retention and investment to a strategy of shareholder value maximization. Until the 1960s, corporation would retain their earnings and reinvest in fixed assets and human resource. However, during the 1960s and 70s, corporate strategies started to change. In an increasingly globalizing world, corporations started to grow through mergers and acquisitions, whereas competition – especially from Japan – was becoming more innovative. Larger companies led to more division and more different types of business units, making it difficult for headquarters to make well-informed decisions with respect to the strategy of retain and reinvest, which in turn led to poor corporate performance (Lazonick & O’Sullivan, 2000). According to the agency theory, shareholders are principals and managers are their agents with respect to corporate governance. As a result of this principal-agent relationship, the return on stock would be the new performance measure, aiming for shareholder value maximization, thus providing managers with fewer incentives for long-term investments (Narayanan, 1985; Lazonick & O’Sullivan, 2000). However, instead of building up, shareholder value maximization might rather break down companies and economies (Lazonick & O’Sullivan, 2000). Doyle (1994), early on, urged on reassessing corporate goals and performance measures, following the increasing rate of failure of companies which were regarded to be excellent not long before their failure.
As Porter (1992) addressed in an early stage, firms’ continuous success and ability to compete internationally are dependent on innovation, and on retaining current and attaining
additional competitive advantages, hence long-term investments. Especially in the contemporary world of global competition, investments are an essential factor towards competitive advantage. Following Porter’s line of argument, Segelod (2000) added that firms investing in the long term will have an advantage with respect to capacity and market share, gaining superior market positions in comparison to short-term focused competitors. So why is a short-term temporal focus such an apparent problem if long-term investments are so important for the viability of firms?
The next section first addresses important distinctions, which are the distinctions between temporal myopia, short-termism and temporal focus. The second section, on the causes of the short-termism phenomenon, discusses the reasons for short-termist behavior despite the importance of the long term. In the final section, the focus is shifted from contextual factors contributing to short-term temporal foci, to personal factors which characterize company leaders, specifically.
2.1 TEMPORAL MYOPIA, SHORT-TERMISM AND TEMPORAL FOCUS
Short-termism, with its alleged linkage to financial scandals and crises, has a negative sound. However, prudence is called for when using similar terms with respect to the short term, as there is a difference between lacking the ability to properly value different time horizons and consciously choosing for the short term over the long term. In this context, it is important to make a distinction between myopia and short-termism. Temporal myopia involves focusing on the short-term, as a managerial deficiency of cognitive ability, through the overvaluation of short-term benefits while undervaluing consequences in the distant future (Miller, 2002; Laverty, 2004). Short-termism, however, can be regarded as a systematic characteristic, resulting from culture and routine, amongst others (Laverty, 2004). Laverty (1996) provided insight into five plausible explanations as to the existence of short-termism. Of those five, two are a reflection of management shortcomings: flawed management practice and managerial
opportunism (Laverty, 1996; Miller, 2002). Flawed management practices include techniques discounting the future while emphasizing performance in the short term, hence undervaluing the future. Managerial opportunism involves making choices which may be “an optimal personal intertemporal choice for a manager, but it may be a suboptimal intertemporal choice for the organization” (Laverty, 1996, p. 832).
Furthermore, a distinction between the generally negative construct of short-termism and the individual-level construct of temporal focus is needed. Short-termism is based on decisions and their eventual outcomes, where the decisions made are optimal for the short term but not for the long term (Laverty, 1996). Looking at the individual perspective regarding short-termism, people undergo distinct developments with respect to how they value different time periods. Temporal focus is based on the differences between individuals with respect to how they perceive the past, the present and the future and how these perceptions influence their strategic decision-making, in which both the short and the long term have to be considered (Shipp et al., 2009). Thus, short-termism is a construct focused on the decisions and outcome, whereas temporal focus is a construct focused on individuals.
2.2 CAUSES OF SHORT-TERMISM
Literature to date has explored the capital investment market as a whole, resulting in many different types of forces that encourage short-termism. Initially, the short-termism research was mainly focused on stock market pressure. With stock prices rising shortly after firms announced long-term investments (Segelod, 2000), managerial pay started to be based on stocks, aiming for performance in the long run (e.g. Jackson & Petraki, 2011). However, in practice, managerial pay would still be related to stock prices in the short term. As a result, managers were unlikely to convince shareholders about benefits in the long run, hence making managers short-termist instead (Palley, 1997). The resulting intertemporal tradeoff is a common part of the managerial role, leading to contradictory goals, and therefore contradicting time horizons
for managers and shareholders, respectively (Laverty, 1996). Following the principal-agent theory, shareholders are the dominating party, provided that firms generally reward managers who have success in the short term higher than managers who invest in good long-horizon projects (Laverty, 2004; Jackson & Petraki, 2011). Therefore, Gigler, Kanodia, Sapra and Venugopalan (2014) constituted that short-termism is a problem originating from shareholders rather than from managers, hence a firm characteristic. This point of view has been shared by many other scholars who believe that term behavior by firms is merely a result of a short-term investor base guiding managers’ preferences towards investors’ preferences. As a result, managers become more concerned with meeting short-term earnings than with investing in long-term project such as R&D (Liljeblom & Vaihekoski, 2009; Brochet et al., 2012; DesJardine & Bansal, 2014).
Although the attempt to base managerial remuneration on stocks did not result in managers focusing more on the long term (Jackson & Petraki, 2011), a further debate on the effectiveness of stock-based pay followed. Whereas exercisable stock options would more likely lead to short-term investments, unexercisable stocks were introduced to align the personal interest of managers with those of their organizations (e.g. Souder & Shaver, 2010). While unexercisable stock options may help to lengthen the horizon of managers, many scholars have doubted their effectiveness in general. Where the intention of these stocks is to lengthen decision horizons, they actually contribute to the poorly designed remuneration structures, which eventually negatively affect firm performance (Antia, Pantzalis & Park, 2010). Souder and Bromiley (2012) also found contradictory results as to the positive effect of stock options on managerial compensation, stating that “the cure (options) may worsen the disease (short-termism)” (p. 565).
Another factor playing a major role in short-termism literature are managerial perceptions of the aforementioned stock market pressures. Based on a study in the UK, Demirag
(1998) proved the presence of a relationship between managers’ perceptions and the effect of this perceived pressure. What is most peculiar is that short-termism is apparent as much as managers feel pressure, even if there is no actual pressure, such as in case of long-termist capital market shareholders. However, if the pressure actually exists, managerial tend to be faced with more financial controls, eventually resulting in short-termism (Demirag, 1998).
Other factors that have been explored in relation to short-termism are the frequency of reporting, poor short-term performance, and takeover pressure. A high frequency of reporting urges managers to perform and meet the expected numbers every single period. The resulting pressure is yet another one for managers to develop short-termist behavior, which directs their perspectives towards the short term (Gigler et al., 2014). Souder and Shaver (2010) investigated the effect of poor short-term performance on making long horizon investments. Their research showed that poor performance in the short term inhibits firms from investing in long-term projects. Firms lose their competitive advantage through focusing on annual or quarterly profits while losing sight of adding value to the firm in the long run. The resulting poor short-term performance, in turn, increases the need for better short-term results afterwards, risking ending up in a vicious circle. Last, takeover pressure, as a result of poor performance and thus poor financial numbers and low stock prices, increases the need for managers to significantly increase short-term performance as well, neglecting the long-term future of the firm once more (Stein, 1988).
2.3 COMPANY LEADERS’ CHARACTERISTICS
Next to the contextual factors, managers’ individual temporal foci have had an impact on short-termism. According to Souder and Bromiley (2012), the American economy has seen a decrease with respect to competitiveness as a result of term managerial temporal foci. These short-term foci are, in turn, a result of the overvaluation of results in the near future in comparison to results in a more distant future. This ‘short-termism’ phenomenon can be regarded as a
consequence of personal characteristics. Literature to date has researched several managerial characteristics.
First of all, a major subject is culture, which is widely regarded as a distinguishing factor between people in general. Institutions, both formal and informal, shape individuals. Formal institutions are rules and laws to which an individual has to live, whereas informal institutions are norms and values which determine how individuals interact socially (Nielsen & Nielsen, 2013). These informal institutions, the culture, can be attributed to individuals specifically and can, therefore, be treated as a personal characteristic for company leaders. Human interaction is highly influenced by culture, whereby different national cultures lead to the aforementioned distinguishing factor (Mamman & Saffu, 1998), as different countries bright forth people with different cultures (Chow, Kato & Merchant, 1996).
Another major subject in personal characteristics literature is average tenure (e.g. Palley, 1997, Huang, 2013). Average tenure refers to the number of years a company leader has spent on a job on average, taking into account all the relevant jobs he or she has had over the span of his or her career. This thesis discusses a company leader’s personal temporal focus, connecting the company leader to his or her respective current employer instead of overviewing the entire span of his or her career. Therefore, rather than average tenure, this thesis considers loyalty as a characteristic for company leaders. Loyal company leaders may be more concerned with the company overall, being committed and having built relationships with many stakeholders, and therefore may be more long-term oriented rather than being interested in short-term personal gains (Buchholtz & Ribbens, 1994). Short-term focused people, on the other, may be concerned more with enhancing their CV for future employers (Brand, 1998).
Several scholars have studied the importance and significance of personal characteristics before, although in different settings. Barker and Mueller (2002) focused on R&D spending, as a fundamental part of top managerial decision-making, since R&D
contributes to the enhancement of competitive advantage in the rather distant future. Huang (2013) researched corporate sustainable development in this respect and also found supporting evidence for major demographic variables affecting strategic choices. One important characteristic of managers in these studies is educational background. However, Gottesman and Morey (2006a) have investigated both the effects of the prestige of schools attended by CEOs and the types of their degrees, finding no relation with the performance of these CEOs. CEO characteristics may as well evolve over time and be related to successes throughout earlier stages of careers, and personality traits may be developed socially rather than through education (Gottesman & Morey, 2006a). Despite the mixed results with respect to education as a personal characteristic, it appears to be an important aspect in literature to date and is, therefore, worth looking into further with respect to temporal focus.
Finally, when discussing multiple major demographic characteristics, age cannot be neglected. Similar to the evaluation of the importance of education, many scholars have pinpointed age as an important influence on strategic decision-making (e.g. Barker & Mueller, 2002; Huang, 2013). In line with age as a characteristic, Matta and Beamish (2008) have researched CEO career horizons, which are typically associated with the different career stages. Shorter horizons are more apparent when CEOs are nearing retirement, thus have a higher age, where short horizons are closely connected to short-termist behavior. In line with this reasoning, Antia et al. (2010) stated that this behavior is not only apparent when CEOs approach the age of retirement but also when the likelihood of being replaced is high. CEO turnover and CEO age appear to be related as well, as turnover rates tend to rise when CEOs get older (Antia et al., 2010).
3. THEORETICAL FRAMEWORK
Literature comprises both contextual factors, at the levels of the firms and industries, and personal factors. Focusing on company leaders, personal characteristics will be addressed to
investigate whether short-termism can be attributed to company leaders personally, resulting from their temporal foci. In the next section, the first characteristic to be discussed is cultural background. In the section thereafter, loyalty will be addressed, referring to the numbers of years company leaders have spent both within their respective company and at their current job, followed by their educational backgrounds. In the final section of this chapter, the possible relation between age and temporal focus will be discussed.
3.1 CULTURE
National culture is a significant factor in shaping human interaction. Due to variety in national and cultural factors, managerial behavior varies across different countries as well (Mamman & Saffu, 1998). Managerial behavior is formed by the respective nation of origin as a result of formal as well as informal institutions. As formal institutions are merely rules and laws, it is hard to relate these to people personally, especially in a globalized environment in which many people find employment outside the borders of their nation. However, culture – informal institutions – is considered to be a vast pattern of norms and values which are unlikely to change over time, as these patterns originate from childhood, when people learn and assimilate the most (Nielsen & Nielsen, 2013). Attitudes of executives, as well as their interpretation and decision-making with respect to strategic choices, are heavily influenced by their national culture (Schneider & DeMeyer, 1991; Geletkanycz, 1997).
Temporal focus is an aspect of these cultural differences that affects the time horizons over which people aim to achieve successes. Short-term successes, in this respect, are based on financial achievements in the near future, whereas managers who aim for long-term success seek to develop skills and capabilities to ensure firm success in the distant future. Based on this, Harris and Carr (2008) argued that cultures which emphasize the long term will more likely bring forth managers that aim for the achievement of success in the long term rather than the short term. Carr and Tomkins (2002) also stressed the importance of cultural differences and
found support for the impact on financial approaches and styles of management at a strategic level. Based on Hofstede’s (1991) dimensions of national culture, individualistic behavior can be regarded as a measure of cultural differences (e.g. Carr & Tomkins, 2002). Hofstede (1991) used the dimension of Individualism vs. Collectivism to measure the extent to which individuals are an integral part of groups. Individualists do not value attachment to other individuals; individuals should take care of themselves and their immediate family only. Collectivists, on the other hand, like to experience more cohesion between individuals, like strong relations within groups (Hofstede, 1991) These distinctive attitudes towards either the self – which could be a reflection of short-termist behavior – and towards the collective – which could be a reflection of valuing what is good for the firm and all of its stakeholders, hence long-termist behavior – can be interpreted as a characteristic for company leaders. Based on the assertion that high scores for Hofstede’s (1991) Individualism measure are positively related to short-term temporal focus, the hypothesis is as follows:
Hypothesis 1: The higher Hofstede’s Individualism score for the CEO’s nation, the more short-term the temporal focus of a company leader from that nation will be.
3.2 LOYALTY
In the traditional deal between employees and employers, loyalty and commitment can be expected from the employees, whereas employers offer employment and career prospects (Baruch, 2001). From the psychological point of view, the underlying concept of such a contract is that employees contribute to the performance and benefits of their company, after which they are being rewarded by the company. This mutual commitment ensures that people are attached to their company, thus creating loyalty and commitment (Baruch, 2001). In his case study on knowledge management and innovation at 3M, Brand (1998) states that loyalty of top managers, following from promotion within the company, ensures that these managers will know and understand other people within the company better. Gaining experience in multiple
countries and departments, a wide range of personal ties can be created throughout the company, leading to employees helping each other without expecting personal gains. On the contrary, people employed on a short-term basis may prefer boosting their CV for future employers, hence focusing on their profession and personal benefits, rather than creating a commitment of loyalty between the employer and employee (Brand, 1998). This view is supported by Buchholtz and Ribbens (1994), who found that length of tenure is positively related to CEOs’ commitment to their firms.
This thesis measures loyalty as a company leader’s personal characteristic in terms of loyalty to both the company and the current job. The number of years a company leader has spent ensures personal ties with many different types of stakeholders and can, therefore, lead to the preference of long-term company performance rather than short-term gains from a personal point of view. Additionally, regarding the relationship between the number of years a company leader has spent in his or her current job and short-termism, a possible moderating effect can be whether or not he or she had been employed within the current company before assuming this job. It may be expected that outsider company leaders are, in line with Brand (1998), more interested in personal short-term gains in terms of wages or CV. Based on the assertion that more years lead to higher commitment and loyalty, hence more focus on the long-term, the hypotheses regarding loyalty are as follows:
Hypothesis 2a: The more years a company leader has spent within the current company, the less short-term the temporal focus of that company leader will be.
Hypothesis 2b: The more years a company leader has spent in his or her current job, the less short-term the temporal focus of that company leader will be.
Hypothesis 2c: If a company leader had been employed by the current company before assuming the current job, then the effect of the direct relationship in hypothesis 2b will be stronger.
3.3 EDUCATION
Many scholars have studied demographic variables before with respect to executives. Besides national culture and loyalty, education is another important characteristic for company leaders. When researching educational backgrounds, it is not only important to address the degree of education but also the specialization, hence the type of education (Hitt & Tyler, 1991; Tyler & Steensma, 1998; Barker & Mueller, 2002; Manner, 2010; Huang, 2013). Both the degree and the type of education has an effect on the development of cognitive assessments of company leaders, eventually influencing strategic choices during their careers (Hitt & Tyler, 1991; Tyler & Steensma, 1998; Manner, 2010). In their study on managers’ degrees and fund performance, Gottesman and Morey (2006b) found that managers who have obtained an MBA from top schools perform better than managers who graduated from lower ranked schools, resulting from superior ability to see and evaluate the big picture. Regarding education type, upper echelons theory states that managers’ backgrounds, in part, predict strategic choices as a result of the formation of different cognitive bases of skills and knowledge (Hambrick & Mason, 1984). The type of formal education a company leader has followed defines the type of information this company leader focuses on when evaluating different strategic options (Hitt & Tyler, 1991). People are educated in types of studies like business, economics and management tend to develop rather short-term temporal foci (Hambrick & Mason, 1984; Manner, 2010), while engineering or technical studies rather foster long-term commitments within people (Hambrick & Mason, 1984; Tyler & Steensma, 1998). Based on the assertions that higher educated company leaders are more able to see the big picture, thus take into account all stakeholders, and that company leaders who have followed business-like education are more short-term
focused than engineering-like education, the following hypotheses regarding education will be tested:
Hypothesis 3a: The higher the degree a company leader has obtained, the less short-term the temporal focus of that company leader will be.
Hypothesis 3b: Company leaders who have obtained a degree in business-like education will have a more short-term temporal focus than company leaders who have obtained engineering-like education.
3.4 AGE
Similar to educational background, age has been identified as a major personal characteristic with regards to strategic decision-making (e.g. Barker & Mueller, 2002; Huang, 2013). One view is that short-termist behavior, thus short-term temporal foci, emerge when CEOs are becoming older, with their retirement approaching sooner (Beamish, 2008). In the research by Barker and Mueller (2002), on R&D spending, evidence was found that CEOs are less willing to spend money when they are older, as their incentives are different from relatively young CEOs. Because of the risk of R&D spending, profitability in the near future may be endangered, while pay-offs might only show off when the CEOs are not around anymore, hence benefiting their successors (Antia et al., 2010). The contrary view is that older individuals have a long-term temporal focus (DesJardine & Bansal, 2014). Narayanan (1985) addressed experience, related to a company leader’s age, as an important factor. Experienced company leaders are more aware of their abilities, which reduces the incentives for quick short-term gains. The potential benefit of projects that ensure rather late returns are higher, lowering the probability that an older and more experience company leader will look for short-term gains (Narayanan, 1985). With regards to the perception of pressure, Brunzell, Liljeblom and Vaihekoski (2012) argue that younger managers are more sensitive to short-term pressures. Older managers are
more independent and, therefore, do not feel that pressure as much. Kang (2014) has addressed two different rationales to investigate whether older CEOs are more interested in short-term gains as long as they are in the position or whether they are more concerned about the legacy and the reputation they leave behind. From an economic point of view, CEOs may look to boost their value, whereas from a cognitive view, CEOs want to maintain their legacies and reputations (Kang, 2014). In spite of the mixed findings, age appears to be an important characteristic in decision-making and, subsequently, might be an indicator for company leaders’ temporal foci. Based on the assertion that older company leaders obtain shorter decision horizons, the hypothesis is as follows:
Hypothesis 4: The higher the company leader’s age, the more short-term the temporal focus of that company leader will be.
Analyzing the content of chairman letters to shareholders in annual reports will result in a measure to decide on the temporal focus as displayed by a company leader. This temporal focus can then be analyzed to discover relationships with company leaders’ personal characteristics – culture, loyalty, education, and age.
The goal of this analysis is to be able to determine whether these characteristics are indicative for company leaders’ personal temporal foci, implying that short-termism within firms could be ascribed to individuals rather than to firms as a whole. The hypotheses described above, to answer the research question, are summarized into a scheme as shown in figure 3.1 above.
4. METHODOLOGY
In this methodology chapter, the data set will be described first. In the following section, a description will be given of the sample used in this thesis. Next to that, the data collection process will be explained, including the data sources used to complete the data. In the second section, the content analysis method, which has been used to gain a measure with respect to the time horizons of company leaders, will be explained for this thesis. Hereafter, the measures used are depicted, specified into the dependent variable, independent variables, moderating variable and control variables. The fourth and final section of this chapter introduces the models used to analyze the data and test the hypotheses.
4.1 DATA SET
The sample used in this study is based on the Fortune Global 500. In a cross-sectional research
design, based on the data of the 2012 fiscal years ending March 31st, 2013 at the latest, the
world’s 500 biggest companies with respect to revenues in the list of 2013 have been included to examine the effect of company leaders’ personal characteristics on the temporal foci they display. The choice for the Fortune Global 500 is based on the diversity of company leaders
nationalities, as a global sample, and the size of the companies. In this list, with the size of companies being measured in revenues, pressures to perform in the short term may be expected to be present, as short-termism is in essence about investing, or omitting to do so, in the short term while neglecting long-term company performance (Kwee, Van Den Bosch & Volberda, 2011; Brochet et al., 2012; Marginson, McAulay, Roush & Van Zijl, 2010). Furthermore, the data are recent, which is important as short-termism and temporal focus are relatively new subjects in literature and are specifically relevant in light of the recent financial crises (e.g. Gonzalez & André, 2014). However, of the 500 companies in total, only 397 produced an annual report including a letter to shareholders in English, which is a requirement to obtain a comparable measure of temporal focus. Of these 397 remaining letters, 65 were excluded because either the letter was not written by one specific company leader and could therefore not be attributed to one person’s characteristics or, in rare cases, the personal characteristics of the company leader could not be ascertained, once more making it impossible to link the temporal focus measure to company leaders’ personal characteristics. Therefore, in the final sample, a total of 332 companies and their respective company leaders have been included. An overview of the companies included in the final sample can be found in appendix A.
This study focuses on company leaders’ personal characteristics, which makes it difficult to extract large amounts of data from commonly used databases in business research. Therefore, data on company leaders are gathered from various specific databases like
Bloomberg, World of CEOs, The Wall Street Journal, NNDB and China Vitae. Additional
sources are company websites or personal websites like LinkedIn and indirect sources obtained via Wikipedia.
To obtain a measure with respect to the temporal focus as displayed by company leaders, archival data have been used. Specifically, annual reports – including the letters to shareholders – are largely available and easily accessible through the internet. Archival data, unlike surveys,
interviews and accounting data used in prior research, are rather objective in that they are not subject to response rates and have been proven to be reliable (DesJardine & Bansal, 2014). These qualitative textual archival data, which letters to shareholders in annual reports are, are transformed into quantitative data by the use of linguistic analysis in order to be able to obtain a measure of company leaders’ cognition with respect to their time horizon, hence their temporal focus (Chatterjee & Hambrick, 2007; DesJardine & Bansal, 2014).
4.2 MEASURES
In this section, the different types of variables used for the statistical analyses in the next chapter will be discussed. First, the dependent variable regarding temporal focus will be explained. Hereafter, both the independent and the moderating variables to test the hypotheses will be discussed. Finally, six variables are controlled for to limit the confounding effect of these variables on the established relationships in the hypotheses.
4.2.1 DEPENDENT VARIABLE
To compute an analyzable measure of issues that are difficult to study, such as short-termism and, specifically in this thesis, temporal focus, the content analysis method has been used to codify textual data into categories of quantitative data (e.g. Duriau, Reger & Pfarrer, 2007; DesJardine & Bansal, 2014). Based on the assumption that analyzing written content gives insight into people’s cognition, prior research shows that word counts are a vital indicator (Duriau et al., 2007).
Using the letters in the fiscal year 2012 annual reports of the Fortune Global 500, a measure of temporal focus has been generated through word counts of DesJardine and Bansal’s (2014) dictionary of key words, categorized as referring to either a short-term or long-term temporal focus, whereby the line can be drawn at one year (e.g. Brochet, Loumioti & Serafeim, 2012). This dictionary was constructed through a five-step process (DesJardine & Bansal,
2014). First, an extensive wordlist was compiled, based on different forms of archival data. Not only annual reports, but also conference call transcripts, sustainability reports and other releases were used to include all relevant words, completed by dictionaries from linguistic analysis programs with regards to time horizons. Second, these words were manually categorized as either short-term or long-term, where possibly ambiguous words were excluded. Next, the dictionary was validated based on the context of the words included, where further ambiguous words were once more excluded, such as ‘now’. Fourth, after word counts of the validated dictionary were possible, a ratio was computed to capture a measure for temporal focus:
𝑇𝑒𝑚𝑝𝑜𝑟𝑎𝑙 𝑓𝑜𝑐𝑢𝑠 = # 𝑜𝑓 𝑙𝑜𝑛𝑔 − 𝑡𝑒𝑟𝑚 𝑡𝑒𝑚𝑝𝑜𝑟𝑎𝑙 𝑓𝑜𝑐𝑢𝑠 𝑤𝑜𝑟𝑑𝑠
# 𝑜𝑓 𝑠ℎ𝑜𝑟𝑡 − 𝑡𝑒𝑟𝑚 𝑡𝑒𝑚𝑝𝑜𝑟𝑎𝑙 𝑓𝑜𝑐𝑢𝑠 𝑤𝑜𝑟𝑑𝑠 + # 𝑜𝑓 𝑙𝑜𝑛𝑔 − 𝑡𝑒𝑟𝑚 𝑡𝑒𝑚𝑝𝑜𝑟𝑎𝑙 𝑓𝑜𝑐𝑢𝑠 𝑤𝑜𝑟𝑑𝑠
Finally, this measure was manually validated by multiple researchers, leading to a 90% inter-rater reliability, which can be regarded acceptable. An overview of the dictionary can be found in appendix B (DesJardine & Bansal, 2014).
The word counts for the 332 letters in the final sample were constructed in two steps: first, the annual reports were retrieved in PDF format. Because this format cannot be used for text analysis, the letters to shareholders have been copied to Notepad files. Second, a computer-aided text analysis tool, Aaron F. McKenny’s CAT Scanner, was used to analyze these files and retrieve the word counts needed for the temporal focus ratio above.
The ratio varies between 0 and 1, whereby a lower ratio indicates a more short-term temporal focus on behalf of the respective company leader. This temporal focus measure will be analyzed with regards to company leaders’ personal characteristics such as culture, loyalty, education, and age, to explore possible relationships between these characteristics and the extent to which company leaders’ temporal foci are short-term.
4.2.2 INDEPENDENT AND MODERATING VARIABLES
To test the different types of hypotheses, several independent variables are included in the statistical models. The hypothesis with respect to culture includes one variable which is linked to Hofstede’s Individualism score. For the hypotheses regarding loyalty, there is one variable with regards to the number of years a company leaders has spent within his or her current company and one variable with regards to the numbers of years in the current role. Additionally, there is a dummy variable stating whether the company leader is an insider or not, as a moderator in the relationship between years in current job and temporal focus. The education hypotheses require two independent variables: the level of education and the type of education. Last, for the hypothesis regarding age, there is one variable based on both the dates of birth and
the cross-section at March 31st, 2013.
Hofstede’s cultural Individualism score. One of Hofstede’s (1991) dimensions of cultural
values comprises of Individualism vs. Collectivism. More individualistic company leaders would rather look out after themselves, hence personal benefits in the short term, than after all stakeholder involved in the company, hence long-term company performance. The scores for each nation are linked to the country of origin of the company. In most cases, obviously, this is the country of birth. However, in cases in which this was not necessarily representative for their nationality, a judgment was made based on all the relevant and available information – name, country where the company leader was raised, country of education, start of career, etcetera. In rare cases, this judgment could not be made adequately, leaving this data unit empty. For the N = 328 for which these data are available, the lowest score was 14, indicating a long-term orientation, and the highest score was 91. With M = 63.03 (SD = 26.027), on a scale of 0-100, company leaders tend to be slightly individualistic on average.
Number of years within the company. Loyalty with respect to company leaders’ tenure
data with respect to company leaders’ first jobs are rather limited, this variable was limited to a year number. With the cross-section being made in 2013, the number of years a company leader has spent within a company is calculated by the difference between 2013 and the year of commencement. Of the N = 323 (M = 21.53, SD = 13.791) for which these data are available, the minimum is 0, indicating a company leader who has just begun his or her tenure shortly before the point of cross-section. The maximum number of years a company leader has spent within the company is 70.
Number of years in the current role as a company leader. Similar to prior research on average
tenure, this variable covers the number of years of company leaders’ tenure in their current position. With data on current positions being more recent and of lower variance, with respect to number of years, these are accurate to months. Including the month in which the company leader has started in his or her position, the total number of years has been calculated by full years, as far as applicable, and partial years for the year running, accurate to months up until March 2013. For this variable, N = 330 units have been collected (M = 6.5977, SD = 6.86158). The minimum is 0.25, indicating a current tenure of three months, and the maximum is 43.25.
Insider company leader. This study distinguishes between insider and outsider company
leaders. Insider company leaders have been employed within the company before and are thus promoted to their current role. Occasional exceptions are founders who are still these companies’ leaders and have been categorized as insider company leaders. Outsider company leaders, on the other hand, have no prior commitment to the company. Using this dummy variable, the direct relationship between the number of years in the current role and temporal focus can be moderated. Data on N = 323 company leaders (M = 6.6561, SD = 6.91137) are known for this variable, including 66 outsiders (M = 5.7891, SD = 5.97772) and 257 insiders (M = 6.8787, SD = 7.12499).
Level of education. When distinguishing based on education, one obvious distinction that can
be made is the level of education. Company leaders can either have a bachelor’s degree, a master’s degree or a PhD degree. Classified as a dummy variable, company leaders have only been assigned to one category, based on the assertion that one cannot obtain a master’s degree without having a bachelor’s degree, et cetera. Of the N = 317 company leaders for which education data is available, 112 have obtained a bachelor’s degree as their highest degree. Furthermore, 151 have obtained a master’s degree and 49 even obtained an PhD. The remaining 5 have not completed any of these degrees. The reference category will be ‘bachelor’s degree’.
Type of education. The second obvious distinction to be made with respect to education is the
type; a dummy variable for company leaders’ educational background was computed. Based on the largely specific types of education in business, three categories were made initially: Business and Economics, Beta, and Other. Business and Economics also includes backgrounds such as Finance, Management, Accounting, Commerce and Marketing. The Beta category comprises, next to Engineering, also Physics, Science, Chemistry, Biology and Mathematics. The Other category includes all the other types of backgrounds which vary from Foreign Languages and Law to Politics and History. From these three main categories, two extra combined categories were computed since many company leaders with some sort of background in Business and Economics also have a background in either the Beta category or the Other category. Last, again, there are
five who have no degree and are therefore not attributed to an education background. The reference category for this variable will be the Other category. An overview of the frequencies per category can be found in
Company leader age. Since ambiguity remains with respect to the effect of age on
short-termism, this variable is included in this study to further address this issue. This variable was computed similar to the variable concerning the number of years of company leaders in their current role. For company leaders born between April and December, their age variable was computed through subtracting from 2012 their year of birth. Company leaders who were born between January and March have had an extra birthday in 2013 due to the cross-section at
March 31st. For the full sample of N = 332 (M = 57.85, SD = 7.646), the age variable has been
computed, ranging from the youngest company leader of 35 years old to the oldest at age 86.
4.2.3 CONTROL VARIABLES
In order to limit the confounding effect of other potential effects on the relationships established in the hypotheses, this study comprises six control variables. First, gender is controlled for. The second control variable is the headquarters home region of the respective companies. Third, company age is included. Furthermore, the industries in which the companies operate, with their particular characteristics and conditions, are controlled for. Another control variable concerns whether the companies are publicly listed or not. Finally, the firm size, expressed in the number of employees, is controlled for.
Gender. Although males are overly represented in the final sample of N = 332, with 97%, men
and women are widely known to be different in many ways, possibly influencing their strategic decision-making and thus their temporal focus (e.g. Huang, 2013). Therefore, gender is controlled for, as an important personal characteristic which is, due to the low amount of females company leaders in this sample, difficult to investigate more thoroughly.
Headquarters home region. Similar to the independent variable of Hofstede’s Individualism
score on cultural characteristics of company leaders, home region is controlled for. Whereas informal institutions shape human interaction, companies are subject to formal institutions, not
in the last place in their home country (Nielsen & Nielsen, 2013). Different institutions lead to different types and forms of shareholder pressures. Bushee (1998) characterizes Asian and European firms as rather long-term oriented due to their corporate governance and dedicated shareholders, in comparison to the United States. The dummy variable used consists out of five categories: North America; South and Central America; Europe; Middle East and Asia; and Oceania, where Russia is categorized under Europe. The reference category used for this dummy variable will be Europe.
Company age. The third control variable will be company age, calculated by subtracting the
year of establishment from 2013, since this might be a potential influence on temporal focus due to its effect on both short-term and long-term performance levels. Outdated technology and management techniques are more likely to be present in older firms than in younger firms (De Jong & Van Houten, 2014).
Industry type. Furthermore, another dummy variable in the form of industry type is controlled
for. Different industry are typically associated with different time horizons due to their specific characteristics. For example, where the financial industry can be regarded to be short-termist, the mining and quarrying industry thrives on long-term contracts. This may influence the type of shareholders companies attract and therefore also the way in which company leaders, on behalf of these companies, communicate with their shareholders (Hitt, Ireland & Stadter, 1982). The categories for industry type are based on the NACE rev. 2 code, through which the European Community statistically classifies companies based on their primary economic activity. The final sample contains seven categories: mining, quarrying, agriculture, forestry and fishing; manufacturing (primary sector); electricity, gas, steam and air conditioning supply; wholesale, retail trade, repair of motor vehicles and motorcycles; information and communication; financial and insurance activities; and a rest category ‘other’, being the reference category.
Publicly listed. Whether or not a company is publicly listed is another dummy variable that can
have an impact on temporal focus. As opposed to listed firms, privately-held firms have no shares owned by the public but rather by committed owners, founders or family. Therefore, in comparison, listed firms are subject to more shareholder pressures which could lead to a more short-term temporal focus on the behalf of company leaders.
Number of employees. The last control variable is the size of companies, expressed by the
number of employees. The choice for this measure of size is based on the logic behind Hofstede’s (1991) Individualism dimension, which distinguishes between people who take care of themselves and people who value intergroup relationships. Being a leader of a company of many people, may emphasize collectivist behavior by company leaders who score low on the Individualism score. Individualistic behavior, on the other hand, might be amplified for company leaders who score high on Individualism and lead rather small companies.
5. STATISTICAL ANALYSES AND RESULTS
5.1 STATISTICAL ANALYSES
In table 5.1 below, a summary of the descriptive statistics and the correlations regarding the variables to be included in the various analyses to test the hypotheses can be found. A full overview of the descriptive statistics and correlations of all variables can be found in appendix C. Before testing these hypotheses, a check for multicollinearity has been performed. Bowerman, O’Connell and Murphree (2009) argue that many statisticians consider multicollinearity to be severe when one or more simple correlation coefficients are 0.9 or higher, where other statisticians utilize 0.7 or higher. From the correlations between the variables in this study, there are only two simple correlation coefficients exceeding the limit of 0.7. The matrix shows a coefficient between a bachelor’s degree and a master’s degree of -0.705. However, since these are two distinct categories of one dummy variable, it is unlikely
that these two are dependent on or related to each other, which would be the matter in case of multicollinearity (Bowerman et al., 2009). The other high coefficient is the one between Hofstede’s cultural dimension of Individualism and headquarters home regions Middle East and Asia (-0.791) and North America (0.656). These two variables might capture a similar measure and, therefore, an extra test has been performed to be able to exclude multicollinearity: the variance influence factor (VIF). Multicollinearity is regarded to be present between two variables if the largest VIF exceeds a value of 10 or if the mean VIF of the factors included is substantially exceeds a value of 1 (Bowerman et al., 2009). The VIF for the cultural measure is 5.666, while the VIFs for the headquarters home regions do not exceed 10 either. All the other VIFs are just higher than 1, meaning that multicollinearity may be expected to be only slightly present in this study, although this issue should always be taken into consideration. Due to the fact that Hofstede’s (1991) scores are generalizable for a population, the scores cannot only be attributed to the countries of origin of company leaders but also to the countries belonging the different regions. Therefore, as will be shown in the results section, a separate reference model has been included to test the hypothesis regarding culture, in order to rule out multicollinearity completely.
Reviewing table 5.1 more thoroughly, several correlations will be highlighted. First, the correlation between company leader age and the number of years a company leader has spent within his or her current company is 0.410 (p <.01). This relatively high correlation could have been expected, since older company leaders have generally spent more years of their careers already, partially within their current company. Also, current company and current job are significantly related (0.324, p <.01), which is no surprise either, as the number of years a company leader has spent within his or her current job are automatically included in the number of years he or she has served the company as well. Furthermore, the only significant correlation with respect to the independent variables is the one about culture (0.215, p <.01), fostering
Variable M SD N 1. 2 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 17. 20. 1. Temporal focus 0,47 0,20 332 1 2. Loyalty (company) 21,53 13,79 323 0,018 1 3. Loyalty (job) 6,60 6,86 330 0,045 0,324** 1 4. Loyalty (insider) 0,80 0,40 323 0,029 0,576** 0,064 1 5. Age 57,85 7,65 332 0,028 0,358** 0,410** -0,019 1 6. Culture (Individualism) 63,03 26,03 328 0,215** -0,075 -0,021 0,004 -0,075 1 7. Education (bachelor) 0,35 0,48 317 0,018 0,221** 0,011 0,131* 0,078 0,053 1 8. Education (master) 0,48 0,50 317 0,058 -0,128* -0,026 -0,075 -0,119* 0,138* -0,705** 1 9. Education (PhD) 0,15 0,36 317 -0,099 -0,163** -0,066 -0,092 0,049 -0,259** -0,316** -0,408** 1 10. Education (B and E) 0,43 0,50 317 0,092 0,021 0,035 -0,010 -0,014 0,156** 0,156** -0,029 -0,127* 1 11. Education (Beta) 0,21 0,41 317 -0,067 0,010 -0,004 -0,022 -0,005 -0,161** 0,007 -0,092 0,140* -0,456** 1 12. Education (other) 0,11 0,31 317 -0,014 0,027 -0,103 0,081 0,089 0,030 0,130* -0,086 -0,037 -0,307** -0,184** 1 13. Education (B and E & Beta) 0,13 0,33 317 0,02 -0,025 -0,065 -0,003 -0,056 -0,050 -0,202**
0,152**
0,072 -0,332**
-0,199**
-0,134*
1 14. Education (B and E & other) 0,10 0,30 317 -0,064 -0,102 0,024 -0,062 -0,014 -0,010 -0,138* 0,143* 0,000 -0,292** -0,175** -0,118* -0,127* 1 15. No education 0,02 0,13 317 -0,012 0,127* 0,248** 0,065 0,029 -0,011 -0,094 -0,121* -0,055 -0,110* -0,066 -0,045 -0,048 -0,042 1 17. HQ North America 0,31 0,47 332 0,167** 0,074 0,108 0,152** -0,004 0,656** 0,046 0,127* -0,224** 0,082 -0,169** 0,054 0,018 0,033 -0,035 1 20. HQ M iddle East/Asia 0,31 0,47 332 -0,259** 0,245** -0,010 0,108 0,156** -0,791** 0,065 -0,151** 0,130* -0,174** 0,178** 0,066 0,011 -0,027 -0,024 -0,456** 1,000 ** p < .01 * p < .05
expectations regarding support for the culture hypothesis.
Looking at the descriptive statistics, the independent measure gives a mean of 0.47, indicating slightly short-termist time horizon. Regarding culture, company leaders score, on average, little over 63 on Hofstede’s Individualism scale, based on their nationalities, indicating a slightly individualistic sample overall. Furthermore, company leaders have, on average, spent 21.53 years within their company as a whole and 6.60 years in their current role, while 80% had been employed within the company before taking on their current position. With respect to education, 35% have obtained a bachelor’s degree as their highest level, whereas 48% also obtained a master’s degree and 15% even a PhD degree. The remaining 2% have had no form of education in this sense. Concerning the type of education background, the majority has obtained a Business and Economics degree (43%). Further, 21% obtained a Beta degree and 11% another type of education. Furthermore, almost a quarter (23%) has a combined background in Business and Economics and either Beta (13%) or another type of education (10%). Again, 2% have followed no specific education type. Last, on average, company leaders are 57.85 years old.
Regarding the control variables, 97% of the sample are males. With respect to home region, most companies are based in Europe (32%), followed by North America and the Middle East and Asia region (both 31%), South and Central America (3%) and Oceania (2%). The average company age is 75.78 years. With respect to the type of industry, the manufacturing industry (36%) and financial industry (27%) are most prominent, followed by the trade and motor industry (12%), the primary industry (8%), the IT industry (6%) and the energy industry (4%). The rest category of other industries occupies the final 8%. Moreover, 87% of the companies in the sample are publicly listed. Finally, the average number of employees amounts to 138,607.
5.2 RESULTS
Multiple regression analyses have been performed for the different models used to test the hypotheses. These analyses have been performed to investigate the ability of measures regarding culture, loyalty, education and age, to predict the temporal foci as displayed by company leaders in letters to shareholders. During these analyses, gender, company age, whether a company is publicly listed or not, the company size expressed in number of employees, headquarters home region and industry type are controlled for.
The results in table 5.2 show all the models used for the hypotheses. Two reference models have been included. Reference model 1b includes all the control variables as discussed earlier. However, reference model 1a has excluded headquarters region due to the probable multicollinearity with respect to the hypothesis testing the relationship between culture and temporal focus as displayed by company leaders. Hypothesis 1 investigates the effect of culture, expressed in Hofstede’s score on the cultural dimension of Individualism, on company leaders’ temporal foci. From the results, it can be concluded that a significant effect appears to be present, providing support for hypothesis 1 (Model 2: β = 0.209, p <.001). Again, the reference model used here is model 1a in order to rule out multicollinearity, whereas from now on the reference model will be 1b, including headquarters home region as a control variable. Looking at loyalty, hypothesis 2a was not supported, since the independent variable concerning the years a company leader has spent within his or her company does not show a significant value (Model 3: β = 0.086). This suggests that loyalty towards a company throughout a company leader’s career does not influence the temporal focus as displayed in the letter to shareholder in annual reports. Model 4, testing hypothesis 2b, also does not provide significant results regarding loyalty, this time with respect to job loyalty. The number of years a company leader has spent in his or her current position has no influence on temporal focus (β = 0.044). Although there appears to be no direct relationship, whether a company leader used to work within the company
before his or her current occupation has been used as a possible moderator. However, like the previous hypotheses on loyalty, there is no support for hypothesis 2c to be found in model 5, since both the main effect of the number of years in the current job (β = -0.028) and the interaction variable (β = 0.083) are not significant. Now, turning to the hypotheses on education, the effect will be tested on two dimensions: education level and education background. Model 6 presents the results regarding the first hypothesis. Hypothesis 3a states that higher education
allows a company leader to be able to better see the big picture and, therefore, to take into
Table 5.2 – Regression analyses
account all stakeholders of the company as a whole. However, the results provide no significant relationships with temporal focus for company leaders who have obtained a master’s degree (β = -0.443) or even a PhD degree (β = -1.002). Second, hypothesis 3b in model 7 investigates the
Model 1a Model 2 Model 1b Model 3 Model 4 Model 5 Model 6 Model 7 Model 8
Dependent variable: temporal focus Reference Hypothesis 1 Reference Hypothesis 2a Hypothesis 2b Hypothesis 2c Hypothesis 3a Hypothesis 3b Hypothesis 4
Constant 0.372*** 0.285*** 0.363*** 0.353*** 0.355*** 0.359*** 0.374*** 0.351*** 0.272* Independent variables Culture (Individualism) 0.209*** Loyalty (company) 0.086 Loyalty (job) 0.044 -0.028 Loyalty (insider) 0.009 Education (master) -0.443 Education (PhD) -1.002 Education (B and E) 0.022 Education (beta) 0.017
Education (B and E + beta) 0.021
Education (B and E + other) -0.088
Education (none) -0.011
Age 0.065
Interaction variable
Loyalty (job) * Loyalty (insider) 0.083
Control variable: company leaders
Gender 0.032 0.034 0.060 0.052 0.057 0.053 0.060 0.062 0.057
Control variables: firms
Company age 0.098 0.045 0.034 0.027 0.039 0.041 0.033 0.032 0.031
Publicly listed 0.042 0.044 0.064 0.064 0.064 0.062 0.062 0.070 0.066
Employees -0.047 -0.030 -0.031 -0.027 -0.035 -0.035 -0.026 -0.035 -0.033
Control variables: headquarters home region
North America 0.103 0.084 0.097 0.090 0.093 0.110 0.099
South/Central America 0.112 0.102 0.111 0.106 0.107 0.111 0.115*
M iddle East/Asia -0.188** -0.219** -0.189** -0.194** -0.190** -0.185** -0.200**
Oceania 0.009 0.010 0.009 0.009 0.008 0.011 0.016
Control variables: industry type
Primary 0.057 0.055 0.015 0.017 0.021 0.022 0.016 0.038 0.018
M anufacturing 0.015 0.011 0.005 -0.003 0.010 0.007 0.007 0.007 0.005
Energy 0.105 0.106 0.086 0.101 0.091 0.093 0.093 0.097 0.083
Trade and motor 0.002 -0.025 -0.031 -0.029 -0.024 -0.026 -0.035 -0.025 -0.029
IT 0.041 0.022 -0.005 -0.003 -0.002 0.002 -0.005 -0.001 -0.008 Financial 0.127 0.115 0.106 0.110 0.111 0.111 0.109 0.120 0.106 Model fit N 332 328 332 323 330 323 317 317 332 Adjusted R2 0.005 0.043 0.069 0.071 0.068 0.063 0.064 0.062 0.070 * p < .05, ** p < .01, *** p < .001
