The universality of narcissistic leadership : a study into the role of national culture in the narcissism debate

The Universality of

Narcissistic Leadership

A study into the role of national culture in the

narcissism debate

Supervisor: Dr. Ilir Haxhi

Second reader: Prof. Suzana Rodrigues Final version: 07/07/2014

Student number: 6057004 Master in Business studies

Abstract

Narcissistic leadership is often referred to as a universal concept. This study questions the universality of narcissistic leadership by analyzing the extent to which narcissistic leadership behavior is equally distributed across borders and by determining whether cultural

differences influence the effectiveness of narcissistic leadership. Previous research already determined that leadership perceptions are culturally endorsed and this research adds to our understanding of the narcissism concept by concluding the same for narcissistic leadership. For a sample of 80 CEOs covering eight European countries we check the influence of national culture on narcissistic leadership behavior. Culture was divided into ten dimensions, eight national culture dimensions of the GLOBE project (2004) and two dimensions of Hofstede (2011).The premises of this approach is that the aggregate effect of national culture is illustrated by the partial effects of the selected culture value dimensions.

The first effect of national culture we propose is that it has an effect on the presence of narcissistic behavior at the CEO level. This study shows that the extent to which a culture is masculine instead of feminine, is centered around the individual instead of the collective, and is not threatened by the uncertainty surrounding us in our daily life, positively influences the presence of narcissistic behavior at the CEO level.

The second proposed effect is that national culture moderates the effect a narcissistic leader has on firm performance. In other words, in a country that is more accepting towards narcissism, a narcissistic CEO is able to perform better. This effect was predicted because previous research determined that certain leadership traits require a similar environment to fully flourish (Brodbeck et al., 2000). In contrast with the first effect, of which clear evidence was present in the analysis, no evidence was found of the second predicted effect.

This thesis contributes to the field of narcissism research by showing the cultural dependency of the concept. Narcissism is more in line with some cultures than other, and in such cultures narcissistic leaders are much more common.

1 Table of content - 1. Introduction 2 - 2. Literature Review 5 o 2.1 Leadership matters 5 o 2.2 Narcissistic leadership 6

o 2.3 The effect of narcissism 9

o 2.4 Role of culture 10

 2.4.1 Culturally endorsed leadership 10  2.4.2 The concept of culture 11  2.4.3 Culture and narcissism 12

o 2.5 The Hypotheses 19 o 2.6 Causal model 23 - 3. Research design 24 o 3.1 Sample 24 o 3.2 CEO narcissism 24 o 3.3 Firm performance 27 o 3.4 National culture 27 o 3.5 Control variables 28 o 3.5 Method 29 - 4. Results 30 o 4.1 Descriptive analysis 30 o 4.2 Correlation 31 o 4.3 Multicollinearity 34 o 4.4 Regression 35 - 5. Discussion 42 o 5.1 Findings 42 o 5.2 Limitations 45 o 5.3 Future research 45 - 6. Conclusion 46 - 7. Appendix 48 - 8. Literature 50

2 1. Introduction

Most of the grand leaders this world has known could be seen as narcissistic. Not only the names you would expect such as Adolf Hitler and Joseph Stalin (Glad, 2002: 33), but also figures as Steve Jobs are seen as being narcissistic in nature (Maccoby, 2000: 73). Maccoby (2000) describes narcissism as a double edged sword. Next to the negative implications that are often connected to narcissism it also has its advantages. This explains the presence of narcissistic individuals at top positions. Narcissism is a personality trait that encompasses grandiosity, arrogance, self-absorption, entitlement, fragile self-esteem, and hostility (Rosenthal & Pittinsky, 2006: 617). These traits influence the way a CEO runs his organization. In this thesis the extent to which country differences play a role in the

narcissism debate stands central. By adding national culture into the equation it is possible to compare the role of narcissism across countries.

One assumption this thesis is build upon is that leaders make a difference. Two different opposing views dominate the research debate concerning this question. The first view, also referred to as the inertial organization perspective, sees little leadership effect on the outcome of an organization. Other factors, such as firm size, the industry a firm is operating in and the overall economic conditions are seen as more appropriate to explain the variance in firm outcome (Lieberson & O’Connor, 1972: 122). Although this view has

dominated the leadership debate for years it has many critics. Thomas (1988) pointed out that the inertial organization perspective was not wrong in stating that these external factors have a much bigger influence on the outcome of a firm, but he emphasized that leaders could be responsible for all the variation within the firm (Thomas, 1988: 398).

One concept that stands central in this view is bounded rationality. By assuming that nobody has a fully rational view on reality you give power to the person in charge. Because of the inability to see the entire picture leaders do not make decisions on a technocratic basis and the way they see reality greatly influences the direction an organization goes (Hambrick & Mason, 1984: 194). The way reality is interpreted is dependent on who the leader is. Every leader has different traits and these traits influence the decision making process.

The narcissistic behavior of a CEO influences the way that reality is interpreted and consequently the outcome of the organization. The direction of this influence however is debatable. The elements that are connected to narcissism give away the negative connotation of the concept. However, besides all these negative effects, some authors have emphasized that the extreme confidence and dominance of narcissistic leaders can push an organization

3

forward. Narcissism, despite all its potential negative effects, therefore is also seen as a requirement for effective leadership (Higgs, 2009: 172).

These different effects a narcissistic CEO has on firm performance have been the subject of previous research. Narcissism is connected to a higher chance of scandals and financial downfall (Cools, 2005). Furthermore, other findings suggest that narcissistic CEOs favor bold actions that attract attention, resulting in big wins and big losses (Chatterjee et al., 2005: 376). Rijsenbilt (2011) found that when linking the level of narcissism to the

performance of an organization an interesting relationship surfaces. As previously mentioned, narcissism is often seen as a perquisite for effective leadership. If it becomes too extreme however, it can be damaging for the organization. Rijsenbilt found evidence of this

relationship taking the shape of a concave parabola, where to little or too much narcissism has a negative effect on the organization (Rijsenbilt, 2011: 122).

These previous studies picked their sample from the S&P 500. Although this brings many practical benefits it also makes it difficult to take one important factor into account. Because they generalize their results using only an American sample they make the mistake of seeing effective leadership as a universal concept (Steers et al., 1996: 423). There is not a single right way of having a positive effect on firm performance. Many elements have an influence on the conceptualization of leadership. Rijsenbilt et al. (2011) suggest that future research into the role of narcissistic leaders should take different countries and cultures into account to find out to what extent the role of narcissistic behavior differs between countries (2011: 69).

Previous research that focused on the differences of leadership perceptions between countries found that leadership traits are culturally endorsed (Koopman et al, 1999; Brodbeck et al., 2000). Even when countries inside the European Union are compared cultural

differences create a different view on what is seen as effective leadership (Koopman et al., 1999: 515). National culture had two effects on the role of different leadership traits. First of all, Koopman et al. (1999) show the difference in leadership perceptions between European countries. There especially is a clear distinction between North-West Europe and South-East Europe. Leadership in the South-Eastern European countries is focused more on power distance and masculinity than is the case in the North-Western countries (Koopman et al., 1999: 515). We argue that the same goes for narcissism. In other words, the degree to which narcissistic behavior is accepted is culturally endorsed. As a consequence, national culture directly influences the degree to which narcissistic behavior is present.

4

al. (2000) found that besides the fact that leadership perceptions are culturally endorsed, the effectiveness of different leadership traits also is culturally dependent. When certain

leadership traits align with the overall values present in a culture this enhances the

effectiveness of such leadership traits (Brodbeck et al., 2000: 12). We argue that the same goes for narcissistic behavior. The alignment of narcissistic behavior and the culturally endorsed leadership perceptions brings forward a situation in which a narcissistic CEO can perform best. These two effects of national culture are brought together in the following research question:

To what extent does national culture affect the presence and performance of narcissistic CEOs?

This thesis contributes to the field of narcissism research by introducing national culture into the debate. Previous narcissism research was conducted in one country and therefore failed to take country effects into account. As mentioned above, leadership

perceptions are culturally endorsed and in this study we determine to what extent this also is the case for narcissism. The core argument of this thesis is that narcissism is not a universal concept that is applicable the same way across borders. To discover such a relationship between national culture and narcissism a cross-national research using a sample of 80 CEOs covering eight European countries is conducted. By taking multiple countries, and therefore multiple cultures, into account we are able to see the direct and indirect effect of national culture. Three concepts stand central in the analysis; narcissism, national culture and firm performance.

Narcissism is often measured using the Narcissistic Personality Inventory (NPI). This measurement uses questionnaires to determine the degree to which a person shows

narcissistic tendencies. However, such a method is not very practical when researching CEOs and therefore unobtrusive indicators of narcissistic tendencies are used in this thesis

(Chatterjee & Hambrick, 2007: 362). These measures of indicators enable narcissism

research from second hand data, and therefore direct interaction with the research unit is not necessary. This makes it ideal for research at the CEO level.

To determine the influence of national culture on the degree of narcissism present at the CEO level cross-national differences are analyzed. National culture is operationalized by dividing culture into 10 dimensions. These dimensions are taken from the Global Leadership and Organizational Behavior Effectiveness (GLOBE) project (2004) and the work of

5

determine the extent to which narcissism is affected, and more importantly, which elements of a country’s national culture create the effect.

Besides the direct effect of national culture we predict an indirect influence on the manner in which narcissistic CEOs affect firm performance. We operationalize firm

performance by taking the average Return on Assets (ROA) and Tobin’s Q of 2011 and 2012. The ROA shows the profitability of a firms assets and Tobin’s Q is a market measure which is used to proxy the organizational performance (Rijsenbilt et al., 2011: 72). ROA and Tobin’s Q are complementary measures as ROA focuses more on previous results whereas Tobin’s Q is a measure of expected future performance.

The remainder of this thesis is accordingly structured. In the first section, the

literature review is discussed. The literature review starts off by explaining why research into leadership traits is relevant. Afterwards the main concepts of this thesis are discussed and several hypotheses are constructed. In the second section we explain the data collection and data analysis methods that are used in this thesis. The next chapter centers around the results of the data analysis. The theoretical implications of these results are then elaborated in the discussion. Finally, the conclusion summarizes the findings of this thesis.

2. Literature Review

2.1 Leadership matters

A very important question that has to be answered before we proceed is whether leaders matter within an organization. The first prominent authors who tried to answer this question were Lieberson and O’Connor (1972). They determined that only a small variance in organizational performance can be attributed to persons in top leadership positions

(Lieberson & O’Connor, 1972: 122). They used three independent variables to determine the effect of leaders. First of all they looked at the year of measurement to take into account the effect of general economic conditions. Secondly, they looked at the industry a firm was competing in. The last control variable was the firm itself. This variable should determine whether the size, location, facilities and reputation of a firm influences its performance (ibid.: 121-124). They found that these control variables influenced firm performance to a much greater extent than leadership did (ibid.: 124).

Although this view, that is often revered to as the inertial organization (Hambrick & Mason, 1984: 194), has kept a prominent role in leadership research for quite some time, the number of critics grew. Thomas (1988) discarded these critiques as being too negative. It was

6

not the outcome that was wrong, but the method that created wrongful insights. The control variables used by Lieberson and O’Connor create a situation in which it is almost impossible for a CEO to be responsible for a large fraction of the variance in performance. As Thomas points out:’ Leaders may account for all of the variation within firms while accounting for

little of the total variation if the bulk of this is generated by differences between firms’

(Thomas, 1988: 398). Looking at all these external factors, such as economic conditions and relative size of the firm does not do justice to the internal effects that a CEO has. Thomas compares CEOs with racing drivers. Car differences play a vital role, but it is often the driver that makes the difference between a victory or a loss (ibid.)

In the eighties research moved away from the inertial organization perspective and new theories were created which looked at the role of CEOs. One of the most cited is the Upper Echelons theory (Hambrick & Mason, 1984). Bounded rationality, multiple and conflicting goals are only a view of the reasons that limit the extent to which complex decisions cannot be made on a technocratic basis (Hambrick & Manson, 1984: 194). A manager has to guide a firm through a maze of uncertainties and he has a great influence on the direction. A manager cannot make such decisions on an entirely rational basis. First of all, a manager only has a restricted field of vision. Secondly, a manager only selectively

perceives some of the phenomena that present themselves within his field of vision. Finally, these phenomena that the manager does perceive are also interpreted by him (ibid.: 195). This means that the way top management sees reality is highly influenced by the sort of managers that are present.

2.2 Narcissistic leadership

The Upper Echelons theory tries to link visible leadership traits to managerial decision-making by focusing on the age, functional tracks, other career experiences, education, socioeconomic roots, financial position and group characteristics of top managers. They predicted that these elements have an effect on the way that managers interpret the reality and consequently it has an effect on the strategic choices made by management (Hambrick & Mason, 1984: 198). These traits were chosen because they are easy to observe and therefore easy to collect. Hambrick and Mason however conclude that it was not a desirable way of researching top management and more attention should be paid to the psychological side of managers (ibid.: 203).

Narcissism is a psychological concept that is often linked to people in leadership positions. Emmons (1984) divides narcissism into four factors. The extent to which these

7

factors are applicable to a person determines their level of narcissism. The first factor is called entitlement. Narcissistic personalities have the feeling that they are entitled to more than others and they are willing to use extreme measures to make sure that they get what they want. The second factor emphasizes the notion that narcissistic individuals enjoy being in leadership positions. This coincides with the fact that people in leadership positions are often seen as narcissistic. The third factor looks at the superior feeling of narcissistic individuals. They have the feeling that they are untouchable. Consequently they believe that they will always find a way out of complex situations. A consequence of this immensely arrogant feeling is that they do not accept criticism, because they denounce the notion that other people possess more know-how than they do. In the worst of cases a narcissistic person even sees advice as an attack on their superiority. Finally, narcissists are very much self-absorbed. They truly admire themselves and they like to show this. In previous times narcissists would have built statues or named roads after themselves. Nowadays the media is the big mirror in which a narcissistic leader can admire himself (Emmons, 1984: 291-294). When taking all these factors into account and linking these to certain requirements we hold for good leadership something interesting comes up. Although narcissism is often seen as a negative personality trait, some elements are also seen as requirements for individuals in leadership positions. Emmons refers to this notion as the narcissistic paradox (ibid.: 298-299).

The four factors of narcissism constructed by Emmons illustrate the most extreme form of narcissism. It is important to note that narcissism takes on different forms. Kets de Vries and Miller (1985) make the distinction between three forms of narcissism. The prominence of the aforementioned narcissism factors differs between these different forms. The first form is called constructive leadership. Kets de Vries and Miller refer to constructive leadership as the most mild form of narcissism. Constructive leaders are very ambitious and they do not hesitate to use some form of manipulation to reach their goal. They are not very fond of criticism but their high self-esteem enables them to deal with it. Constructive leaders are good listeners. They do not mind listening to the ideas of other people as long as they can claim responsibility (Kets de Vries & Miller, 1985: 595, 598).

The second form of narcissistic leadership is the self-deceptive leader. Their main objective is to be liked by all people close to them. Because they cannot handle criticism very well, they collect people around them who look up to them and who think twice before going against anything they say. This grand desire to be liked is the reason that it is not likely that self-deceptive leaders take big risks. Their biggest concern is to do their job well and to be liked by the people around them (ibid.: 595, 597). Therefore, although this sort of narcissism

8

is quite severe from a psychological perspective it does not cause the big effects on the organization that stand central in this thesis.

The final form of narcissism is the reactive leader. This is the most destructive form. They are cold, ruthless and have a strong desire to control the situation. Mainly due to their extreme form of self-confidence they stand above reality and most decisions are therefore made only based on their intuition. Such a leader rarely takes on the advice of other people because he is strongly convinced that he knows everything best (ibid.: 596). If a situation occurs where something goes wrong, he is the last to admit his mistakes and he does not accept the subsequent critique directed at him. Just as the self-deceptive leader he solves such problems by collecting an inner circle of people around him. This high tendency to avoid punishment can have severe consequences. Reactive leaders are tempted to cover up mistakes in order to keep up the appearance of untouchability (ibid.: 599). When the previously

discussed narcissism factors are compared to the different forms of narcissism it becomes apparent that reactive narcissism is the most extreme form of narcissism. Especially their feeling of superiority that disconnects them from reality can create problems in the long run.

Over the last decade narcissism research redirected its focus from the negative side of narcissism, and put much more emphasis on the positive side. The downsides of narcissism closely align with some aforementioned elements. Downsides are the fact that narcissist are hypersensitive and amoral. Their hypersensitivity can create anger bursts which can severely damage the internal environment of an organization, whereas amorality can lead to fraudulent acts (Rosenthal & Pittinsky, 2006: 620-621). The hypersensitivity aligns with the

aforementioned fact that narcissistic leader cannot handle critique whereas the amoral downside of narcissism can be linked to the reactive narcissists feeling of superiority.

Contra to the downsides of narcissism are the upsides. Narcissistic leaders are often praised for their great vision. This enables them to see opportunities before they even arise (Maccoby, 2000: 72). These leaders also have the capability to motivate their employees and therefore create a score of followers (ibid.:72). Rosenthal and Pittinsky even go as far as to suggest that without these narcissistic leaders we would live in a world where bold innovation and social change would not be as present as it is today (Rosenthal & Pittinsky, 2006: 630). Coinciding with this statement is the concept of positive self-esteem, or hubris. This concept is a clear element of the superiority factor, and it creates the same paradox as narcissism as a whole. On the one hand hubris is seen as a requirement to seize opportunities and motivate the organization, where on the other hand the same CEOs engage in uninformed or excessive risk-taking, grandiose initiatives, and acts of intimidation (Hiller & Hambrick, 2005: 297).

9

2.3 The effect of narcissism

Whether or not narcissism is a negative leadership trait is best tested by connected narcissism to organizational performance. Chatterjee and Hambrick (2007) measured the relative

narcissism scores of CEOS and connected these scores to the performance of firms. Performance was measured by looking at the difference between firm performance and industry average. They found that the higher a CEO scores on narcissism, the more they favor bold actions that attract attention. These actions either result in big wins or big losses

(Chatterjee & Hambrick, 2007: 376). Their study encompasses the positive and the negative sides of narcissistic leadership. It is the extreme positive effects of narcissistic leadership that make them so attractive for companies. However, it is the extreme negative side of

narcissistic leadership that makes it such an academic hot topic.

Rijsenbilt et al. (2011) did not focus on the extremes of performance and therefore give some other insights in the effect of different levels of narcissism. They refer to narcissistic CEOs as Sun kings which gives away the negative baggage of the concept. Rijsenbilt et al. look at the ROA and the Tobin’s Q of a firm to determine the performance of an organization. They find that a certain amount of narcissism is necessary if you want to reach the top in an organization (Rijsenbilt et al., 2011: 68). Their results also aligns with the aforementioned notion that a high level of narcissism can be destructive for an organization (ibid.: 69).

In the elaboration of this research something very interesting came up. The relationship between CEO narcissism and performance takes on the shape of a concave parabola. As Rijsenbilt explains: ‘The conclusion can be drawn that moderate levels of CEO

narcissism show higher financial performances while very low or very high levels of CEO narcissism show lower financial performance. An intricate connection between leadership and narcissism exists: having too little narcissism can destruct the CEO effectiveness; having too much may destroy the CEO ability as well.’ (Rijsenbilt, 2011: 122). These findings align

with the theory that narcissism is a requirement for effective leadership. However, too much narcissism creates the negative effects that are connected to the reactive leadership style. Rijsenbilt concludes by saying that the objective of every organization should be to find this equilibrium of CEO narcissism and keep it under constant monitoring in order to reap the positive effects (ibid.).

10

2.4 Role of culture

Previous research concerning narcissism was primarily conducted in the U.S. and it therefore failed to take cross-national differences into account. By introducing national culture into the debate we try to fill this gap. National culture is the average pattern of beliefs and values, around which individuals in a country vary (Hofstede, 1983: 78). It is the premises that sound management practices are not universal that creates a role for national culture in management research. National culture influence the perception and effectiveness of different leadership traits (Hofstede, 1983: 75).

2.4.1 Culturally endorsed leadership

The GLOBE project, a large study into the relationship between leadership and national culture already made the distinction between implicit leadership theories, which are build on the premises of universal leadership traits, and culturally endorsed implicit leadership theories. According to the implicit leadership theory, individuals hold a set of beliefs about the kinds of attributes, personality characteristics, skills, and behaviors that contribute to or impede outstanding leadership. These belief systems, in the broader social cognitive

literature, are assumed to affect the way an individual accepts and responds to leadership (Javidan et al., 2006: 72). In these theories cross-national differences do not play a role and it could be said that previous narcissism theories fall into this category.

The GLOBE project does not renounce the concept of implicit leadership theories. However, they believe that there are also leadership theories that are culturally endorsed. The main hypothesis of these theories is that that members of cultures share a common frame of reference regarding effective leadership (House et al., 2004: 711). Several studies already used the data collected by the GLOBE project to highlight the importance of culture in the leadership debate. Koopman et al. (1999) show the difference in leadership perceptions between European countries. There especially is a clear distinction between North-West Europe and South-East Europe. Leadership in the South-Eastern European countries is focused more on power distance and masculinity than is the case in the North-Western countries (Koopman et al., 1999: 515).

Brodbeck et al. (2000) compared the leadership prototypes of several European countries. They found that, besides the fact that leadership perceptions are culturally

endorsed, the effectiveness of different leadership traits is culturally dependent. When certain leadership traits align with the overall values presented in a culture this enhances the

11

effectiveness of such leadership traits (Brodbeck et al., 2000: 12). In this thesis we argue that these indirect and direct cultural effects of leadership traits are also applicable to the concept of narcissism.

Other studies focused on specific leadership traits. Den Hartog et al. (1999) found evidence of the universality of charismatic leadership (Den Hartog et al., 1999: 249). This is of specific interest for this study because narcissistic and charismatic leadership are often seen as two sides of the same coin (Maccoby, 2000: 72). In the GLOBE research charismatic leadership is defined as a leadership dimension that reflects the ability to inspire, motivate, and to expect high performance outcomes of others on the basis of firmly held beliefs

(Javidan et al., 2006: 73). This conceptualization of charismatic leadership closely aligns with the upsides of narcissistic leadership mentioned by Rosenthal and Pittinsky (2006: 630). However, such research into the universality of charismatic research does not take the dark side of narcissism into account (Higgs, 2009: 175). This research contributes by relating the four factors of narcissism by Emmons (1984) to the different perspectives of national culture.

2.4.2 The concept of culture

When talking about national culture in the field of business one name always comes up, Geert Hofstede (1983; 2011). Although his theories still take on a prominent place within the field of national culture research, it is not without critique (Javidan et al., 2006: Shenkar, 2001). Because the growing criticism over the last few years we must explain why we made the choice to use Hofstede in this thesis. There are three advantages for the use of Hofstede that are applicable to this research. First, the values of Hofstede are the most widely used in the measurement of national culture which enhances their legitimacy. Second, the work-related nature of Hofstede’s values activated in work organizations matches well with our object of study, culturally endorsed leadership perceptions. Third, and finally, our analysis is carried out at the country level, matching the nature of Hofstede’s value dimensions, which are based on national means (Haxhi & van Ees, 2010, 712).

According to Hofstede (2011), the effect of national culture on organization practices is profound, and can be captured by six dimensions that reflect business and work-related values (Hofstede, 2011: 7-8). The first dimension looks at the degree of individualism versus collectivism in a country. This dimension centers around the question to what extent a

country evolves more around the ‘I’ or the ‘we’ (Hofstede, 1983.: 79). The second dimension is labeled power distance. This dimension looks at the level of inequality in a country and the manner that a country reacts to this inequality (ibid.: 81). The third dimension has to do with

12

uncertainty avoidance. Every culture reacts differently to the fact that the future is unknown. Some cultures try to minimize the uncertainties that lie ahead whereas other cultures accept it and live much more by the day (ibid.: 83). The fourth dimension is concerned with

masculinity. Hofstede makes a distinction between masculine and feminine cultures. More masculine cultures are extrovert and more materialized, whereas feminine cultures focus much more on the quality of life and on social interaction (ibid.: 85). The fifth dimension makes the distinction between pragmatic and normative cultures. This dimension looks at the extent to which societies accept that a lot of what happens cannot be explained (Hofstede, 2011: 21). The final dimension makes the distinction between countries that accept

indulgence and countries that feel that you should restrain gratification and focus much more on keeping up with social norms (ibid.: 16).

Hofstede’s work is used as the starting point for the most elaborate research ever conducted concerning leadership and culture, the GLOBE project. The overarching goal of the GLOBE project is to develop an empirically based theory to describe, understand, and predict the impact of specific cultural variables on leadership and organizational processes and the effectiveness of these processes. More specific, they ask themselves if leadership behavior, attributes, and organizational practices are accepted and effective in only some cultures (House et al, 2001: 492). A selection of the national culture dimensions of Hofstede (2011) and the GLOBE project (2004) are used in this thesis as proxies for narcissism. These dimensions are clustered into three groups. The premises of this approach is that the

aggregate effect of national culture is illustrated by the partial effects of the selected culture value dimensions.

2.4.3 Culture and narcissism

When secondary data is used in a different manner than its original form it is important to clearly explain the grounds on which these choices are made. The GLOBE project

dimensions of national culture and the pragmatic and indulgence dimensions of Hofstede are used as proxies for national culture in this research. The choice to only include the final two Hofstede dimensions is based on the theoretical foundation of the GLOBE dimensions. Several of these dimensions measure the same elements of national culture as the Hofstede dimensions (House et al., 1999: 16). For instance, the masculinity dimension which is missing in the GLOBE project is replaced by assertiveness and gender egalitarianism (ibid.). The indulgence and pragmatic dimensions are the only two that are not present in the

13 dimensions is presented in Appendix B.

We clustered the dimensions into three groups which enable us to connect the dimensions more easily to the national culture and narcissism theories. The first group is labeled masculinity. The dimensions present in this group determine to what extent a country is more feminine or masculine. Previous research already established the importance of this distinction. In Europe, there is a clear distinction between the northern countries that are more feminine and the southern countries that are much more masculine (Koopman et al., 1999: 515). Power distance, assertiveness, humane orientation, gender egalitarianism, and performance orientation are included in this group.

The second group focuses on the degree to which a country emphasizes the individual or the collective. Some cultures put much more emphasis on the collective in the way they set goals, remunerate, and organize their organization. The feeling of superiority and

self-absorbance of a narcissistic CEO stand opposite to such collective feelings (Emmons, 1984: 293). Next to the collectivism dimension the indulgence dimension from the Hofstede study (2011) is included.

Finally, the relative extent to which a society accepts uncertainty or tries to avoid it is included as a group. In short, narcissistic CEOs do not act on facts and hard data. They prefer to use their superior overview of the situation and make decisions based on their gut feeling (Emmons, 1984: 292). This aligns with countries that score very low on uncertainty

avoidance. Next to the uncertainty avoidance dimension, future orientation and the pragmatic dimension are included. The theoretical relationship between narcissism and each individual dimension is discussed below.

Hofstede (1983) used one dimension to determine the degree of masculinity. The GLOBE project (1999) split this dimension into two separate dimensions, although they emphasize that other dimensions in their research are also placed on the

masculinity/femininity continuum (House et al., 2004: 513, 569, 244 ; Hofstede, 1983: 566). The first dimension is power distance. Societies differ in the extent to which they accept and endorse authority, power differences, and status privileges (House et al., 2004: 513). Power distance specifically looks at the degree of inequality in power between less powerful individual and more powerful ones within the same social system (Mulder, 1977: 90). Furthermore, in high power distance countries, it is expected that certain positions higher up on the social latter are unattainable for people who are at the bottom of society. Other

countries with lower power distance believe much more that everyone should have a fair shot at promoting to the next professional or social level. Low power distance countries base their

14

distaste of large power differences on the notion that power corrupts and eventually result in the abuse of power (House et al., 2004: 518).

According to House et al. (2004) the variation between countries on this score is rooted in the opposition between catholic and protestant tradition. The catholic church has always been coordinated through a highly formal and hierarchical system in which religion and politics were very much intertwined. As a consequence catholic culture has always been associated to high power distance. The protestant tradition on the other hand focuses much more on the individual and their personal relationship with the Christian religion. Protestant societies are therefore less accustomed to power distance. Centuries later, this distinction is still visible between European countries which are dominantly protestant or catholic (House et al., 2004: 520).

Power distance closely aligns with the dimensions of narcissism. Narcissistic leaders feel that they are entitled to more than others and they therefore fit better within a system in which high power distance is accepted. In the same light, the feeling of superiority of narcissistic leaders aligns with cultures where power is unevenly distributed. As a consequence, these claims of superiority are accepted by the people below them in the organization. Such a feeling of superiority make a CEO very hostile towards any feedback or advice. They see this as a critique directed at them personally. Within high power distance societies, it is less common to ask questions or give advice to someone higher up in the organization for the very reason that this probably is seen as critique (House et al., 2004: 559).

The second dimension related to the masculinity/femininity continuum is

assertiveness. The assertive dimension shows to what extent a culture stimulates assertive, aggressive, and tough behavior in social relations (ibid.: 395). Interesting in this

conceptualization of the assertive dimension is the integration of aggressive behavior. Aggressive behavior has a positive connotation in masculine countries whereas it is seen as negative in more feminine countries (ibid.: 396). Furthermore, highly assertive cultures value dominant tough behavior, and value the strong in an organization. They also emphasize performance oriented behavior and therefore appreciate leaders that take initiative and set demanding and challenging goals for themselves and others (House et al., 2004: 405).

The third dimension on this continuum looks at the degree to which a culture is humanely oriented. This dimension is defined as the degree to which a society encourages and rewards individuals for being fair, altruistic, friendly, generous, caring, and kind to others. (House et al., 2004: 569). Although love, friendship and care are universally accepted

15

values the prominence they get differs between cultures (ibid.: 592). Cultures that have a low humane orientation are highly masculine in the eyes of Hofstede (1983: 566).

This dimension is closely related to the previously discussed narcissism dimensions. According to House et al., self-enhancement and self-gratification can be interpreted as being less humane oriented (House et al., 2004: 566). These elements align with the entitlement and self-enhancement dimensions of Emmons (1984). It can therefore be said that narcissists are better suited at places which are less humanely oriented.

The fourth dimension on the masculinity/femininity continuum looks at gender

egalitarianism. At the core of this dimensions lies the question whether a society beliefs that a member’s gender should determine the roles that they play in their homes, business

organizations and communities. Gender egalitarian cultures do not allocate the roles in society according to the sexes of its members (House, 2004: 347). It is important to know whether a country holds more feminine or masculine traits as the standard. Countries

dominated by male leaders focus more on hierarchical and competitive interpersonal behavior where female leaders put more emphasis on collaborative and consensus building behavior (House et al., 2004: 384). Cultures that are less egalitarian have the tendency to place more men at top positions and therefore their leadership practices depend much more on the aforementioned masculine leadership behavior. As the masculinity of a culture positively relates to narcissistic behavior we propose that societies that are less gender egalitarian have a positive effect on the presence and effectiveness of narcissistic leadership.

The final dimension in the first group looks into the performance orientation of a culture. This reflects the extent to which a community encourages and rewards innovation, high standards, and performance improvement. Cultural differences influence the way people define success and set goals (House et al., 2004: 239, 243). In this dimension the distinction between catholic and protestant countries also plays a role. Max Weber was amongst the first to point out the difference between catholic countries, where the emphasis was on doing good, and protestant countries where working hard was seen as a godly calling (Weber, 1904). Weber was one of the earliest academics who emphasized that performance

orientation was not a universal concept. However, because of the diminishing role of religion in many countries the distinction made by Weber was not supported by the GLOBE project (House et al., 2004: 258).

Despite the insignificant role of religion within this dimension the distinction between high en low performance oriented countries is still present. High performance oriented

16

once, which means that it should be used efficiently. Communication in such cultures is therefore much more direct. Furthermore, the focus in such cultures is oriented towards winners and achievers. Social relations are of less importance and the task at hand is valued most (House et al., 2004: 244-245).

When connecting the concept of performance orientation to narcissism it becomes clear that narcissistic behavior is much more suited for performance oriented cultures. The focus is on the CEO who sees himself as the winner and the realization that time is a valuable asset enables the CEO to act more directly without consulting other organizational members. House et al. also emphasize the need for visionary, inspirational, decisive and performance oriented leadership in such cultures (2004: 276). Traits that are connected to narcissistic CEOs (Maccoby, 2000: 72).

The second group of dimensions centers around the collectivism/individualism continuum. In the GLOBE project they assigned this role to one dimension and in this thesis another dimension is added. The essence of collectivism is that individuals are interdependent and have duties and obligations to other group members (House et al., 2004: 438). In

individualistic societies however, ties between individuals are loose and everyone is expected to look out only for themselves and the people close to them (ibid.: 440). As a consequence group goals take precedence over individual goals in collective societies and vice versa in individual societies (ibid.: 454).

The degree to which a culture is collectively or individually based affects the

perception on leadership to a great extent and influences the way that narcissistic leaders are perceived. In individually based societies compensation and promotion are based on an equity model in which an individual is rewarded in direct relationship to their contribution (ibid.: 459). This closely aligns with the narcissistic CEO’s view of entitlement. Collective remuneration would go against every instinct of such a CEO. Furthermore, accountability for failure and success rests with individuals (ibid.). This means that CEOs stand in front of the group and get all the praise when times are good and the blame when times are tough. This aligns with the fact that narcissistic CEOs enjoy being the leader, and they enjoy getting all the praise for the success of their organizations.

The other dimension which is connected to the collectivism/individualism continuum looks at the extent that societies show restraint or fully indulge themselves when something good happens. Societies that are on the indulgence side of the spectrum allow the

gratification of basic and natural human desires related to enjoying life and having fun. Societies on the restraint side of the spectrum however feel a need to control these impulsive

17

feelings of gratification in the name of strict social norms (Hofstede, 2011: 15).

The conceptualization of this dimension is related to two of the narcissism factors. First of all, narcissistic leaders feel that they are entitled to more than others. This means that they want to be paid more and they want to get services that show the rest of the world their importance. This is in line with the indulgence side of the spectrum that states that life should be enjoyed and it goes against the restraint side because such societies do not approve such a mentality. The second dimension of narcissism that comes up is again superiority. Countries that emphasize the need to restrain feelings of gratification value the social norms that are seen as binding for every citizen. A narcissistic leader’s feeling of superiority creates an individual mindset that stands opposite to the notion of social norms.

The final group of dimensions centers around the role of knowledge in a society. Some cultures are better equipped to deal with the uncertainties we are confronted with every day. The degree to which a culture accepts these uncertainties or tries to minimize them determines their position on this continuum. The first dimension within this group looks at the degree of uncertainty avoidance in a culture. Uncertainty is a force that is always present in almost every aspect of life and the way a society deals with these uncertainties determines their score on this dimension (House et al., 2004: 602). Uncertainty can be canalized by the issuance of laws, procedures and consistency of actions and in some cultures they are tempted to use such precautions (ibid.: 603).

Uncertain situation create different reaction within an organization and have different effects. People who are less open to uncertainty are more dependent on other members of their organization for feedback in order to understand the consequences of their actions (House et al., 2004: 604). For an organization as a whole it has an effect on the way they plan future actions and more importantly, on the extent to which they can be innovative.

Companies that are innovative need to deal with a certain degree of uncertainty (ibid.: 606). Narcissistic leaders are everything but uncertainty avoiders. The positive side of narcissism, mentioned by Maccoby (2000) and Rosenthal and Pittinsky (2006), is their vision to act on opportunities that have not yet presented themselves. As mentioned above, the absence of uncertainty avoidance also create an environment within an organization that opposes the use of feedback within an organization. This aligns with the feeling of superiority of narcissistic leaders who see feedback as a direct attack directed at his capabilities.

The second dimension related to the uncertainty continuum looks at the relative future orientation of a culture. This dimension centers around the subjective experience of time that differs between cultures. It is the degree to which society encourages and rewards

future-18

oriented behavior such as planning and delaying of gratification (House et al., 2004: 282). In general, we speak of three temporal frames: past, present and future. The degree to which a culture is focused one has implications for the way they act in the present (ibid.: 285). For instance, cultures with a high focus on the present are much more spontaneous and have a higher tendency to seek hedonic pleasures. At the same time they are less willing to plan ahead and do not appreciate people reminding them of their responsibilities (ibid.). In

contrast, cultures with a strong orientation on the future integrate future goals and aspirations in their present actions. As a consequence they are much more equipped to maintain self-control over short-term desires and keep their focus on the long-term (ibid.).

This dimension has an influence on the perception of narcissism in a culture. Here the double edged sword of narcissism becomes apparent. On the one hand narcissistic leaders are praised for their visionary view which enables them to see opportunities before they even arise. This closely aligns with the mentality present in highly future orientated countries whom are comfortable basing their actions in the present on outcomes which become apparent in the future. On the other hand entitlement is one of the main pillars of Emmons’ narcissism (1984). This pillar stands opposite to the delaying of gratification. This creates a ambiguous role for this dimension in the narcissism debate.

The pragmatic dimension looks at the way a country copes with the uncertainty of not being able to answer all questions in life and it therefore is positioned on the uncertainty avoidance continuum. On the one side you have normative cultures. These societies have a strong desire to explain everything. They are always searching for a higher truth and they want to minimize the unanswered questions in their environment. As a consequence they have a great need for personal stability. On the other side of the spectrum are pragmatic societies. These societies do not want to know everything and they accept this because they understand that many things cannot be explained. The challenge is not to know the truth but to live a virtuous life (Hofstede, 2011: 21).

There is a clear comparison between narcissism theory and the pragmatic dimension. Especially Emmons’ superiority dimension shows the position of a narcissistic leader within this dimension. Narcissists do not have an inner urge to know everything. Because of their feeling of superiority they do not need all the information to make decisions. They stand above facts and reality and fully trust their gut feeling. Their trust in themselves also makes them less careful. In normative societies stability is the key and this goes against everything a narcissistic leader is.

19

2.5 The hypotheses

On the basis of the connection between the narcissistic theory and the ten national culture dimensions six hypotheses are constructed. We argue that every group of dimensions influences the narcissism debate directly and indirectly. The direct effect looks at the influence of national culture on the presence of narcissistic behavior at the CEO level in a country. The premises that national culture influences the presence of narcissistic behavior is built upon previous research concerning the culturally endorsed nature of leadership traits in general. We test whether the findings of Koopman et al. (2000) are also applicable to

narcissistic behavior.

The first group of dimensions centers around the masculinity/femininity continuum. The narcissism factors of Emmons (1984) and the dimensions that are clustered in this group have a clear relationship with each other. The feeling of superiority, self-enhancement, and entitlement align with the notion of high power distance, an assertive leadership style, and performance orientation and it misaligns with gender egalitarianism and humane orientation. The mentality of a narcissistic CEO is focused on performance, his actions are assertive, and he feels best in a highly hierarchical system (Emmons, 1984: 291-294). The degree to which such traits are accepted is country dependent. Previous research found that such masculine leadership traits are preferred in the south of Europe, whereas northern countries are much more focused on feminine leadership traits (Koopman et al., 1999: 515) Feminine cultures do not share the same mentality with narcissistic CEOs. For instance, the degree to which a society is humanely orientated will decrease the presence of narcissistic behavior, because narcissistic leaders are to self-absorbed to be oriented on other human beings (Emmons, 1984: 294). In sum, we argue that cultures that score high on the masculine dimensions and low on the feminine dimensions will create the ideal environment for narcissistic behavior:

h1a: The degree to which a society is masculine positively influences the presence of narcissistic behavior at the CEO level.

The second pair of dimensions illustrates the distinction between collective and individual cultures. One of the main pillars in the narcissism literature is the fact that narcissistic individuals feel superior to the people around them (Emmons, 1984: 292). Therefore, the behavior of narcissists is very much centered around the individual. They do not believe in teamwork because the rest of a team will only slow them down. Furthermore, they also believe they are entitled to more. Because their actions are superior to those of

20

others they want to be rewarded accordingly. They therefore expect high salaries and high bonuses. This stands opposite to the mentality present in collectivistic societies who believe that resources should be distributed equally (House et al., 2004: 438).

The indulgence dimension is the other dimension that is included in this group. Opposite to the collectivism dimension this dimension aligns more with the feeling of entitlement of a narcissistic CEO. In such societies the gratification and celebration of landmarks is accepted (Hofstede, 2011: 15). This creates an environment that promotes narcissistic behavior that is centered around the entitlement of the CEO.

On the basis of the alignment between individual cultures and narcissistic behavior we argue that narcissist behavior is positively influenced by the degree to which a society is centered around the individual instead of the collective. They believe they are superior, they are entitled to more, and they love to show these elements to the rest of the world (Emmons, 1984: 291-294). Such behavior is more accepted in societies that are centered around the individual. We therefore hypothesize that cultures that score low on collectivism and high on indulgence positively influence the presence of narcissistic behavior at the CEO level:

h1b: The degree to which a society is focused around the individual positively influences the presence of narcissistic behavior at the CEO level.

The final group looks at the degree to which societies try to avoid uncertainty. Pragmatic societies are very accustomed to dealing with uncertainty. Countries that score high on this dimension do not panic in situations where they are unable to attain all the facts (Hofstede, 2011: 21). This aligns with the notion that narcissistic CEOs are praised for their vision and for their capability to act on events that have not yet presented themselves (Rosenthal & Pittinsky, 2006: 630). The pragmatic dimension, and the narcissism factors, stand opposite to the mentality present in uncertainty avoidant societies. This dimension stand on the opposite side of the continuum and in such cultures people design their

environment in such a way as to minimize the uncertainty of their surroundings (House et al., 2004: 603). Narcissistic CEOs differentiate themselves in situations in which they are able to use their vision to guide an organization. Such a leadership style is not accepted in countries that score high on uncertainty avoidance.

The future orientation dimension in this group had an ambiguous relationship with the concept of narcissism and it therefore was difficult to clearly hypothesize a directional

relationship between this dimension and narcissistic behavior. However, based on the other two dimensions in this group we argue that the extent to which cultures accept the lack of

21

knowledge in their environment positively influences the presence of narcissistic behavior at the CEO level:

h1c: The extent to which cultures accept uncertainty positively influences the presence of narcissistic behavior at the CEO level.

Next to the direct effect of national culture on the presence of narcissism at the CEO level we also hypothesize an indirect effect. Brodbeck et al. (2000) already established that the effectiveness of certain leadership traits is culturally endorsed. This means that national culture influences the relationship between narcissism and firm performance. We argue that national culture has the same moderating effect on the relationship between CEO narcissism and firm performance. The main premises for the upcoming hypotheses is that narcissistic CEOs are better able to perform in cultures that align with their own mentality. In other words, the actions of narcissistic CEOs are more effective when they are accepted by the people around them.

The direct effect of the masculine dimensions on the presence of narcissism is discussed in the previous section. We propose that the relative alignment between the narcissism factors and the national culture dimensions also affects the performance of these narcissistic CEOs. One of the positive sides of narcissism is that such leaders are able to motivate their employees (Maccoby, 2000: 72). However, the leadership style, which is described as aggressive and tough, does not align with the leadership perception present in feminine societies (House et al., 2004: 396). We therefore expect the effectiveness of the motivational capabilities of narcissistic CEOs to be culturally dependent. Furthermore, their drive for results aligns with the mentality in highly assertive cultures. A leadership style that is primarily focused on the outcome of an organization could have the opposite effect in feminine societies. Such societies are more focused on quality of life and social interaction (Hofstede, 1983.: 85). On these grounds we argue that the degree to which a society is masculine positively moderates the relationship between the narcissism score of a CEOs and the performance of their organization:

h2a: The degree to which a society is masculine positively moderates the influence of a narcissistic CEO on firm performance.

In contrast to the masculine dimension, the collectivistic misalign with the narcissism factors. First of all, collectivism goes against the individualistic mindset of the narcissistic CEO. A narcissistic CEO prefers an organization in which he is solely in charge. Such a

22

hierarchical design of the organization in which one man is responsible for the decisions is not accepted in countries that score high on collectivism. Such countries emphasize the importance of internalizing all stakeholders in the decision making process. This takes away a lot of the maneuverability of narcissistic CEOs what diminishes their effectiveness to act upon opportunities.

The second point centers around the motivation of narcissistic CEOs to perform well. The praise for good results should be directed at them and they therefore do not like to share any reward for such results with other executives. Such a mentality is based on the equity model. A reward system that pays by performance which is most commonly used in societies centered around the individual (House et al., 2004: 459). The indulgence dimension, which promotes the gratification of basic and natural human desires coincides with the mindset of rewarding performance instead of delaying gratification in line with social norms (Hofstede, 2011: 15).Collectivistic remuneration on the other hand decreases the motivation of a narcissistic CEO to maximize performance, because the reward does not directly reflect the importance of the CEO (Emmons, 1984: 293). Because of the misalignment between the narcissistic behavior of CEOs and the mindset present in collectivistic societies we propose a negative moderating effect of the collectivism dimension and a positive effect of the

indulgence dimension:

h2b: The degree to which a society is focused around the individual positively moderates the influence of a narcissistic CEO on firm performance.

The final hypothesis centers around the indirect effect of the uncertainty avoidance dimensions on the performance of narcissistic CEOs. The vision of narcissistic CEOs to act upon events that have not yet occurred is seen as one of the big upsides of narcissistic leadership (Rosenthal & Pittinsky, 2006: 630). In cultures that try to avoid uncertainty it is much harder for a narcissistic CEO to leverage their vision. Such behavior requires an environment that accepts the fact that the actions of a narcissistic CEO are not based on hard facts. In uncertainty avoidant societies however, uncertainty is canalized by the issuance of laws, procedures and consistency of actions (House et al., 2004: 603). Narcissistic CEOs see these precautions as a burden which negatively influences their capability to perform.

Furthermore, we argue that the role of knowledge within a culture has an effect on the way a narcissistic CEO influences the performance of the firm. Because of their feeling of superiority narcissistic CEOs do not base their actions on factual knowledge. They see it as a resource other people need, but what they can do without because of their superior capability

23

to act upon opportunities in the market (Emmons, 1984: 292). This aligns with the mentality present in pragmatic societies. Pragmatic societies are comfortable with the inexistence of total knowledge (Hofstede, 2011: 21). Narcissistic behavior, which is rarely based on facts and much more often on the intuition of the CEO, is therefore more appropriate in such cultures. All in all, we argue that the degree to which a culture accepts uncertainty positively moderates the performance of narcissistic CEOs:

Hh2c: The extent to which cultures accept uncertainty positively moderates the influence of a narcissistic CEO on firm performance.

2.6 Causal model

Two roles for national culture are presented in the model. The first role hypothesizes a direct effect of the different national culture dimensions on the existence of narcissism at the CEO level. The second role of national culture is as a indirect moderator of the relationship between narcissistic leadership and firm performance. A moderating variable affects the relationship between an independent variable and a dependent variable (Saunders et al., 2012: 424). In the case of this thesis the independent variable is CEO narcissism and the dependent variable is firm performance. All these variables and the associated hypotheses are brought together in model 1 below.

24 3. Data and Method

3.1 Sample

The main goal of this thesis is to find out what role national culture plays in the narcissism debate. A cross national sample therefore is essential. We chose a sample which consists of CEOs of traded European firms. The sample only consists of traded firms because some of the narcissism indicators are only applicable in such cases. Furthermore, these firms are obligated by law to be more transparent, which makes it much easier to collect the data. European firms were chosen for several reasons. First of all, the transparency of European firms made them ideal for this research. Secondly, previous research concerning narcissistic CEOs focused on North American firms and it is of added value to also use western countries in this study.

It is possible to bundle national cultures into several clusters. Previous research already showed that leadership prototypes align with these cultural clusters (Brodbeck et al., 2000: 12). Of all the cultural clusters, the support for the European clusters was the strongest (Ronen & Shenkar, 1985: 542). According to Ronen and Shenkar western Europe consist of four clusters; the Anglo, Latin, Nordic and Germanic Clusters (House et al., 2004: 191).An important point they make is that different cluster sets may be equally valid depending on the viewpoint that is taken (ibid.). To prevent any ambiguity in the choice of clusters they are only used to spread the sample in a logical matter.

During the analysis the emphasis is on the national level. Within each cluster two countries are picked and in every country ten CEOs of different traded companies are

included. This brings the total amount of research units to 80. The Forbes Global 2000, a list of the 2000 largest companies in the world is used to select the companies (Forbes website). Because this list already consists of the largest firms present in a country we chose the CEOs within each country randomly. Looking at other aspects, as CEO gender or industry is not possible for practical reasons. Because several countries only have a small amount of firms present in the Forbes Global 2000 there is no room to use other requirements in the selection process. The total list firms and their CEOs is added in the Appendix (A).

3.2 CEO narcissism

The most widely accepted way of researching narcissism is the NPI by Raskin and Hall (1981). This survey-based index divides narcissism into four categories; authority, self-admiration, superiority and entitlement (Emmons, 1987: 12). The problem with the NPI is

25

that it is very difficult to apply such a method in research concerning CEOs. It is close to impossible to find CEOs who are willing to cooperate, let alone on a large scale (Rijsenbilt et al., 2011: 64). This is why another method is used in this thesis that deals with such

problems. By focusing on unobtrusive indicators of a CEO’s narcissistic tendencies it becomes possible to eliminate problems of reactivity. Unobtrusive indicators look at the psychical traces, non-participant observation, documentary sources, and the written and spoken words of subjects as ways to learn about their preferences, perceptions, and personalities (Chatterjee & Hambrick, 2005: 362). This enables research that measures narcissism at the CEO level without them having to participate directly in the research. Another benefit of this method is that all indicators can be found in open sources.

Chatterjee and Hambrick used five indicators, that closely align with the four categories used by Emmons, to measure narcissism at the CEO level (ibid.: 365). The first indicator looks at the prominence of the CEO’s photograph in the annual report. The way a CEO is portrayed in the annual report strongly reflects their position within the company. Although the visual presence of the CEO in the annual report is standard for most companies there is no universal rule that says what size it should be and who it should contain. The CEO’s picture also greatly differs in size. The more prominent the CEO is present, the higher he scores on this indicator (ibid.: 363). A four point scale is used to rate this indicator: four points if the photo of the CEO is of him or her alone and occupies more than half a page, three points if the picture is of the CEO alone and occupies less than half a page, two points if the CEO is photographed with one or more of his fellow executives, and one point if there was no picture of the CEO present.

The second indicator looks at the prominence of the CEO in the media. The press is a modern outlet for narcissistic CEOs to showcase their vanity. A CEO that scores high on narcissism wants people to think of him when the company is mentioned in the press. He therefore actively presents himself as the spokesman of the company. To measure this prominence the lexis nexus media database is used. By searching articles concerning the company between two time units, that either mention the CEO or did not, a CEO prominence score was created (Rijsenbilt, 2011: 60-61). This is not the original prominence indicator used by Chatterjee and Hambrick. They looked at the prominence of the CEO’s name in press statements released by the company itself (2007: 363-364). This however brings up certain difficulties during the operationalization. In contrast to the Chatterjee and Hambrick research this thesis has a multi-lingual sample. The English press releases of the ‘non-native’ speaker companies were only of a technical nature where the English and Irish sample was much

26

more personal. To level the playing field the variable was replaced for another variable that tries to uncover the same phenomena, CEO prominence. This indicator was taken from the Rijsenbilt (2011) study.

The third indicator measures the percentage of first-person singular pronounce relative to all first-person plural pronounce. This shows the degree to which a CEO sees himself as the most important individual within an organization. A CEO that scores high on this indicator scores high on self-absorption (Chatterjee & Hambrick, 2007: 363-364). The more often a CEO speaks of I, me, mine, my or myself instead of we, us, our, ours, ourselves, the higher he scores on this indicator and consequently on narcissism (ibid.). The score is constructed by taking the amount of first person singular pronounce and dividing them by all the singular pronounce used. This is measured by analyzing the foreword of the annual reports of the organization in the year 2011.

The final two indicators are related to the relative pay of a CEO. A narcissistic CEO believes he is more valuable than any other employee and therefore he deserves a higher salary. This closely aligns with the entitlement dimension of narcissism. The fourth indicator looks at the relative cash pay of the CEO compared to the second-highest-paid executive in the company. Cash pay refers to base salary as well as bonuses paid out in cash. The bigger the difference between a CEO and the second highest paid executive the higher a CEO scores on this variable. The fifth indicator includes all non-cash paid rewards of the CEO relative to the second-highest paid executive. Non-cash pay refers to deferred income, stock grants, and stock options (ibid.: 364).

Chatterjee and Hambrick tested for the coherence amongst the indicators. They did so to find out if all the indicators contributed to the measurement of narcissism. They calculated the Cronbach’s alpha of all the variables (Chatterjee & Hambrick, 2007: 366). With an alpha score of 0.75 it was above the level acceptable for forming a new index (Nunnally, 1978). In this thesis the Cronbach was just below the accepted level of 0.5. The relative low alpha score of my sample can be assigned to the small sample (n=80). It is commonly suggested that a minimum sample size is needed of around 300 to calculate the alpha (Yurdugul, 2008: 7), and our sample therefore is not large enough. This is why we rely on the high alpha score of the

original study by Chatterjee and Hambrick (2007).

The five indicators are used to generate a narcissism score. The score is created in two steps. First, all the indicators are standardized separately. Afterwards, the mean of all five indicators is calculated which gives us an individual score for every CEO in the sample (Chatterjee & Hambrick, 2007: 366). This means that CEOs that scored one standard