Is the composition of CEO compensation dependent on the national culture?

(1)

Master Thesis:

Is the composition of CEO compensation

dependent on the national culture?

A research on the relation between the culture values of Hofstede and the annual salary, the annual bonus, the accumulated equity compensations and the long-term incentive plans of a CEO.

Dirk R. Rozendaal

Author: D. R. Rozendaal Studentnumber: 1475118

Date: 10-11-2011

Place and country: Groningen, the Netherlands University: Rijksuniversiteit Groningen Faculty: Economics and business Master: MSc Accountancy Supervisor: Dr. B.Qin

(2)

Foreword

Dear Reader,

In front of you is my master thesis. The impact a national culture can have on the CEO compensation is chosen as a subject because of my personal interest. It was during my time as an editor for the magazine of my study association that the CEO compensations found my interest. I was intrigued by the dazzling numbers I saw while doing some research. And after reading the book “Barbarians at the gate” I knew I would write my master thesis on this subject.

After some travels to Central-America and Asia I was intrigued at the big differences that are still visible in the world. I think that the impact of these different cultures is often neglected when we discuss economical problems. While writing this thesis there are discussions by the politicians in Europe on how to prevent Greece from dragging the rest of Europe in a bankruptcy. Many discussions on this subject include an analysis of the Greece culture, so there is actually no better time to point out the still existing major differences in national cultures.

By examining the impact a national culture can have on CEO compensations I will contribute to the general understanding how a CEO compensation is comprised. From the viewpoint of one culture it is sometimes hard to understand the underlying thoughts in another culture when the CEO compensation is determined. With the findings of this research it becomes clearer what cultural values are capable of explaining the differences.

This is the final product of seven years of studying. Luckily, those seven years consisted out of more than just studying. In these seven years I made a lot of new friends and kept some good friends from earlier years. After four years of study I decided to start working at an large auditor firm, because to my opinion a study is much clearer when you know how things roll in real-life. Luckily, my employer gave me enough opportunity to work and study next to each other.

Some special thanks are certainly in place here. First of all I would like to thank my supervisor Dr. B. Qin for his patience, understanding, flexibility and especially his very fast and useful feedback and Dr. R.B.H. Hooghiemstra for his and time and effort as co-assessor. Someone else that had to bring up a lot of patience in the last months is Eline, who deserves a special thanking from me because of that. She made sure I kept on working when the motivation from my side was low and had to spent a lot of evenings alone in the living room because of that. And I should off-course thank my dear friend Jasper, who helped me out with his knowledge of the English language.

Last, I would like to thank you as a reader for your time and interests in the research I conducted.

Dirk Rozendaal

(3)

Executive Summary

Recently there seems to be more interest in the excessive levels and rise in the CEO compensations. There are multiple reasons that could be thought to cause this rise in interest, but most theories trying to explain the rise are based upon studies conducted in the U.S. Since different cultures have a different view on money, they will also have a different view on the CEO compensation. The objective of this thesis is therefore to identify whether cultural values could have an influence on the CEO compensation. The main research question is: What is the relation between the values of a culture and the composition of the CEO compensation?

To answer this main research question three important theories are used as a basis for the reasoning; the principal–agent theory, the expectancy theory and the tournament theory. A big interference for these theories is the managerial power theory. A CEO compensation is composed out of an annual salary (AS), an annual bonus (AB), non-performance-based equity pay (EC) and performance-based equity pay (LTIP). The purpose of the equity compensations is to align interests of the CEO with those of the shareholders. This is being achieved with the total accumulated equity compensations and not the annual granted equity compensations, so the total accumulated equity compensations is used in the analysis.

Since each country has its own national culture, differences in CEO compensations between countries can be explained by examining cultural differences in relation with CEO compensations. The studies of Hofstede are used as a basis to classify a culture into values.The five values used to describe a culture are uncertainty avoidance (UA), individualism (IND), power distance (PD), masculinity (MAS) and long-term orientation index (LTO).

The CEO compensation components are analysed as a ratio of the annual granted compensations (TC) to make the international comparison more equally. The identified relations can be summarised as follows (table extracted from Chapter 5). The main research question is answered with these results.

Table 7: Conclusions

Variable Abbreviation AS/TC ratio AB/TC ratio EC/TC ratio LTIP/TC ratio

Independent

Uncertainty Avoidance UA n.a. n.a. (-) (-)

Individualism IND (-) n.a. n.a. n.a.

Power Distance PD (-) n.a. (+) n.a.

Masculinity MAS (-) (+) n.a. n.a.

Long-term Orientation LTO n.a. n.a. n.a. n.a.

The four most important conclusions are:

• The culture value uncertainty avoidance has a negative impact on the CEO’s equity part of the compensations, which are the total compensated equity and the total compensated long-term incentive plans.

• If the culture value masculinity has a higher value in a country, annual salaries are expected to be used less as a compensation while the use of the performance compensation annual bonuses are expected to be used more often.

• If the culture value power distance has a higher value in a country, the use of direct compensations (in particular annual salary) will be less while the use of equity compensations will be higher.

• The CEO’s annual salary is dependent on the culture values individualism, power distance and masculinity.

(4)

Chapter 1: Introduction

The problem that will be researched is introduced in this chapter. This is being translated in a research objective. Also, the relevance of contributing to a solution of this problem is described and the main results are shortly presented. Last a description of the organization on this thesis is provided.

1.1 The public debate on CEO compensations

“Not really responsible…” that is how former ING Group CEO Cees Maas commented on the bonus that Jan Hommen, the present CEO ING Group, received. In March 2011 the Dutch society showed distress when one of the largest banks in Holland, the ING Group announced to reward its CEO Jan Hommen a bonus of 1.2 million Euros, next to his fixed salary of 1.3 million Euros. At the moment of the announcement the ING Group had a public debt of 5 billion Euros with the Dutch Government, originating from the recent financial crisis. Because of the public’s anger, the Dutch Minister of Finance Jan Kees de Jager announced a new law by which companies are not allowed to pay a bonus to executives while the company still has an outstanding public loan (NRC Handelsblad, March 20, 2011).

During the last years it is not unusual to hear both national and international politicians actively debate on the subject CEO compensation. In March 2009 President Obama publicly expressed his anger at the 165 million dollar bonuses pledged to executives of insurer AIG, calling the payments “an outrage” (BBC News, march 17, 2009). Keep in mind that AIG received bailout payments from the US Government totalling 180 billion dollar, since its collapse in 2008.

According to Murphy (1999) there are three major reasons why societies show more interest and criticism on the rising CEO compensations.

1. First there is the steep rise in the level of CEO compensations, much more than the rise in the average salaries in a country. The steep rise in the CEO compensation is visible in Figure 3 and 4 in the Appendix. The median cash compensation paid to S&P 500 CEO’s has more than doubled since the seventies, while the median compensation including cash rewards nearly quadrupled (Murphy, 1999).

2. The second reason is societies perception that high CEO compensations go hand in hand with major lay-offs, the closing of company locations and corporation downsizing. This perception has been created because in the seventies and eighties business decisions started to reflect the view that the formerly established compensation packages were inappropriate in the economies of the eighties and nineties. In these economies the focus was on creating shareholder value and unions started making concessions like downsizing and massive lay-offs. Concessions that were unthinkable in earlier years. Therefore the labor unions, who faced an inevitable decline in their economic and political influence, and frustrated workers who faced possible unemployment started to show more critique on the rising CEO compensations (Murphy, 1997).

3. The so-called Bull Market in the 1990’s is the third reason. This is a period in which the securities prices constantly rose. The CEO compensations are increasingly tied to company stock-price performance and in combination with the Bull Market of the nineties this resulted in steep increases in the equity based compensations a CEO receives (Murphy, 1999). Although this is a perfectly legitimate reason and it only shows CEO’s are smart enough to tie their compensation to the results of a company when it matters, it did not help to diminish the public criticism.

(6)

1.2 Different cultures…different opinions

Although there are enough anecdotes of prominent people that disapprove the seemingly high CEO compensations and the steep rise in it in the last years, the matter of CEO compensation is more complicated than some ethic sense of what the public (and therefore also politicians) think is fair. Most of the empirical research conducted with the purpose to give an opinion whether the CEO compensation is efficient is rooted in the principal-agent theory. Chapter 2 will elaborate more on the base theories of CEO compensation. The growing focus in the past decades on this theory slowly shifted compensation schemes toward a greater use of incentives whereby the CEO would profit when shareholders did (Wirtz, 2006). Alongside the Bull Market of the nineties this resulted in a steep rise in the CEO compensation, as can be seen in Figure 3 (Appendix).

The empirical research that tries to explain the increase in the CEO compensation during the last decades leads to mixed conclusions. Sundaramurthy et al. (2005) state that the empirical evidence regarding the relation between CEO compensations and firm performance is quite mixed (Sundaramurthy et al., 2005). The existing theories do not provide a fully convincing explanation for the remarkable change that started in the seventies (Frydman & Jenter, 2010). Murphy (1999) states the following: “Although the theoretical and the empirical literature on CEO compensation is fairly well developed, it is far from complete and there are many issues worthy of continued research. For example, while the recent increase in CEO pay level is well documented, the factors underlying the trend (…) are not” (Murphy, 1999, p. 53).

The majority of the studies on CEO compensation have been conducted in the US and are focused on a US context (Barkema & Gomez-Mejia, 1998). The existing theoretical approaches seem most applicable to large companies located in the US and the U.K. Especially in the last decade academic literature questions the applicability of the existing theories (e.g. the principal-agent theory) in different (institutional) environments of an organisation (Bruce et al., 2005; Fligstein & Freeland, 1995). The principal-agent theory is even being mentioned as under-socialized (Aguilera & Jackson, 2003), mainly because the compensation for CEO’s in Europe is subject to different national corporate governance models which are based on the values of the United States, even though it is questionable whether these are applicable in other countries. Zeitling and Herrigel (2000) called this adaption the ‘Americanization’ of CEO pay in Europe in their book “Americanization and its limits”. This is being stressed by Fligstein & Freeland (1995) who conclude there is no convergence between societies. “…different societies define property rights and the rules of competiotion and cooperation in different ways” (Fligsteind & Freeland, 1995, p.40). This leads to different types of interdependencies and different ways of managing and resolving these problems. (Fligstein & Freeland, 1995). All in all, this makes it questionable whether the available academic literature is usable in other cultures besides the one it originates from. The level and composition of CEO compensation can be better explained by means of a social comparison process than the economic performance of a firm (O’Reilly III et al., 1988).

According to Zelizer (1994) money can have different meanings in different cultures. It is not possible to compare different kinds of income, even if the total sum is equal. Forms and amounts of payments also have a symbolic value. A Christmas bonus or an annual bonus perceived for achieving a target can have a different meaning, while the amount may be the same. A payment by results is more restrictive than a payment by time when looking at the autonomy a CEO receives (Tosi & Greckhamer, 2004). Since different cultures have a different view on money, they will also have a different view on the type of CEO compensation (Bloch & Parry, 1998; Gomez-Mejia & Welbourne, 1991).

1.3 Research objective

As can be read above there is a significant change in recent years in both the level and the composition of CEO compensation. The rising popularity of the agency theory combined with the Bull Market from the nineties resulted in an extensive rise of the proportion of equity based compensation to total compensation. In Figure 4 it is shown that in 1992 the direct compensation (salary and bonus) took about three quarters of the total compensation. In 2008 the proportion of equity based

(7)

compensation (stocks and options) is almost twice as large as the proportion of direct compensation. Although many CEO compensations packages are based upon the agency theory, including incentives (like equity based compensation) whereby the CEO would profit when shareholders did (Wirtz, 2006; Murphy, 1999), “…there is now recognition that many boards have employed compensation arrangements that do not serve shareholders’ interests” (Bebchuk & Fried, 2006). It seems that the agency theory is not applicable in different institutional environments (Bruce et al., 2005; Fligstein & Freeland, 1995).

According to O’Reilly III et al. (1988) the CEO compensations can be better explained by means of a social comparison process. Since there is no convergence between different societies, it is questionable whether the existing theories could be used in every culture. Types and amounts of payments can have a different meaning in a different culture (Tosi & Greckhamer, 2004).

The objective of this thesis is therefore to identify how a background culture of a CEO could have influence on the CEO compensation. Since culture is a broad understanding the focus will be on specific cultural values, which are more extensively described in the theoretical framework. The theoretical framework also describes why the research question is focused on the composition of the CEO compensation.

The main research question is:

What is the relation between the values of a culture and the composition of the CEO compensation?

To answer the main research question and develop a research method, the following sub-questions will be answered in the next chapter.

1. What comprises CEO compensation?

2. Which acknowledged theories lie at the base of CEO compensation? 3. What is culture and how can it be defined?

4. Which models describe culture values?

1.4 Relevance

Along with many other researchers, Choi & Meek (2005) and Kreitner et al. (2002) showed that cultural differences in values, attitudes and behaviours among companies operating in different countries are shown to have an impact on business activities, but they did not specifically look at the effect on CEO compensations. In recent years, not only the level of CEO compensation, but also the amount of papers written about it took a steep rise (Figure 3 in appendix). Still, Kabir (2008) expressed a recommendation that more research should be conducted to study the relationship between country specific characteristics and the compensation of the CEO. “Societal culture reflects the institutions of society, but is represented in the relatively stable values, attitudes, and behavioral assumptions of individuals. By focusing on this more fundamental and stable construct we present an opportunity to understand systematic variation in individual responses that has relevance for international management” (Thomas & Au, 2002).

Pennings (1993) was one of the first researchers to conduct research on the cross-cultural context in relation with CEO compensations. Pennings calls for more cross-cultural research, since his own focus was on countries from which the cultural values are not very divergent. Before, many of the research had a US origin (Pennings, 1993). Besides the relation between CEO compensation and a national culture, there is also a growing realization that there should be conducted more research at the influence of stock and option holdings in a CEO compensation (Abowd & Kaplan, 1998).

(8)

Multinational companies need to realise that they need to adapt values of a national culture in their decision taking in order to achieve high business performance. Numerous examples show that US management practices will not be effective elsewhere (Newman & Nolan, 1996). It is as important as adaption to customers’ needs to adapt local shares and values for a company to operate effectively (Leung et al., 2005). Lazear and Gibbs (2009, p. 341) state: “Overall, the evidence suggests that there can be substantial loss of value when companies do not pay adequate attention to incentives and governance of top management. And, the opposite also appears to be true: When management incentives are well designed, performance increases, and firms are more innovative and dynamic (Lazear & Gibbs, 2009).

Much of the existing literature is US based and is therefore based upon the US cultural values (Tosi & Greckhamer, 2004). By examining which cultural values have a relationship with the total CEO compensation, management can anticipate to which extent the results of the existing literature can be used in countries other than the US. This way, the dialogue on CEO compensation can be broadened in the context of other cultures (Fligstein & Freeland, 1995).

This thesis is a contribution to the academic literature on CEO compensation. In many articles the difference between the CEO compensation in relation with national culture is researched in a small amount of countries (mostly two or three). This research will include the data of a large amount of countries, which makes it possible to formulate a more general conclusion on the relation between the national culture and the CEO compensation. Hofstede was one of the first researchers that developed measurable cultural values of a nation culture. There are a lot of academic papers with conclusions in favour of these values, but also those that are not. By using his values in this thesis, it is possible to contribute to the discussion on the reliability of this values.

1.5 Main results

This research shows the relationship between different cultural values and the composition of a CEO compensation. Whether it is a politician in a public debate, a researcher on CEO compensations or cultural values, a member of a compensation committee or a CEO himself, the findings of this thesis enlightens the reasons why different countries have differences in the composition of their average CEO compensations.

The four most significant conclusions are:

• The culture value uncertainty avoidance has a negative impact on the CEO’s equity part of the compensations, namely the total compensated equity and the total compensated long-term incentive plans.

• The culture value masculinity creates a shift in the CEO’s direct part of the compensations, namely from the annual salary to the annual bonus when the value is higher.

• The culture value power distance is a predictor for the shift from direct compensation to equity compensation when the value is higher.

• The CEO’s annual salary is dependent on the culture values individualism, power distance and masculinity.

1.6 Organization

The remainder of this thesis is organized as follows. The answers to the formulated sub-questions will be answered in Chapter 2 and will be summarized at the end of Chapter 2. In Chapter 3 the identified theoretical framework is used to formulate the hypotheses, after the multiple variables are identified and described. Last, the research design including the multiple regression models is formulated. In Chapter 4 the results from the statistical tests are described. These are analyzed and adjusted if necessary. In Chapter 5 the conclusions are drawn, while the most significant conclusions are already presented in this Chapter. Last, the identified limitations and suggestions for future research are presented.

(9)

Chapter 2: Theoretical Framework

The sub-questions that were introduced in the previous chapter will be answered in this chapter. In the first part the composition of CEO compensation will be explained. In the next part the basic theories which are relevant for CEO compensation are summarised. These include the principal-agent theory, the expectancy theory and the tournament theory. The managerial power is a big interference for these theories, and will therefore also be discusses in this part. In the third part (national) culture is defined and reviewed and in the fourth part the cultural values defined by Hofstede are summarised.

2.1 CEO compensation

The CEO compensation can consist out of five basic components: the annual salary, the annual bonus, received stocks , received options and the long-term incentive plans. Next to the basic elements a CEO can receive contributions for his or her pension plan and a variety of other benefits (Bebchuc & Fried, 2010; Frydman & Jenter, 2010). Each of these has in total and separately different connotations, meanings and implications that will be a function of the cultural values that reflect CEO compensation (Pennings, 1993, Tosi & Greckhamer, 2004).

The composition and the level of the CEO’s compensation have endured a lot of change in the last 30 years as can be seen in Figure 3 and 4 (appendix). Till the fifties of the past century a CEO compensation was mainly composed out of a salary and an annual bonus. In the sixties the long-term incentive plans, which are bonus plans based on multi-year performance, took a remarkable rise in the composition of CEO compensation. In the early eighties the use of stock compensations rose, with the purpose to align the interest of the shareholder and the CEO. In the nineties stock options became the largest component of the CEO’s compensation until 2006, in which restricted stocks took a larger share in the total compensation (Frydman & Saks, 2010).

CEO compensation increases with the size of an organization, although recent data suggest that the relation has weakened over time. The elasticity of cash compensations to company size is constant between different countries (Murphy, 1999). When making an international comparison there is evidence that a US CEO is being paid more than his international colleagues. He is also paid differently in terms of composition. For example, a US CEO receives a larger fraction of his pay in options and less in salary compared to any CEO in other countries (Murphy, 1999).

The annual salary

The public, but also the CEO himself, devotes substantial attention to the annual salary. Although the relative proportion of the annual salary to total compensation is declining the last years, the amount of attention is not. The other components are most of the time expressed on basis of the annual salary. Bonuses and pension plans are usually expressed as a percentage of the salary while options and long-term incentive plans are expressed as a multiple of the annual salary (Murphy, 1999).

The annual bonus

The typical annual bonus consists out of a threshold performance and a cap. When performing below this threshold there are no bonuses paid. The threshold marks the point where a minimum or no bonus at all is being paid. The target bonus is paid in relation with the performance standard until a certain cap (usually a percentage of the total compensation) is reached (Lazear & Gibbs, 2009; Murphy, 1999). The annual salary and the annual bonus together are called the direct compensations (Hall & Liebman, 2000).

Stock and options

The equity linked compensation contains the stocks and options included in the CEO compensation and can be divided in a short-term period and a long-term period (Bebchuk & Fried, 2010; Core et al., 2003). The short term stocks and options are an important part of the total CEO compensation contract. “Options are contracts with a CEO that give him the right to buy a share of stocks at a pre-specified exercise-price for a pre-pre-specified term” (Murphy, 1999, p. 16). The equity linked

(10)

compensation has become a popular part of CEO compensation (Core et al., 2003). The stock options provide a direct link between the rewards of the CEO and the performance of the company (Frydman & Jenter, 2010). When the capital markets acknowledge that a company has performed well, this will be visible in the share price of the company. With a rising share price the restricted option from the CEO gains value. It is important to make a distinction in the cost for a company for granting the option (intrinsic value) and the value to a CEO from receiving the option (expected value), since these are two fundamentally different valuations (Murphy, 1999). The intrinsic value is the difference between the underlying stock price and the strike price. But if the owner of a option believes the value of the underlying stock will rise or fall (depending it is a call or putt option) the option can have a bigger value to the owner. A model to measure this value that is commonly used is called the Black Scholes model (Core & Guay, 2002).

Long-term incentive plans

Many companies offer long-term incentive plans (LTIPs) to their CEO. In most cases they consist out of restricted stock and option plans. It can also be a bonus plan that is based upon average cumulative performance for a longer period, mostly three to five years (Bebchuk & Fried, 2010; Lazear & Gibbs, 2009; Core et al., 2003). The granted stocks and options are restricted if the CEO does not have the outrights for the stock. Ownership of the stock vests over time, so the CEO has an incentive to stay with a company. At the time the vesting period ends the CEO has the right to cash out the stock (Bebchuk & Fried, 2010). Recently the debate about CEO compensation started to focus on the

proportion of the LTIP to the total CEO compensation (Bruce et al., 2005).

Pension and other benefits

Although there are a variety of ways and forms of other benefits, they are usually not part of the data used in research because it is hard to determine the current value and they are not equally disclosed (Murphy, 1999). An exception is the pension benefits. Bebchuk and Jackson (2005) found that there are large pension claims in a small sample of S&P 500 CEO’s, around 35 percent of the total compensation throughout her tenure. “Ignoring the pension benefits can cause a serious underestimation of total CEO pay” (Frydman & Jenter, 2010). It depends whether it is a defined contribution pension or a defined benefit pension to classify the pensions respectively as direct income or a part of a LTIP (Bebchuk & Fried, 2010; Core et al., 2003).

2.2 The basic theories on CEO compensation

The amount of literature on CEO compensation is immense and is several centuries old, although the majority of the studies is performed during the last eighty years (Barkema & Gomez-Mejia, 1998; Murphy, 1999). Some of the earliest research conducted on the subject can be read in Adam Smiths’ book “The Wealth of Nations” (Smith, 1776). Although this book originates from the year 1776, Smith’s views remain central to our current western society. In his book Smith describes that a business owner can hire a trustworthy manager to completely separate himself from the managing duties. This way he will diminish himself solely to the profits of his interest ownership (Smith, 1776). Berle and Means (1933) were one of the first to mention that this separation could lead to incentive problems since decisions are taken by someone else than the owner (Berle & Means, 1933). There is a need to align the interest of the owner and the manager (Scott, 2006). A substantial body of theoretical articles, starting with Jensen Meckling (1976), shows that pay for performance plans can be useful in aligning the incentives of mangers with those of shareholders (Erickson et al., 2005). Figure 3 shows that in line with the compensations CEO's received, there is a distinctive rise in the amount of studies conducted since mid-1970 (Frydman & Jenter, 2010). There are however three classic studies, on which a lot of the following studies are based, that is the principal-agent theory, the expectancy theory and the tournament theory (Tosi & Greckhamer, 2004).

Principal-agent theory

Probably one of the best-known articles about the incentive problem is written by Jensen & Meckling (1976) and is called the principal-agent theory, or more commonly used, the agency theory. Most of the empirical research conducted with the purpose to give an opinion whether the CEO compensation

(11)

is efficient in relation with the firm performance is rooted in the principal-agent theory (Murphy, 1999). This theory describes the problem of motivating the agent to perform certain acts that are costly to him but are useful for the principle, while a direct observation by the principle would be costly. The relationship between the principal and the agent is called the agency relationship and can be defined as a contract to motivate a rational agent to act on behalf of a principal when the agent’s interests would otherwise conflict with those of the principal (Scott, 2006; Jensen & Meckling, 1976). Because a direct observation would be too costly for the principal, he does not know to what extent a contract has been satisfied while the agent on the other hand knows this exactly. This information problem is called information asymmetry and is closely related to the moral hazard problem (Holmström, 1979). The moral hazard problem may occur when a party paying for the negative consequence of a risk (the principal) has less information than the party that is taking the risk (the agent), since this last party does not take the full consequence for his actions and therefore has a tendency to act less carefully (Holmström, 1979). By establishing an appropriate mix in incentives and monitoring the agent the principal will be able to limit these deviations. This optimal executive contract avoids opportunism and malfeasance by the agents (Conyon, 2006).

The conflict of interest between the shareholders of a publicly owned company and the CEO of the same company can be defined as a principal-agent problem. In some recent research done by Jensen himself he concludes that a way of attracting, maintaining and motivating the CEO to pursue shareholders interest is a performance based pay contract (Jensen et al., 2004). If the incentive for the CEO would only include a fixed compensation, such as an annual salary, the agent is able to pursue inappropriate strategies that could create unnecessary risks for the shareholders, while the CEO does not take the full responsibility. The CEO is shielded for the consequences of the risks the CEO himself takes. Therefore the compensation for a CEO would be partly variable, as to shift some of the risks the shareholders bear to the CEO (Pennings, 1993). According to Bebchuk and Fried (2006) it is not the absolute level of CEO compensation that is the subject of the principal-agent theory, but the composition of the CEO compensation package (Bebchuk & Fried, 2006).

Since the principal-agent theory forms the base of the CEO compensation plan, this thesis will only consider the composition of the compensation plan, and not the absolute level.

Although the main idea behind the principal–agent theory is to align the interests of the shareholders of a company with those of the management, Bebchuk and Fried (2006, p.6) state that: “…there is now recognition that many boards have employed compensation arrangements that do not serve shareholders’ interests”. The existing theories like the principal-agent theory do not provide a fully convincing explanation for the remarkable change that started in the seventies (Frydman & Jenter, 2010). Murphy (1999) states the following: “Although the theoretical and the empirical literature on CEO compensation is fairly well developed, it is far from complete and there are many issues worthy of continued research. For example, while the recent increase in CEO pay level is well documented, the factors underlying the trend (…) are not” (Murphy, 1999, p. 53).

Expectancy Theory

Although accounting researchers have the tendency to use the agency theory as a basis for explaining the level and composition of CEO compensation, the theory invokes a microeconomic premise, treating individuals in a standardized fashion. A theory that’s more based on the individual characteristics of a CEO and mainly employed by psychologists is the expectancy theory (Pennings, 1993). While the agency theory states that a CEO is being motivated to execute his job only because he is bound to do it by means of a contract, the expectancy theory states that a CEO is motivated by the satisfaction he gets from receiving a compensation he expects. In short, expectancy theorists examine rewards as motivators of performance while agency theorists examine how rewards can shift the risk from the principal to the agent so to align both parties interests (Pennings, 1993). Kesselman et al. (1974) measured a few key variables associated with the expectancy model developed by Porter and Lawler in 1968 in order to validate them. Therefore they developed a model that contains a serie of links in a chain. This model of work motivation can be summarized as:

(12)

Figure 1: The model of work motivation used in the expectancy theory

• Link 1: The effort portion; the expectancy that certain performance will lead to a certain outcome. This will increase the probability that such behaviour will occur.

• Link 2: The performance portion; the performance will in turn lead to rewards (outcome) which in turn will result in satisfaction.

• Link 3: The outcome portion of the model; because of the satisfaction received by achieving goals, the expectancies in respect to future outcomes will be modified (Kesselman et al., 1974). Although Kesselman et al. (1974) proved the existence of the modified model, the available research is mainly aimed at lower-level employees. Lawler (1983) on the other hand has described and researched executive compensations systems in accordance with the expectancy theory. He researched whether executives perceive a relationship between their performance levels and their compensation. When answered yes, and a high level of compensation is being paid to them, executives will be motivated to higher performance levels (Pennings, 1993).

Tournament Theory

The CEO compensation can be better explained by means of a social comparison process than the economic performance of a firm (O’Reilly III et al., 1988). Both the agency theory and the expectancy theory state that compensation should be based on the level of individual output. Differences in CEO compensations are thereby explained in terms of marginal performance. The tournament theory on the other hand states that in many cases the differences occur because of relative performance differences, and not absolute differences (Tosi & Greckhamer, 2004). Kale et al. (2009) found that there is a positive relation between tournament incentive and firm performance (Kale et al., 2009). It may be in the interest of the company to structure compensation in a manner where the winner takes very large sums as a way of spurring on those lower in the hierarchy. In the end, there can be only one winner, or in the case of a company there can only be one CEO (Lazear & Gibbs, 2009; Main et al., 1993; Lazear & Rosen, 1981). An agent’s efforts increase with the size of the promotion reward (Prendergast, 1999). The primary reason for the high pay agreed with the CEO may be to give those lower in the hierarchy an incentive to work hard, not to give the CEO himself the incentive to perform well. The excessive differences between CEO compensations and those of executives on the immediate lower level could theoretically be justified by the tournament theory (O’Reilly III et al., 1988).

In the tournament theory the executives directly below the CEO compete with each other to become the next CEO, or maybe even to show to the principals that they will be a better CEO than the current one. If the promotion probability is very high, the employee has incentives to slack off since incremental reductions in effort are unlikely to cause the employee to lose the promotion. But since the probability to become a CEO is low, the lower placed executives have a great incentive if the prize is large enough (Lazear & Gibbs, 2009). Researchers in labour economics use the compensation gap between the CEO and the next executive in the hierarchy as the measure for tournament incentive (Kale et al., 2009). If the promotion probability becomes smaller (because for example last year a young CEO got appointed), the tournament theory still holds. Because a CEO has to be very competitive to “win” the position of CEO, this characteristic will not diminish when the position has been obtained. The CEO will indulge in a competition with other CEO’s to satisfy his competitive urge. The comparison of compensation is a good way to measure which CEO is the best. Therefore the

(13)

incentive for the highest compensation or the best composition of the compensation will still be present, even if the lower-level companies think there is a small change for them to be the next CEO (Lazear & Gibbs, 2009; Malmendier & Tate, 2005).

For the tournament theory to work adequately it is necessary for the participants to understand the performance appraisals and they should be capable of giving constructive feedback. The risk with relative performance evaluation is a more frequent use of lower ratings as it can encourage sabotage between employees. An organisation needs to establish a corporate culture in which such behaviour is not common. The implicit contract in an organisation is that all participants are expected to contribute appropriately and constructively to the giving and receiving of constructive feedback. This implicit contract is part of the culture of an organisation (Lazear & Gibbs, 2009). Culture implies that one way of acting is preferable to another; executives may be less able or willing to perform well because their deeply held values are inconsistent with the management practices (Newman & Nolan, 1996). Since the norms about constructive feedback is different in each organisation, this suggests that culture in an organisation can be created and changed (Lazear & Gibbs, 2009).

Managerial power

The managerial power approach explains that a CEO uses his power to influence the compensation contract (Bebchuk & Fried, 2006). Mainly because of the managerial power CEO’s have, they have the power to extract a compensation that exceeds the market value. This exceeding amount is called the rent for a CEO. The executive compensations have grown beyond the increase that could be explained by firm size performance (Bebchuk & Grinstein, 2005). Because the CEO has more interests than solely his remuneration, for example an upcoming re-election, psychological factors including friendships, loyalty and a desire for authority the resulting bargaining with the shareholders will not be at arm’s length. Besides that, some members of the compensation oversight committee might favorite a CEO without a legitimate reason. What also adds to the belief that the committee might not bargain at arm’s length is the limited amount of available time and information, the use of compensation consultants and the possibility that the remuneration setting process starts after the election of a CEO (Bebchuk & Fried, 2006). When the CEO is appointed before the compensation oversight committee is appointed he is likely able to exert influence over them (Hallock, 1997).

Despite his extensive research and his critique on the CEO compensations Bebchuk and Grinstein are not able to make an estimate of the rent that is being caused because of this managerial power. It is hard to determine the ‘normal’ CEO compensation without any rent, because it does not exist in such a pure form. All existing CEO compensations have some rent included, although it is not clear to what extent this affects the amount of remuneration (Wirtz, 2006).

2.3 Different societies, different cultures

The environment of an organisation can be categorised by different factors, like demographic factors and economical factors. But it is the social cultural factors in particular that have a big impact on culture. The social cultural factors comprises the way people in a society think and the existing shares and values in a society (Besanko et al., 2004). “Societal culture reflects the institutions of society, but is represented in the relatively stable values, attitudes, and behavioral assumptions of individuals” (Thomas & Au, 2002, p.310). There are a lot of different definitions on culture. Maybe the best description therefore is “a fuzzy, difficult to define construct” (Triandis et al., 1986). The Dutch dictionary Van Dale gives the following explanation (translated): “Culture: The cumulative

programming of the mind of a country, nation etc.”(Van Dale Online, viewed January 16th_{). For this}

thesis the following explanation is preferred: “the collective programming of the mind that distinguishes the members of one category of people from those of another. Culture is composed out of certain values, which shape behaviour as well as one's perception of the world” (Hofstede & Bond, 1988, p.6).

Organizations are embedded in the societies where they are stationed (Chui et al., 2002), societies which have been developing shares and values for a long time so they are deeply rooted. But also the attitudes and behaviour of their members have been developing a long time so that a specific culture is established in the society (Tosi & Greckhamer, 2004). Cultures are deeply embedded in societies

(14)

and are therefore hard to change (Newman & Nolan, 1996). Culture can be defined on different levels, there is for example a group culture, a corporate culture and an (inter)national culture in a society. Therefore an organisation has to cope with more than one culture. All cultures have an effect on the existing practices and on the way decisions are made in an organisation, therefore the distinction between the different kinds of cultures is important (Schuler and Rogovsky, 1998). This thesis will mainly focus on multinational firms; to them the difference in national cultures is the most relevant one, as pointed out by Evans (1992).

National culture implies that one way of acting is preferable to another; executives may be less able or willing to perform well because their deeply held values are inconsistent with the management practices (Newman & Nolan, 1996). Since each country has its own national culture, differences in CEO compensations between countries can be explained by examining cultural differences in relation with CEO compensations (Pennings, 1993). According to Bloch and Parry (1989) the relationship between cultural values and CEO compensation illustrates the difference in the view on money between cultures (Tosi & Greckhamer, 2004). According to Zelizer (1994) money can have different meanings in different cultures. It is not possible to compare different kinds of income between cultures, even if the total sum is equal. Forms and amounts of payments also have a symbolic value. A Christmas bonus or an annual bonus perceived for achieving a target can have a different impact, while the amount may be the same. A payment by performance is more restrictive than a payment by time when looking at the autonomy an executive receives (Tosi & Greckhamer, 2004). The symbolic value is dependent of the deeply rooted shares and values formed by the national culture of a CEO. CEO pay is a combination of base pay, bonuses and other short-term incentives and long-term incentives. Since different cultures have a different view on money, they will also have a different view on the composition of the CEO compensation (Bloch & Parry, 1998; Gomez-Mejia and Welbourne, 1991).

Many researchers have tried to make international comparisons between compensation schemes in an organisation. See for a non comprehensive literature list Murphy (1999) and Tosi and Greckhamer (2004). Despite that, Tosi and Greckhamer (2004) state that only two studies have directly addressed culture and CEO compensation. The first is performed by Pennings (1993), in which he concludes that differences in CEO compensation are attributable to the cultural differences. The second is a research conducted by Schuler and Rogovsky (1998) in which they prove there is a relation between the cultural values defined by Hofstede and the different human resource systems in different countries (Tosi & Greckhamer, 2004; Schuler & Rogovsky, 1998). The research is difficult to perform because of the substantial heterogeneity in available data. This makes it hard to uniformly describe regression specifications which include the definitions of the dependent and the independent variables (Murphy, 1999).

When looking at the first part of the theoretical framework, two limitations of this thesis can be formulated. The objective and relevance described in the introduction are mainly aimed at multinational organisations. These organisations are constantly expanding in new countries, with their own national culture. When the corporate culture does not connect with the national culture, value will be lost for the company. Since a corporate culture can be created and changed, it can be formed in a way to optimally connect with the national culture (Lazear & Gibbs, 2009). This thesis will therefore be aimed at national cultures from now on, so to fulfil the objective optimally. The second limitation is aimed at the CEO compensation. It is hard to uniformly describe the definitions of the value of money, since each country gives its own meaning to it (Tosi & Greckhamer, 2004). Next to that it is difficult to internationally compare CEO compensations because of the substantial heterogeneity in data and the institutional details such as national taxes and the impact of the exchange rates (Murphy, 1999). Therefore this thesis will further focus on the composition of CEO compensation instead of the level. The ratios in the composition of CEO compensation will be better comparable than the absolute amount. According to different academic researchers the focus in other research is often aimed at the level of CEO compensation, while the composition is actually the subject (Lazear & Gibbs, 2009; Bebchuk & Fried, 2006).

(15)

2.4 Values of natural culture

As described before, differences in culture emerge from a variety of forces and are reflected in the social systems and institutions of a society (Shaw, 1999). The largest study attempting to classify a culture into values is conducted by Hofstede (Goktan & Saatçıoğlu, 2011; Hofstede, 2001). Besides Hofstede, there are other theoretical approaches to the structure of values at the cultural level, such as Triandis and Schwartz (Gouveia & Ros, 2000; Triandis et al., 1988). Values are considered to be the core concepts that define culture (Hofstede, 2001; Schwartz, 1999). Hofstede states that a culture can be defined in four value dimensions, after conducting a study among employees of a multinational organisation with branches in more than fourty countries (Gouveia & Ros, 2000). His work is still extensively used in research (Goktan & Saatçıoğlu, 2011, Westerman et al., 2007; Tosi & Greckhamer, 2004; Gouveia & Ros, 2000; Johnson & Lenartowicz, 1998) and has been updated with subsequent studies validating the earlier studies (Hofstede, 2002). The four basic values to describe a culture are uncertainty avoidance (UA), individualism (IND), power distance (PD) and masculinity (MAS). The fifth value Long-Term Orientation index (LTO) was added in later years after conducting an additional international study with a survey instrument developed with Chinese employees and managers (Hofstede et al., 2010; Hofstede, 2001, Johnson & Lenartowicz, 1998).

The work of Triandis et al. (1988) was based upon the cultural values described by Hofstede, but they specifically focused on the individualism and collectivism dimension (Triandis et al., 1988). Schwartz

(1994) develops his own etic1_{values of culture paralleling the values described by Hofstede (Chui et}

al., 2002). This is because he, just like Triandis, thinks the individualism value has obscured some important differences associated with the factors used to measure the value. According to him cultures can be accounted for by seven basic cultural values, namely conservation, hierarchy, intellectual autonomy, affective autonomy, competency, harmony and egalitarian compromise (Gouveia & Ros, 2000).

Research on which model is more appropriate to use does not give a decisive answer (Johnson & Lenartowicz, 1998; Drogendijk & Slangen, 2006). Researchers should therefore carefully consider which cultural distinction is more appropriate for their study (Imm Ng et al., 2007). Gouveia and Ross (2000) compared twenty countries with scores in both the Hofstede as the Schwartz model in relation with a group of macro-social and macro-economic variables in a country. Examples of social variables are the birth rate, human development and illiteracy rate. Examples of macro-economic values are gross-national product, rate of unemployment and rate of inflation. Their results show that the Hofstede model has a stronger relation with the macro-economic variables and the Schwartz model with the macro-social variables (Gouveia & Ros, 2000).

The third limitation in this thesis is thereby identified. The total CEO compensations are analyzed through an accounting and economic scope, therefore the Hofstede model will be used for further analysis. Each cultural value from Hofstede will be described briefly.

Uncertainty Avoidance

UA stands for the degree of tolerance for uncertainty and ambiguity a society has. It gives an indication to what extent the members of culture feel (un)comfortable in unstructured situations, such as an unknown and surprising situations. This leads the members to support certainty by means of strict laws and rules, safety and security measures. Some of the common traits found in countries that score high on the UA index are cultures with a long history; a non multicultural population; risks are avoided in business; and new concepts or ideas are difficult to introduce. Generally speaking Protestant countries and those with Chinese influences score low, while Catholic, Buddhist and Arabic speaking countries tend to score high on the UA index (Hofstede et al., 2010; Tosi & Greckhamer, 2004; Gouveia & Ros, 2000).

1_{Etic is a professional term used in cross-cultural research that refers to the study of behaviour from an outside view and is often related} with the study of many cultures at the same time (Chui et al., 2002).

(16)

Individualism

Societies that have a high IND index have loose ties among its members. The opposite of individualism is collectivism, the degree to which individuals are integrated into groups. When the IND index is high, it is understood by its members that individuals must take care only for themselves and their immediate family members. People emphasize only their individual needs, concerns and interest over their own group or organization. When the IND index is low and the collectivism is high the members have a high dependence on groups. These groups protect individuals throughout their lifetimes in exchange for unquestioning loyalty. The United States and the Western Europe countries are cultures with a high IND index, where personal achievements are emphasized. The collective cultures like China and Japan emphasize the group mentality (Goktan & Saatçıoğlu, 2011; Hofstede et al., 2010; Tosi & Greckhamer, 2004; Gouveia & Ros, 2000).

Power Distance

If a society has a high PD, the members expect and accept that the interpersonal power is distributed unequally among them. Both the followers and the leaders expect and accept this. Good examples of countries with a high PD are the Arabic speaking countries, but also China and Russia. Characteristics are that people with authority do openly demonstrate this; subordinates do not perform the important work, but they take the blame if anything goes wrong; and the relationship between the CEO and the employees is rarely personal. In a low PD country the employees will feel less discomfort and stress when disagreeing with the CEO, the CEO and its subordinates will interact socially and discuss work with each other (Goktan & Saatçıoğlu, 2011; Hofstede et al., 2010; Tosi & Greckhamer, 2004; Gouveia & Ros, 2000).

Masculinity

A masculine society has a preference for typical masculine values like competiveness, acquisition on wealth, heroism and severity as opposed to feminine values like relationship building, modesty, attention for the weak and quality of live. A society is called feminine when emotional gender roles assigned to men and woman overlap and both men and woman are supposed to be modest, tender and concerned with the quality of life (Goktan & Saatçıoğlu, 2011; Hofstede et al., 2010; Gouveia & Ros, 2000). In a highly masculine culture people show more concern for job performance and less for the quality of the working environment. They also show more value to aggressive ways of the acquisition of material possessions. In a highly feminine culture the emphasize goes to working conditions and employee participation (Tosi & Greckhamer, 2004; Gomez-Mejia & Welbourne, 1991).

Long-Term Orientation

The LTO is the fifth and last cultural value added by Hofstede. In his original study the difference in thinking between the classic East and West was not recognized. The thinking in the East is influenced with the teaching of Confucius, a Chinese thinker and social philosopher that lived around the year 500 BC. The philosophy of Confucius emphasized personal and governmental morality, correctness of social relationships, justice and sincerity and has developed into a system of philosophy known as Confucianism (Bloom & Solotko, 2003). Futurologist Herman Kahn called this modern thinking the neo-Confucian. He believed that the countries of East-Asia have this neo-Confucian as a common cultural root that under the world market conditions of the last decades has constituted a competitive advantage for successful business activity (Hofstede & Bond, 1988).

In short it gives an indication to what extent a culture appreciates virtue above self preservation. Members of a culture with a high LTO index emphasize characteristics like persistence, ordering relationships by status, thrift and having a sense of shame. In countries with a low LTO index characteristics like personal steadiness and stability and “protecting your face” are emphasized. Both the positively and the negatively rated values of this dimension are found in the teachings of Confucius; however, the dimension also applies to countries without a Confucian heritage (Hofstede et al., 2010; Hofstede & Bond, 1988).

(17)

Criticism on Hofstede

Although Hofstede was one of the first authors to study cultural values and his work represents the largest study to classify nations on differences in their cultural value, there are other authors who express their criticism on this classification (Goktan & Saatçıoğlu, 2011; Hofstede, 2002). Read for example McSweeny (2009; 2002), Baskerville (2003), Roberts & Boyacigiller (1984) and Cooper (1982). Hofstede (2001) reacts that it is normal for a paradigm shift in science (like he made) to meet strong initial resistance (Hofstede, 2002; 2001).

One of the standard critiques on the Hofstede Study is the generalization of a culture by equating nation states with cultures. The assumption made by Hofstede, according to the critics is that every citizen in a country acts like the cultural values will predict. Every citizen of an Arabic country will expect that interpersonal power is distributed unequally among the citizens of a country, because of the high PD index (McSweeny, 2009; 2003; Hofstede, 2002). A country can also have more than one culture. Africa for example has 98 cultures identified while there are 48 countries and in North America there are 147 Native American cultures and nine North American folk cultures (Baskerville, 2003).

As can be read in the previous part of this chapter, culture can be defined in more than one way (Triandis et al., 1986). In this thesis culture has been defined as: “the collective programming of the mind that distinguishes the members of one category of people from those of another. A culture is composed out of certain values, which shape behaviour as well as one's perception of the world” (Hofstede & Bond, 1988, p.6). Culture can be defined on different levels, for example a group culture, a corporate culture and (inter)national cultures, so it is logical there are more cultures in one country (Schuler and Rogovsky, 1998). Hofstede chose to make a distinction in cultures based upon country borders because these borders are well described and generally accepted (Hofstede, 2002). Besides that, in the research conducted by Schuler and Rogovsky (1998) they prove there is a relation between the cultural values defined by Hofstede and the different human resource systems in different countries (Tosi & Greckhamer, 2004; Schuler & Rogovsky, 1998). Hofstede (2001) and academics stimulated by his first work identified more than 400 independent cross-national differences that relate with the cultural values (Hofstede, 2002; 2001).

Another standard critique is the use of a subsidiary of a multinational firm to describe the characteristics of an entire culture (McSweeny, 2002). The first Culture’s Consequences that was published in 1980 used the data from a survey of employees at IBM subsidiaries in 66 countries between 1967 and 1973 (Baskerville, 2003). Surveys are not a suitable way of measuring cultural differences according to some authors (McSweeny, 2002). Hofstede agrees partly with this critique by acknowledging that there should be conducted more research with different methods to measure cultural differences (Hofstede, 2001). But as mentioned before, many other academics performed their own research and came up with the same results. Some did, but others did not use a survey as an instrument (Tosi & Greckhamer, 2004; Hofstede, 2002; Schuler & Rogovsky, 1998). The use of a subsidiary of a multinational firm was carefully chosen by Hofstede. His objective was to research differences between national cultures and according to him any set of functionally equivalent samples from national populations can supply information about such differences (Hofstede, 2002)

It is possible to have critique on the work of Hofstede, but it is often found that Hofstede himself was the first to express self-criticism. Although there is a very high level of generalization in the work of Hofstede, he provides all the information about the procedures he used so that one can see which specific empirical data is changed into this generalized data (Hofstede, 2002; Chapman, 1997)

(18)

2.5 Summary

The objective of the theoretical framework was to answer the sub-questions that are relevant in answering the main research question. By summarizing the above theories the answers will be briefly provided.

What comprises CEO compensation?

The CEO compensation can consist out of five basic components: the annual salary, the annual bonus, received stocks and accumulated stocks, received and accumulated options and the long-term incentive plans. Next to the basic elements a CEO can receive contributions for his or her pension plan and a variety of other benefits. The annual salary and the annual bonus together are called the direct compensation, from which the annual bonus is the performance based share. The stocks and options is the non-performance based equity linked part of the compensation, where the long-term incentive plans comprise out of the performance-based equity part of the compensation. Options provide a direct link between the rewards of the CEO and the performance of the company. Since the main research question is aimed at the impact the values of a culture have on the composition of CEO compensation, some of the described components in ratio to the total compensation will be used as the dependent variables.

Which acknowledged theories lie at the base of CEO compensation?

The principal-agent theory is one of the best known and mostly respected theories concerning management incentives. The main idea behind the principal–agent theory is to align the interests of the shareholders (principals) of a company with those of the management, including the CEO (the agents). It is not the absolute level of CEO compensation that is the subject of the principal-agent theory, but the composition of the CEO compensation package. Although the main idea behind the principal–agent theory is to align the interests of the shareholders of a company with those of the management there is now recognition that many boards have employed compensation arrangements that do not serve shareholders’ interests. The existing theories like the principal-agent theory do not provide a fully convincing explanation for the remarkable change that started in the seventies. A theory that’s more based on the individual characteristics of a CEO is the expectancy theory. Expectancy theorists examine rewards as motivators of performance while agency theorists examine how rewards can shift the risk from the principal to the agent so to align both parties interests. The tournament theory states that in many cases the differences in CEO compensations occur because of relative performance differences, and not absolute differences. The compensation gap between the CEO and the next executive is the tournament incentive. The managerial power approach explains that a CEO uses his power to influence the compensation contract. Mainly because of the managerial power CEO’s have, they have the power to extract a compensation that exceeds the market value. The existing theories do not provide a fully convincing explanation for the remarkable change that started in the seventies. Murphy (1999) states the following: “Although the theoretical and the empirical literature on CEO compensation is fairly well developed, it is far from complete and there are many issues worthy of continued research. For example, while the recent increase in CEO pay level is well documented, the factors underlying the trend (…) are not” (Murphy, 1999, p. 53).

What is culture and how can it be defined?

“Culture can be defined as the collective programming of the mind that distinguishes the members of one category of people from those of another. Culture is composed out of certain values, which shape behaviour as well as one's perception of the world” (Hofstede & Bond, 1988, p.6). Since each country has its own national culture, differences in CEO compensations between countries can be explained by examining cultural differences in relation with CEO compensations. This thesis will therefore be aimed at national cultures from now on, so to fulfil the objective optimally. Next to that it is difficult to internationally compare CEO compensations because of the substantial heterogeneity in data and the institutional details such as national taxes and the impact of the exchange rates (Murphy, 1999). Therefore this thesis will further focus on the composition of CEO compensation instead of the level.

(19)

Which models describe culture values?

The studies of Hofstede are used as a basis to classify a culture into values. The Hofstede model has a strong relationship with the macro-economic of a culture and is therefore the best suitable model. The four basic values to describe a culture are uncertainty avoidance (UA), individualism (IND), power distance (PD) and masculinity (MAS). The fifth value long-term orientation index (LTO) was added in later years. Since the main research question is aimed at the impact the values of a culture have on the composition of CEO compensation, the values described by Hofstede will be used as the independent variables. Besides Hofstede, there are other theoretical approaches to the structure of values at the cultural level, such as Triandis and Schwartz (Gouveia & Ros, 2000; Triandis et al., 1988). Research on which model is more appropriate to use does not give a decisive answer (Johnson & Lenartowicz, 1998; Drogendijk & Slangen, 2006). Gouveia and Ross (2000) results show that the Hofstede model has a stronger relation with the macro-economic variables and the Schwartz model with the macro-social variables (Gouveia & Ros, 2000). The total CEO compensations are analyzed through an accounting and economic scope, therefore the Hofstede model will be used for further analysis.

Is the composition of CEO compensation dependent on the national culture?

Master Thesis: