1
Leader narcissism: Bright, dark, or both at the same
time? The role of charismatic leadership, abusive
leadership, and organizational climate
Maud Kruijer 10643125
22-06-2018 Final Draft
MSc Business Administration Track: Leadership & Management First Supervisor: A. De Hoogh Second Supervisor: D. Den Hartog
2 STATEMENT OR ORIGINALITY
This document is written by Student Maud Kruijer who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3 ABSTRACT
This study examined whether leader narcissism can have a bright side and a dark side simultaneously and if this dark side can be diminished through a certain organizational climate. We explored the bright side via charismatic leadership and its relationship with follower being. We explored the dark side via abusive leadership and its relationship with follower well-being. Specifically, we tested whether there is an indirect positive relationship between leader narcissism and follower well-being via charismatic leadership and if there is an indirect negative relationship between leader narcissism and follower well-being via abusive leadership. We also examined if a law and code climate moderates the relationship between leader narcissism and abusive leadership. We collected dyadic data from participants working in Dutch organizations (N = 102). Results of regression analysis suggested that there was an indirect relationship via charismatic leadership, but not via abusive leadership. However, we found that abusive leadership is negatively related to follower well-being. Further, no evidence was found for the moderating effect of law and code climate. Altogether, we did find evidence for the bright side of narcissism, but not for a possible dark side.
4 TABLE OF CONTENTS LIST OF TABLES ... 6 LIST OF FIGURES ... 7 1. INTRODUCTION ... 8 2. LITERATURE REVIEW ... 12 2.1 Narcissism ... 12
2.4 Narcissism and Leadership ... 13
2.3 Leader Narcissism and Charismatic Leadership ... 16
2.4 Charismatic Leadership and Follower Well-Being ... 18
2.5 Indirect Relationship via Charismatic Leadership ... 19
2.6 Leader Narcissism and Abusive Leadership ... 19
2.7 Abusive Leadership and Follower Well-Being ... 22
2.8 Indirect Relationship via Abusive Leadership ... 23
2.9 Moderating Role of Organizational Climate ... 23
2.10 Indirect conditional relationship ... 26
2.11 The whole model ... 27
3. METHOD ... 28
3.1 Procedure and sample ... 28
3.2 Measures ... 29
5
4.1 Data preparation ... 31
4.2 Descriptive statistics and correlations ... 31
4.3 Hypothesis testing ... 34
5. DISCUSSION ... 43
5.1 Limitations and Future Research ... 46
5.2 Practical implications ... 50
5.3 Conclusion ... 51
6. REFERENCES ... 52
6 LIST OF TABLES
Table 1. Means, standard deviations, and correlations………....33 Table 2. Regression results of the indirect relationship via charismatic leadership……….35 Table 3. Regression results of the indirect relationship via abusive leadership………..37 Table 4. Regression results of the moderation effect of law and code climate on
the relationship between leader narcissism and abusive leadership………39 Table 5. Regression results of the indirect conditional effect of law and code
climate on the relationship between leader narcissism and follower
well-being via abusive leadership………...40 Table 6. Regression results of the indirect relationship via charismatic leadership
in the whole model………..41 Table 7. Regression results of the indirect conditional effect of law and code climate
7 LIST OF FIGURES
8 1. INTRODUCTION
Recently, narcissism levels are rising (Twenge & Foster, 2010; Twenge, Konrath, Foster, Campbell, & Bushman, 2008). A narcissist is someone who has a grandiose sense of self-importance (Judge, LePine, & Rich, 2006). There are several (world) leaders who show characteristics of a narcissist. Well-known narcissistic leaders are Adolf Hitler, Steve Jobs, and Michael Eisner (Rosenthal & Pittinsky, 2006). Today, Donald Trump is probably the best-known narcissist in the world. His communication profile suggests that he has a strong self-promotional style that emphasizes narcissistic grandiosity (Ahmadian, Azarshahi, & Paulhus, 2017). Twenge (2013) states that narcissism is especially prevalent in the younger adults, described as ‘Generation Me’.
This rise in levels of narcissism is particularly relevant for organizations because narcissists are prone to be in leadership positions. Firstly, because narcissists have a "desire to ganger the power they need to ‘structure an external world’ that supports their grandiose needs and visions” (Rosenthal & Pittinsky, 2002, p. 618). And secondly, because narcissists tend to emerge as leaders (Brunell et al., 2008; Grijalva, Harms, Newman, Gaddis, & Fraley, 2015; Ong, Roberts, Arthur, Woodman, & Akehurst, 2016). Since narcissists seek leadership positions, they can have great influence in organizations, and thus impact several work-related outcomes.
Narcissism is mostly seen as a dark personality trait, and therefore it is associated with detrimental outcomes (Benson & Campbell, 2007; Braun, 2017; Paulhus & Williams, 2002). However, dark personality traits do not always have to be negative (Smith, Hill, Wallace, Recendes, & Judge, 2018). Whether the impact of a narcissistic leader is positive or negative has been an ongoing conversation between scholars. Although in most studies scholars argue that leader narcissism has negative consequences, there is no consensus about this relationship in the
9 literature. Some studies found negative outcomes of narcissism (Nevicka, Ten Velden, De Hoogh, & Van Vianen, 2011), some studies found positive outcomes of narcissism (Galvin, Waldman, & Balthazard, 2010; Harms, Spain, & Hannah, 2011; Volmer, Koch, & Göritz, 2016), and others did not find a relationship at all (Volmer et al., 2016).
Regarding follower well-being, there are inconsistencies as well. Some scholars argue that
a narcissistic leader has a negative effect on well-being (Sankowsky, 1995), while Volmer et al. (2016) did not find a detrimental impact of narcissistic leaders on follower well-being. These inconsistencies in findings indicate that there might be other factors that play a role in this relationship. Outcomes may depend on which leadership style is shown by the narcissistic leader.
For example, several scholars have suggested that narcissistic leaders possess a strong charisma and engage in charismatic leadership (Conger & Kanungo, 1998; King, 2007; Rosenthal & Pittinsky, 2006). However, this is only based on conceptual research. Empirical evidence regarding the link between leader narcissism and charismatic leadership is weak (Ouimet, 2010). In this study, we test this relationship in an empirical setting. In addition, the suggested charismatic leadership that narcissistic leaders can show may be positively related to follower well-being. It is already shown that followers of charismatic leaders do have more positive moods (Bono & Ilies, 2006) and that it positively influences follower satisfaction (Shamir, House, & Arthur, 1993; Shamir & Howell, 1999). Taken together, it might be that narcissistic leaders have a positive influence on the well-being of their followers through engaging in charismatic leadership, suggesting a bright side of leader narcissism.
On the contrary, the typical behaviors that narcissists can show, such as exploitation of others and arrogant behavior, are associated with abusive leadership (Rosenthal & Pittinsky, 2006; Tepper, 2000). Scholars already suggested that narcissism might predispose leaders to abusive behaviors (Hansbrough & Jones, 2014; Tepper, 2000). But also in this research area, the connection
10 between leader narcissism and abusive leadership is underdeveloped (Waldman, Wang, Hannah, Owens, & Balthazard, 2018; Wisse & Sleebos, 2016). In this study, we try to explicate on this. In addition, the suggested abusive behaviors that narcissistic leaders can show, can, in turn, negatively affect follower well-being (Tepper, 2007; Bono, Foldes, Vinson, & Muros, 2007). These things considered, narcissistic leaders may have a negative influence on follower well-being via abusive leadership, suggesting a dark side of narcissism.
Knowing how to diminish this dark side of narcissistic leaders (i.e., the detrimental effect on follower well-being) can be particularly interesting for organizations. There might be situations in which narcissistic leaders show less abusive leadership than in other situations. Smith et al. (2018) argue that “trait activation theory might aid researchers in explaining the situational factors that lead to the exhibition or inhibition of trait-based behaviors”. Thus, trait activation theory can explain how dark personality traits influence workplace outcomes (Smith et al., 2018). The principle of trait activation theory is that “behavioral expression of a trait requires arousal of that trait by trait-relevant situational cues” (Tett & Guterman, 2000, p. 398). In other words, the expression of a trait may depend on situational factors. This might also be the case with narcissistic leaders; it depends on the situation to what extend narcissistic leaders show abusive behavior. In this study, we look at organizational climate as the situational factor that can inhibit the abusive behaviors of narcissistic leaders. A law and code climate can be such a climate. In a strong law and code climate, people are highly expected to behave according to the regulations (Tsai & Huang, 2008). Because of this expectation, narcissistic leaders might be less prone to show their abusive behaviors. Again, this can be particularly interesting for organizations, because they can use this to diminish the detrimental effects of narcissistic leaders by creating a certain organizational climate, which is, in this case, a law and code climate.
11 In this study, we examine if leader narcissism can be bright and dark at the same time. Also, we investigate how the dark side of narcissism can be less harmful to organizations. To be specific, it is investigated if leader narcissism is indirectly related to follower well-being via charismatic leadership (bright side of narcissism) and via abusive leadership (dark side of narcissism). Further, based on trait activation theory, we examine the moderating role of a law and code climate in the relationship between leader narcissism and abusive leadership.
We aim to extend current literature by providing empirical research into narcissism in leadership. According to Grijalva et al. (2015) “part of the reason the narcissism–leadership debate has been difficult to resolve is that, although there is a tremendous amount of interest in the topic and a large body of theoretical work speculating on the link between narcissism and work outcomes, there has been surprisingly less empirical work compared to the many theoretical claims made in this field” (p. 29). So, this study provides additional empirical work. Specifically, we extend current literature by investigating whether leader narcissism can be bright (via charismatic leadership) and dark (via abusive leadership) at the same time. Furthermore, trait activation by organizational climate is new in research into narcissism. Overall, with this study, we contribute to building a solid foundation of knowledge on narcissism in organizations.
Figure 1. The conceptual model
Leader Narcissism Charismatic Leadership Abusive Leadership Follower Well-Being
Law and Code Climate
12 2. LITERATURE REVIEW
2.1 Narcissism
Narcissism originates from the Greek myth of Narcissus, who was a young man that fell in love with his own reflection in the water (Rosenthal & Pittinsky, 2006; Grijalva & Harms, 2014; Braun, 2017). Based on this myth, Ellis (1898) was the first one to describe the term narcissism as a clinical condition whereby people perversely love themselves. In 1931, Freud described narcissism as a personality trait, whereby people have an outwardly unflappable strength, arrogance, and confidence (Rosenthal & Pittinsky, 2006). He later described narcissism as a personality disorder (Braun, 2017).
In literature, narcissism appears as a personality disorder and as a personality trait. The clinical and psychiatric literature describes narcissism as a personality disorder: Narcissistic Personality Disorder (NDP). NPD is associated with a grandiose sense of self-importance, preoccupation with fantasies of unlimited success or power, belief in “special” or unique status (including fixation on associating with high-status people or institutions), requirement for excessive admiration, unreasonable sense and expectations of entitlement, interpersonal exploitativeness, lack of empathy, envy, and arrogant behaviors or attitudes. To be diagnosed with NPD, people need to have five out of nine symptoms (American Psychiatric Association, 2000). Besides, people should not be able to establish normal relationships and perform at work. These things cause that a relatively low percentage of people is diagnosed with NPD.
However, symptoms of narcissism, without crossing the line of NPD, are still prevalent in society (Stinson, Dawson, Goldstein, Chou, & Huang, 2008). Therefore, the personality literature describes narcissism as a personality trait which is normally distributed in the population (Foster & Campbell, 2007). In this study, we look a narcissism from a trait perspective. From this
13 perspective, we can capture the more common occurrences of narcissism (Stinson et al., 2008), like narcissists in organizations.
Multiple scholars suggest that there are two distinct forms of narcissism: vulnerable narcissism and grandiose narcissism (Dickinson & Pincus, 2003; Miller & Campbell, 2008; Wink, 1991). Vulnerable narcissists are modest and self-inhibited, and they have a defensive and insecure grandiosity that obscures feelings of inadequacy, incompetence and negative affect. Grandiose narcissism is related to grandiosity, arrogant behavior, entitlement, exploitativeness, enviousness, aggression, and dominance (Dickinson & Pincus, 2003; Miller et al., 2011). When looking at narcissism in organizations, it is presumable that this will be grandiose narcissism; someone who is extraverted, confident, dominant, and charming, but also aggressive, lacking in empathy, and unwilling to take criticism. In this study, we focus on grandiose narcissism, since this type of narcissism is of particular concern to organizations, especially with regard to leadership (Campbell, Hoffman, Campbell, & Marchisio, 2011).
2.4 Narcissism and Leadership
Leadership positions are useful stages for narcissists to achieve their goals of self-enhancement and superiority (Campbell & Campbell, 2009). Therefore, narcissists are prone to be in these positions. An important motivation for narcissists to aim for leadership positions is their "desire to ganger the power they need to ‘structure an external world’ that supports their grandiose needs and visions” (Rosenthal & Pittinsky, 2002, p. 618). Further evidence suggests that fantasies of power and status (Raskin & Novacek, 1991), need for power (Carroll, 1987) and dominance (Emmons, 1984) are also reasons why narcissists seek leadership positions.
These observations are supported by research that linked narcissism and leader emergence. According to the implicit leadership theory, people choose their leaders based on how well
14 someone’s characteristics match the perception of the prototypical leader (Lord, Foti, & DeVader, 1984; Shondrick, Dinh, & Lord, 2010). Since narcissists possess characteristics, such as extraversion, dominance, and high self-esteem, that match these prototypical leadership characteristics, they tend to emerge as leaders (Judge, Bono, Ilies, & Gerhardt, 2002). Several studies indeed found that narcissists emerge as leaders (Brunell et al., 2008; Grijalva et al., 2015; Nevicka, De Hoogh, Van Vianen, Beersma, & McIlwain, 2011; Ong et al., 2016; Sautter, Brown, Littvay, Sautter, & Bearnes, 2008). Brunell and colleagues (2008) found in their study that narcissists rise as group leaders in leaderless discussion groups. Also, Nevicka, De Hoogh et al. (2011) found that in an experimental setting, in which they manipulated the context, narcissists emerge as leaders. Furthermore, Sautter et al. (2008) established that business students, compared with students from other university majors, have the highest levels of narcissism, confirming the suggestion that narcissists seek out leadership positions (Campbell et al., 2011).
There has been discussion about whether it is good or bad to have narcissistic leaders in organizations (Braun, 2017; Campbell & Campbell, 2009; Campbell et al., 2011; Grijalva et al., 2015; Grijalva & Harms, 2014; Higgs, 2009; Rosenthal & Pittinsky, 2006). It is likely that characteristics like charisma, visionary, assertiveness, and charm can be beneficial for followers, and characteristics like lack of empathy, aggression, and arrogance may have detrimental effects on the followers. These positive and negative sides of narcissism are often referred to as the ‘bright' and the ‘dark' side of narcissism (Campbell et al., 2011).
The primary focus of literature has been on the dark side of leader narcissism (Rosenthal & Pittinsky, 2006). Arrogance, lack of empathy, aggression, and exploitative behaviors are often associated with the dark side of narcissism (Campbell et al., 2011, Rosenthal & Pittinsky, 2006). Narcissistic leaders make decisions without consulting others, assuming that their own ideas are always the best. This may have detrimental effects on for example voice behaviors of followers
15 because the benefits do not outweigh the costs of speaking up (Detert & Burris, 2007). Narcissistic leaders also make decisions without concern for others. Their lack of empathy causes that narcissists are seen as cold, leaving followers alone with their struggles (Rosenthal & Pittinsky, 2006; Volmer et al., 2016). Besides, the lack of empathy and exploitative behaviors of narcissistic leaders are associated with unethical behavior (Campbell et al., 2011). Furthermore, exploitative behaviors, (e.g., narcissistic leaders taking credit for achievements of their followers and blaming followers for their own mistakes) may lead to a higher stress level in followers (Volmer et al., 2016).
Since there are so many infamous examples of narcissists, one might ask oneself how a narcissistic leader can possibly be beneficial. However, the prevalence of narcissists in leadership positions over the world suggests that there must be something in narcissists that is positive as well (Rosenthal & Pittinsky, 2006). Scholars suggest that the self-confidence and dominance of narcissists are precisely what may inspire followers (Gladwell, 2002; Hogan, Curphy, & Hogan, 1994). Certain situations call for leaders that have a vision and can take dramatic action. Especially in today’s chaotic situations, society needs leaders who have grand visions and urge innovations; something that narcissistic leaders can do. Further, narcissistic leaders are willing to do what it takes to get the job done and always have the big picture in mind (Maccoby, 2000). Furthermore, Volmer et al. (2016) found that narcissistic leaders are beneficial for follower’s career success.
In summary, scholars have not yet established whether it is good or bad to have a narcissistic leader since both positive and negative sides of narcissism are found. In this paper, we focus on the duality of narcissistic leaders: they can show both a bright and a dark side simultaneously. We explore the bright side via charismatic leadership and its relationship with follower well-being, and the dark side via abusive leadership and its relationship with follower well-being.
16 2.3 Leader Narcissism and Charismatic Leadership
Charismatic leaders are able to communicate an attractive vision that is appealing to followers. These leaders connect with followers on an emotional level. They give meaning to work of followers by infusing work with purpose and commitment (House & Howell, 1992). By enhancing followers’ self-concept, charismatic leaders have the ability to motivate followers and inspire followers to go beyond personal goals and focus on the collective (Shamir et al., 1993). According to House and Howell (1992), charismatic leadership contains "symbolic leader behavior, visionary and inspirational ability, nonverbal communication, appeal to ideological values, intellectual stimulation of followers by the leader, and leader expectations for follower self-sacrifice and performance beyond expectations" (p. 82).
The word charisma is a Greek word, which literally means ‘gift’ (Conger & Kanugo, 1987). Max Weber was the first to describe the concept of charisma in 1947 (Den Hartog & Koopman, 2001). He outlined five components that need to be in place for charismatic leadership to happen: (1) there needs to be some kind of crisis, (2) a leader with a radical vision who presents a solution to the crisis needs to rise, (3) followers need to believe in this vision, (4) repeated successes are necessary, and (5) due to these repeated successes, followers will see the leader as an exceptional leader. Although several scholars have studied charisma and altered theory to explain charismatic leadership in organizations since Weber (1947), these five components are still used in most theories about charismatic leadership (Den Hartog & Koopman, 2001).
There is a common belief that charisma is something that someone has or does not have. Indeed, there are certain personal factors that are argued to predict charismatic leadership. Most often it is argued that physical personal factors (height, appearance, distinctive voice; Willner, 1984; Bryman, 1992), psychological personal factors (self-confidence, loyalty, need for power,
17 conviction of own values and believes; House, 1977; House & Howell, 1992), and ability personal factors (personal skill, intelligence; Locke, 1991) play a role in acquiring and maintaining charisma (Den Hartog & Koopman, 2001). However, this view seems to lack the social complexity of charismatic leadership (Den Hartog & Koopman, 2001). Charismatic leadership is more than just something that a leader possesses. Weber (1947) already described that the social relationship between a leader and a follower is essential in the case of charismatic leadership. If we only look at personal factors, it is impossible to explain this mutual relationship between the leader and follower. Bass (1985) argued that charismatic leaders have referent power and influence. This is because followers want to identify with their leader. They have respect for and trust in their leader, and that is why charismatic leaders can have this great influence (Bass, 1985). This shows that the relationship between charismatic leaders and followers is important.
Narcissists have several characteristics that charismatic leaders also possess. Narcissists are visionaries with grand visions. Through these visions, narcissists are able to attract and inspire a lot of followers (Maccoby, 2000). Confidence, self-esteem, passion, and risk-taking are also characteristics of narcissists that are found in charismatic leaders (Deluga, 1997; Maccoby, 2000). Maccoboy (2000) argued that charisma is one of the most important positive traits of narcissistic leaders. The link between narcissism and charisma is suggested by several other researchers as well (Deluga, 1997; House & Howell, 1992; Rosenthal & Pittinsky, 2006). Deluga (1997) found that, after analyzing speeches of former presidents, narcissistic leaders are seen as more charismatic. Also, Khoo and Burch (2008) found that narcissism is positively associated with aspects of charisma. Since it is argued that narcissists have charisma, they are able to show charismatic leadership. Based on the discussion above, we hypothesize the following:
18 2.4 Charismatic Leadership and Follower Well-Being
Charismatic leadership is known for the great influence it can have on followers. Therefore, it has several organizational outcomes. For example, charismatic leaders are seen as effective leaders (Shamir et al., 1993; Shamir & Howell, 1999). Furthermore, they encourage followers to perform beyond the normal level, and therefore followers engage more in organizational citizenship behavior (Banks et al., 2017; Zehir, Müceldili, Altindağ, Şehitoğlu, & Zehir, 2014), which is helping voluntarily on the job (Bies, 1989). Also, charismatic leaders are able to attract many people with their attractive vision (Bass, 1985). Moreover, charismatic leaders can influence the self-concept of followers positively. This is done by increasing follower’s motivation to make an effort by increasing their self-esteem through the communication of high expectations and confidence, and by inspiring them through the articulation of an attractive vision of a better future (Shamir et al., 1993).
Because charismatic leadership is positive and inspiring, it has presumably a positive effect on follower well-being. The well-being of followers is in the best interest of organizations. It has a major impact on several outcomes at work, such as individual performance, and therewith organizational performance (Harter, Schmidt, & Keyes, 2002). Multiple scholars have studied this effect. They found that higher employee well-being leads to better job performance (Wright & Cropanzana, 2000; Wright, Cropanzano, & Bonett, 2007). It suggests that well-being of employees needs to be an important consideration of organizations since it can affect the performance of organizations.
There are several studies that found that transformational leadership (Bass & Avolio, 1995), of which charismatic leadership is a substantial part, is positively related to job satisfaction and well-being of followers (Arnold, Turner, Barling, Kelloway, & McKee, 2007; Judge & Piccolo,
19 2004; Liu, Siu & Shi, 2010). Besides this, De Hoogh and Den Hartog (2009) found that charismatic leadership is associated with lower burnout. Bass (1985) states that appealing to emotions is what charismatic leaders do. They express more positive emotions than non-charismatic leaders do, which in turn leads to more positive moods by followers (Arnold et al., 2007; Bono & Ilies, 2006). All things considered, we argue that charismatic leadership is associated with higher follower well-being. Based on the discussion above, we hypothesize the following:
Hypothesis 1b. Charismatic leadership is positively related to follower well-being.
2.5 Indirect Relationship via Charismatic Leadership
We predict that leader narcissism is positively related to charismatic leadership and that in turn, charismatic leadership is associated with higher well-being of followers. These relationships suggest that a possible path through which leader narcissism is indirectly positively related to follower well-being is via charismatic leadership. Therefore, we hypothesize the following:
Hypothesis 1c. There is an indirect positive relationship between leader narcissism and follower well-being via charismatic leadership.
2.6 Leader Narcissism and Abusive Leadership
Abusive leadership is defined as “subordinates' perceptions of the extent to which supervisors engage in the sustained display of hostile verbal and nonverbal behaviors, excluding physical contact” (Tepper, 2000, p. 178). There are a few things that should be noted when using this definition. First, it is the subjective assessment of subordinates based on their experiences of their leaders’ behavior. This means that it is the perception of subordinates of abusive leadership. Second, the definition refers to the sustained display of abusive leadership (Martinko, Harvey,
20 Brees, & Mackey, 2013). There needs to be a continuing exposure to nonphysical hostility. Third, leaders show abusive leadership for a purpose, however, what the intended outcomes might be, is not defined. Accordingly, leaders can show abusive leadership, not because they want to harm followers, but because they want to achieve better performance. Thus, abusive leaders may mistreat their followers to achieve goals other than causing harm (Tepper, 2007). Behaviors associated with abusive leadership are blaming followers for mistakes, being rude to followers, angry outbursts, unfair evaluation of followers, and giving an unreasonable workload to followers (Keashly, Trott, & MacLean, 1994; Tepper, 2000, 2007).
There has been some research into what might actually cause that leaders engage in abusive leadership. Several scholars found that when leaders themselves are subject to the abusive supervision of their own superior, these leaders engage more in abusive leadership themselves (Liu, Liao, & Loi, 2012; Mawritz, Mayer, Hoobler, Wayne, & Marinova, 2012). Also, leaders with a history of family undermining engage more in abusive leadership (Kiewitsz et al., 2012). Further, when leaders feel stressed, they are more likely to show abusive behaviors (Burton, Hoobler & Scheuer, 2012). It is also found that certain personal characteristics predict the abusive behaviors of leaders (Zhang & Bednall, 2016).
Although already suggested by Tepper (2007) that scholars should look into whether narcissism is an antecedent of abusive leadership, studies that investigated this are rare. Only a few studies included both narcissism and abusive leadership (Burton & Hoobler, 2011; Park, 2012). According to Waldman et al. (2018), there is a need to explicate on the theoretical connection between narcissism and abusive leadership. Narcissism has already been found to relate to other negative workplace behaviors (Judge et al., 2006; Rosenthal & Pittinsky, 2006), but the empirical link specifically to abusive leadership is underdeveloped.
21 However, from the literature, we can derive arguments that narcissistic leaders are more likely to engage in abusive leadership. In several studies, it is argued that narcissists become aggressive and hostile when they are provoked. According to Baumeister, Smart, and Boden
(1996), narcissists feel especially provoked when their self-esteem is threatened. Because narcissists believe that they are superior to others, there are more possible threats to their self-esteem (Baumeister et al., 1996). Narcissists have a high, unstable self-esteem (Rhodewalt, Madrian, & Cheney, 1998). When facing threats, people with high, unstable levels of self-esteem are more likely to engage in anger and hostility (Kernis, Grannemann, & Barclay, 1989). Moreover, people with high, unstable levels of self-esteem constantly feel that their sense of self is endangered (Kernis, Cornell, Sun, Berry, & Harlow, 1993). So, narcissists tend to react with anger and hostile behaviors when their self-esteem is endangered, which they feel is continuously. Therefore, narcissists are more likely to show anger and engage in hostile behaviors.
Furthermore, it is found that narcissists react with aggressive behavior when they receive negative feedback. They use aggression as a defense mechanism to maintain their positive view of themselves (Bausmeister, Bushman, & Campbell, 2000). Moreover, when narcissists face ego threat (i.e., when their self-esteem is endangered), they also react with displaced aggression. This means that the aggressive behaviors are not only towards those directly involved in the ego threat, but also towards third parties. Narcissists’ need for social dominance and lack of concern for others are reasons why they react with aggression and displaced aggression (Twenge & Campbell, 2003).
Also, studies show that narcissists tend to dominate others and that followers see their leader as abusive when they engage in dominant, hostile, and aggressive behaviors (Aryee, Chen, Sun, & Debrah, 2007; Tepper, 2000).
Additionally, there are characteristics that are both associated with the dark side of narcissism and abusive leadership. For example, narcissism is associated with low agreeableness
22 (Paulhus & Williams, 2002), and Tepper (2007) argued that abusive leaders are low on agreeableness as well. Also, both narcissistic leaders and abusive leaders are likely to take credit for their followers’ success and blame followers for mistakes (Campbell et al., 2011; Chatterjee & Hambrick, 2007; Rosenthal & Pittinsky, 2006). Based on the discussion above, we hypothesize the following:
Hypothesis 2a. Leader narcissism is positively related to abusive leadership.
2.7 Abusive Leadership and Follower Well-Being
Abusive leadership is not uncommon in organizations. Schat, Frone, and Kelloway (2006) found that abusive leadership affects around 13.6% of the U.S. employees. They suggested that abusive leadership incurs at substantial costs. Many studies have been conducted about abusive leadership. Most studies focused on the detrimental effects that abusive leaders can have on their followers (Martinko et al., 2013). Lots of negative consequences have been found to be predicted by abusive leadership. In the first place, it is found that abusive leadership is associated with lower job satisfaction (Schat, Desmarais, & Kelloway, 2006; Tepper, 2000; Tepper, Duffy, Hoobler, & Ensley, 2004), higher turnover (intentions) (Schat, Desmarais, et al, 2006; Tepper, 2000), lower quality of life (Schat, Frone et al., 2006; Tepper 2000), and greater work-family conflict (Tepper, 2000). Also, the perception of abusive leadership relates to dysfunctional workplace behaviors like workplace deviance (Mitchell & Amsbrose, 2007). Besides, followers who perceive that their leader is abusive show lower levels of task performance (Xu, Huang, Lam, & Miao, 2012), and score lower on formal performance appraisals (Harris, Kacmar, & Zivnuska, 2007).
Additionally, abusive leadership can cause psychological distress in followers. For example, abusive leadership is linked to depression (Richman, Flaherty, Rospenda, & Christensen, 1992; Sheehan, Sheehan, White, Leibowitz, & Baldwin, 1990; Tepper, 2000), emotional
23 exhaustion (Tepper, 2000; Yagil, 2006), and burnout (Grandey, Kern, & Frone, 2007; Tepper, 2000; Wu & Hu, 2009; Yagil, 2006). The behaviors that abusive leaders show, like yelling and blaming others, can cause psychological distress because abusive behaviors can be perceived as chronic stressors. (Alexander, 2011). Chronic stressors are persistent and recurrent difficulties in life (Serido, Almeida, & Wethington, 2004) and are found to predict psychological distress (Lepore, Miles, & Levy, 1997). Since abusive leadership holds that abusive behaviors need to be shown on an ongoing basis, the abusive behaviors are chronic stressors. Therefore, such abusive behaviors may lead to lower well-being. Based on the discussion above, we hypothesize the following:
Hypothesis 2b. Abusive leadership is negatively related to follower well-being.
2.8 Indirect Relationship via Abusive Leadership
We predict that leader narcissism is positively related to abusive leadership and that in turn, abusive leadership is associated with lower follower well-being of followers. These relationships suggest that a possible way through which leader narcissism is indirectly negatively related to follower well-being is via abusive leadership. Therefore, we hypothesize the following:
Hypothesis 2c. There is an indirect negative relationship between leader narcissism and follower well-being via abusive leadership.
2.9 Moderating Role of Organizational Climate
Campbell and Campbell (2009) suggested that a way to make sense of narcissistic leadership is to look at the situational context. The role of organizational climate, as suggested by Grijalva et al. (2015), can be such a situation to consider to better understand narcissism. This is
24 because organizational climate is a situation with consequences for behavior of people in that situation (Denison, 1996). Organizational climate can be defined as “the shared perceptions of organizational policies, practices, and procedures” (Reichers & Schneider, 1990, p. 22). The organizational climate thus guides behavior of employees (Grojean, 2004).
For organizations, it might be important to know what inhibits the dark side of narcissism to reduce the detrimental outcomes that narcissistic leaders can have. Trait activation theory provides an explanation for how organizational climate can influence the relationship between leader narcissism and abusive leadership. This theory explains when personality traits are expressed and when they are not. The principle of trait activation theory is that “behavioral expression of a trait requires arousal of that trait by trait-relevant situational cues” (Tett & Guterman, 2000, p. 398). In other words, people only behave trait-like when the situation is relevant for that trait. This means that for example, someone high on aggression does not always show aggression, depending on the situation in which he or she is.
Relevant for trait activation theory are situational trait relevance and situational strength (Lievens, Chasteen, Day, & Christiansen, 2006). The former refers to situations in which the trait is thematically connected by the provision of cues, responses to which (or lack of responses to which) indicate someone’s height on the trait (Tett & Burnett, 2003). The latter refers to the compellingness of situations to behave in a certain way. When there is a strong situation, individual differences are washed out (Tett & Guterman, 2000). These two elements need to be taken into account when using the trait activation theory.
In general, narcissistic leaders will, as argued before, be perceived as abusive leaders. This is due to the aggressive, hostile, and dominant behaviors in which they engage. However, drawing on trait activation theory, we argue that there are certain situations in which the trait of narcissism will be less expressed and that therefore, abusive behaviors are less prevalent. With regard to trait
25 activation theory, Tett and Burnett (2003) already argued that organizational climate is a factor that can provide trait-relevant cues. We apply this to the trait of leader narcissism.
In this paper, we propose that a law and code climate can have an inhibiting effect on the abusiveness of narcissistic leaders. A law and code climate is a climate in which employees adhere to the codes of conduct and regulations of their profession (Tsai & Huang, 2008; Wimbush, Shepard, & Markham, 1997). When someone is a member of a professional organization, he or she should be following the codes of conduct and regulations of that organization in order not to jeopardize his or her membership of that professional organization (Wimbush et al., 1997). It is already suggested that a law and code climate can inhibit several negative behaviors like stealing, lying, and disobedience (Wimbush et al., 1997).
For an organizational climate to be thematically connected to the trait of leader narcissism, the climate needs to be in such a way that it provides an opportunity for narcissistic leaders to behave in the way they want to. This can be a climate in which behaving according to the codes of conduct and professional regulations are not important for employees. This climate can be described as a weak law and code climate. In a weak law and code climate, employees do not adhere to the codes of conduct and professional regulations and thus do not behave according to these. Therefore, the abusive behaviors of narcissistic leaders are likely. We suggest that abusive behaviors like aggressiveness, dominance, and exploitation of followers are expected behaviors of narcissistic leaders when there is no need to follow the codes of conduct and professional regulations.
On the contrary, we argue that when there is a strong situation, thus a strong law and code climate, narcissistic leaders will follow the codes of conduct and professional regulations. Thus, they behave according to the codes of conduct and professional regulations, because this is expected of them. Also, to not jeopardize their membership of the organization, narcissistic leaders
26 will comply with the regulations. The codes of conduct and professional regulations do not include the abusive behaviors in which narcissistic leaders can engage. So, a strong law and code climate decreases the abusive behaviors of narcissistic leaders.
Thus, based on trait activation theory, we argue that a strong law and code climate weakens the relationship between leader narcissism and abusive leadership because a strong law and code climate restricts the expression of the narcissism trait. This means that narcissistic leaders are not able to fully show their (abusive) typical behaviors, like aggression, dominance, and exploitation of followers, because they are expected to behave according to the code of conduct and professional regulations of the organization. Because, as said before, people can jeopardize their position in the organization when not following the regulations, narcissistic leaders will follow the regulations and show less of their typical behavior (which include abusive behaviors). So, in a strong law and code climate, narcissistic leaders will engage less in abusive leadership. Therefore, we hypothesize the following:
Hypothesis 3. A law and code climate moderates the relationship between leader narcissism and abusive leadership, so that the relationship is weaker when there is a stronger law and code climate.
2.10 Indirect conditional relationship
We combine the previous arguments about the indirect relationship via abusive leadership and the moderation of law and code climate on the relationship between leader narcissism and abusive leadership. Based on trait activation theory, we argue that a strong law and code climate can diminish the abusiveness of narcissistic leaders, after which the perception of abusive leadership decreases which in turn may lead to a less lower well-being of followers. Therefore, we hypothesize the following:
27
Hypothesis 4. The moderating effect of a law and code climate on the relationship between leader narcissism and follower well-being is explained trough abusive leadership; when law and code climate is strong, leader narcissism is less positively related to abusive leadership and thereby less negatively associated with follower well-being.
2.11 The whole model
Based on the above argument about the indirect conditional relationship of leader narcissism and law and code climate via abusive leadership on follower well-being, and the argument about the indirect positive relationship between leader narcissism and follower well-being via charismatic leadership, we present support for the whole model. We argue that leader narcissism has a positive indirect relationship with follower well-being via charismatic leadership, based on the charisma that narcissists possess, and the positive effect charismatic leaders may have on follower well-being. Besides, we argue that leader narcissism simultaneously has an indirect negative relationship with follower well-being via abusive leadership, based on the chronic stressors that narcissistic leaders provide and the negative consequences of abusive leadership for follower well-being, and that this relationship is less strong when there is a strong law and code climate in place. Therefore, we hypothesize the following:
Hypothesis 5: Leader narcissism is related to follower well-being via the indirect relationships; charismatic leadership explains the indirect positive relationship between leader narcissism and follower well-being, and abusive leadership explains the interaction between leader narcissism and law and code climate on follower well-being.
28 3. METHOD
3.1 Procedure and sample
For this study, a survey with a questionnaire was used to collect the data. We collected dyadic data. Dyads consisted of a leader and a follower. Due to the limited time available, convenience sampling was used to collect the data. A group of seven master students approached the participants. Either the leader or the follower was approached, who in turn approached their follower or leader. Participants received an email with the link to the questionnaire and instructions about it. Important to note is that is that all participants received a specific number via the email. The leader and the follower of a dyad received the same number so that their answers could be matched. Leaders rated themselves on narcissism. Followers rated the law and code climate, and they rated their leaders on charismatic leadership and abusive leadership. Followers also rated their own well-being. Leaders and followers both filled out some demographics.
In total, 310 surveys were sent out; we sent 155 surveys to the leaders and 155 to the followers. Data was collected over a period of four weeks. Of the leaders, we had 114 respondents, which resulted in a response rate of 73,5%. Of the followers, we had 123 respondents, which resulted in a response rate of 79,4%. In total, we collected 102 complete dyads (overall response rate: 76.5%). Depending on the analysis, however, the sample ranged from 95 to 102 due to missing values.
Participants were Dutch residents, working in a variety of organizations. Of the leaders, 53.9% was male, 43.1% was female, and 2.9% did not indicate their gender. The average age of the leaders was 42.11 years (SD = 10.64, range = 21-62). The average tenure of the leaders was 145.39 months (SD = 131.17, range = 4-471). Most of the leaders had followed higher level education; 83.3% (HBO, University Bachelor, University Master, or PhD). According to the
29 leaders, 53.9% had daily contact with followers, 39.2% on a weekly basis, and 4.9% only monthly. Of the leaders did 2% not indicate their contact rate with their followers.
Of the followers, 44.1% of participants were male, 54.9% was female, and 1% did not indicate their gender. The average age was 33.32 (SD = 12.49, range = 18-61). The average tenure of the followers was 76.06 months (SD = 108.77, range = 1-471). Most of the followers had also followed higher level education; 75.5% (HBO, University Bachelor, University Master). The average time that the leader and follower had been working together was 43.41 months (SD = 77.52, range = 1-540).
3.2 Measures
Leader narcissism. We measured the independent variable leader narcissism with the Narcissistic Personality Inventory (NPI) scale, which is developed by Gentile et al. (2013). This scale consists of thirteen items. The items were measured by a seven-point Likert scale ranging from (1) completely disagree to (7) completely agree. Examples of items are ‘I like having authority over other people’, ‘I like to show off my body’ and ‘I will never be satisfied until I get all that I deserve’. Cronbach’s Alpha was satisfying for leader narcissism (.89)
Charismatic leadership. The mediating variable charismatic leadership was measured by the scale of De Hoogh, Den Hartog, and Koopman, (2004). The scale consists of six items. The items were rated on a seven-point Likert scale ranging from (1) completely disagree to (7) completely agree. Examples of items are: ‘My leader displays conviction in his/her ideals, beliefs, and values’ and ‘My leader has a vision and imagination of the future’. Cronbach’s Alpha for this scale was satisfying (.89)
Abusive leadership. We measured the mediating variable abusive leadership with the scale of Mitchell and Ambrose (2007) which is based on Tepper (2000). The scale consists of five items.
30 The items were rated on a seven-point Likert scale ranging from (1) completely disagree to (7) completely agree. Examples of items are ‘My leader ridicules me’ and ‘My leader tells me my thoughts or feelings are stupid’. Cronbach’s Alpha for this scale was satisfying (.86)
Follower well-being. The dependent variable follower well-being was measured by the job related affective well-being scale. This scale is developed by Warr (1990). The scale consists of eleven items which are either positive or negative. Participants were asked how much of the time they have experienced each of the eleven items. The items were rated on a 7-point Likert scale, ranging from (1) never to (7) all the time. Examples of items are: ‘tense’, ‘uneasy’, ‘cheerful’, and ‘enthusiastic’. Cronbach’s Alpha for this scale was .84, which is satisfying.
Law and code climate. The moderating variable law and code climate was measured by the scale created by Victor and Cullen (1988). The scale consists of four items. The items were measured by a seven-point Likert scale ranging from (1) completely disagree to (7) completely agree. Examples of items are: ‘People are expected to comply with the law and professional standards over and above other considerations’ and ‘In this company, the law or ethical code of their profession is the major consideration’. Cronbach’s Alpha for this scale was satisfying (.86).
Control variables. Because Paulhus (1998) found that narcissists generally make good first impressions and that this good impression can decline over time, we controlled for the tenure that the leader has had with the follower. Furthermore, we controlled for leader gender, because men are generally more narcissistic than women (Maxwell, Donnellan, Hopwood, & Ackerman, 2010; Tschanz, Morf, & Turner, 1998). Also, controlling for gender has become common in research on narcissism (Nevicka, De Hoogh, et al., 2011).
31 4. RESULTS
In this section, the results of the statistical analysis are presented. First, we describe how we prepared the data for analysis. Then, we discuss the outcome of the correlation matrix. After that, the results of the hypotheses tests are discussed. The hypotheses were tested by using linear regression with PROCESS (Hayes, 2013). Because of the priori directional hypotheses, we conducted directional tests to test the expectations. This implicates that we divide the p-values that we get by two and that we use a 90% confidence interval.
4.1 Data preparation
Before we analyzed the data, we recoded the counter-indicative items of the variable follower well-being. Then, we checked whether there were outliers in the data. The variables follower well-being, abusive leadership, and charismatic leadership contained some outliers which we removed from the data. As recommended by Tabachnick and Fidell (2001), we used a threshold of Z = 3.29 to identify extreme outliers only. Outliers were excluded from further analysis. In total,
fives cases were deleted for the analysis. Two cases of charismatic leadership were deleted (Z = -3.9, Z = -3.39). Two cases of abusive leadership were deleted (Z = 4.44, Z = 4.60). One case of follower well-being was deleted (Z = -3.90).
4.2 Descriptive statistics and correlations
The means, standard deviations, and correlations of all variables used in this study are displayed in Table 1. As expected, leader narcissism was positively associated with charismatic leadership (r = .224, p = .025). However, the results did not show a relationship between leader narcissism and abusive leadership (r = .025, p = .803). Furthermore, the results showed a
32 relationship between leadership style and follower well-being: charismatic leadership was positively related to well-being (r = .226, p = .017) and abusive leadership was negatively related to well-being (r = -.210, p = .026). This could indicate that followers feel better at their job when leaders show charismatic leadership and that followers feel worse when leaders show abusive leadership. Furthermore, the results showed a significant negative relationship between charismatic leadership and abusive leadership (r = -.275, p = .004) and between follower well-being and tenure with the leader (r = -.209, p = .039). A significant positive relationship was found between charismatic leadership and law and code climate (r = .233, p = .013).
33 Table 1. Means, standard deviations, and correlations
Variables M SD 1 2 3 4 5 6 7
1. Leader Gender 1.460 .501 _
2. Tenure Leader and Follower 42.775 76.084 -.043 _
3. Leader Narcissism 4.018 .912 .056 -.105 (.89)
4. Charismatic Leadership 5.563 .760 -.041 -.088 .224* (.89)
5. Abusive Leadership 1.381 .585 -.129 .180 .025 -.275** (.86)
6. Follower Well-being 5.499 .615 .115 -.209* -.052 .226* -.210* (.84)
7. Law and Code Climate 5.059 1.190 -.033 .054 .119 .233* .053 -.089 (.86) Note: Tenure is in months, Cronbach Alpha's are indicated on the diagonal, *p<.05, **p<.01, ***p<.001
34 4.3 Hypothesis testing
Hypothesis 1. In the first model, we tested Hypothesis 1a. This hypothesis stated that leader narcissism is positively related to charismatic leadership. Results are displayed in Table 2. This model was not statistically significant (F (3,92) = 1.845, p > .05). However, results showed that leader narcissism is significantly related to charismatic leadership (b = .185, p < .05), which means that Hypothesis 1a is confirmed.
In the second model, we tested Hypothesis 1b and Hypothesis 1c. Hypothesis 1b stated that charismatic leadership is positively related to follower well-being. Hypothesis 1c stated that leader narcissism is indirectly positively related to follower well-being via charismatic leadership. Therefore, it was both tested whether charismatic leadership is related to follower well-being and whether leader narcissism is related to follower well-being. The second model was statistically significant (F (4,91) = 3,058, p < .05) and explained 11.9% of the variance in follower well-being. Results showed that charismatic leadership is significantly related to follower well-being (b = .212,
p < .05), which is in line with Hypothesis 1b. Further, we found no direct relationship between
leader narcissism and follower well-being (b = -.109, p > .05). These findings together confirm an indirect relationship. The indirect relationship via charismatic leadership was also tested with bootstrap estimation approach with a thousand samples. Indeed, results of this test showed that there is an indirect relationship with an effect size of .039, (90% CI [.0060, .1163]), which is significant since the confidence interval does not contain the value of zero. Thus, results suggested that there is a positive indirect relationship between leader narcissism and follower well-being via charismatic leadership. This means that Hypothesis 1c is confirmed.
35 Table 2. Regression results of the indirect relationship via charismatic leadership
b SE p 90% CI F R2
Charismatic Leadership
Model 1 1.845 .057
Constant 4.983*** .434 .000
Leader Gender -.091 .158 .283
Tenure Leader with Follower -.001 .001 .248
Leader Narcissism .185* .087 .019
Follower Well-Being
Model 2 3.058* .119
Constant 4.571*** .539 .000
Leader Gender .144 .126 .128
Tenure Leader with Follower -.002* .001 .038 Charismatic Leadership .212** .083 .006
Leader Narcissism -.109 .071 .066
Indirect effect .039 .034 .0060, .1163
36 Hypothesis 2. In the first model, we tested Hypothesis 2a. This hypothesis stated that leader narcissism is positively related to abusive leadership. Results are displayed in Table 3. In the first model, we tested whether leader narcissism is related to abusive leadership. This model was not statistically significant (F (3,92) = .832, p > .05). Results did not indicate a significant relationship between leader narcissism and abusive leadership (b = .052, p > .05). Therefore, Hypothesis 2a is rejected.
In the second model, Hypothesis 2b and Hypothesis 2c were tested. Hypothesis 2b stated that abusive leadership is negatively related to follower well-being. Hypothesis 2c stated that leader narcissism is indirectly negatively related to follower well-being via abusive leadership. So, it was both tested whether abusive leadership is related to follower well-being and whether leader narcissism is related to follower well-being. The second model was statistically significant (F (4,91) = 2.272, p <.05) and explained 10,7% of the variance in follower well-being. Results showed a significant relationship between abusive leadership and follower well-being (b = -.200, p < .05), which means that Hypothesis 2b is confirmed. No direct relationship was found between leader narcissism and follower well-being (b = -.009, p > .05). However, in order to accept Hypothesis 2c, both Hypothesis 2a and Hypothesis 2b need to be confirmed. Since Hypothesis 2a is not confirmed, Hypothesis 2c cannot be accepted. Indeed, results did not show a significant indirect relationship via abusive leadership (b = -.010, 90% CI [.0530, .0050]). Thus, results suggested that there is no indirect relationship between leader narcissism and follower well-being via abusive leadership. So, Hypothesis 2c is not confirmed.
37 Table 3. Regression results of the indirect relationship via abusive leadership
b SE p 90% CI F R2
Abusive Leadership
Model 1 .832 .026
Constant 1.319*** .325 .000
Leader Gender -.132 .117 .136
Tenure Leader with Follower .001 .001 .214
Leader Narcissism .052 .068 .216
Follower Well-Being
Model 2 2.716* .107
Constant 5.639*** .361 .000
Leader Gender .157 .120 .098
Tenure Leader with Follower -.002* .001 .024
Abusive Leadership -.200* .106 .032
Leader Narcissism -.010 .067 .444
Indirect effect -.012 .020 -.0530, .0050
38 Hypothesis 3. The third hypothesis stated that the relationship between leader narcissism and abusive leadership is moderated by a law and code climate so that the relationship is weaker when the law and code climate is stronger. Results are shown in Table 4. After mean centering leader narcissism and law and code climate and computing the interaction term, we first entered all variables in the model, except for the interaction term. The model was not statistically significant (F (4,91) = .688, p > .05). Result indicated that leader narcissism is not significantly related to abusive leadership (b = .056, p > .05), neither was law and code climate significantly related to abusive leadership (b = -.026, p > .05).
In the second model, we added the interaction term. Again, the model was not statistically significant (F (1,90) = .588, p > .05). Like in the first model, results showed that leader narcissism is not significantly related to abusive leadership (b = .059, p > .05), neither was law and code climate (b = -.027, p > .05). Also, results indicated that the interaction effect of leader narcissism and law and code climate on the relationship between leader narcissism and abusive leadership was not significant (b = .017, p > .05), This means that law and code climate does not moderate the relationship between leader narcissism and abusive leadership. So, the relationship is not weakened by a law and code climate. Thus, Hypothesis 3 is not confirmed.
39 Table 4. Regression results of the moderation effect of law and code climate on the relationship between leader narcissism and abusive leadership
b SE p R2 ΔR2
Abusive Leadership
Model 1 .029
Constant 1.663*** .314 .000
Leader Gender -.134 .118 .129
Tenure Leader with Follower .001 .050 .215
Leader Narcissism .056 .066 .202
Law and Code Climate -.026 .050 .301
Model 2 .030 .001
Constant 1.658*** .317 .000
Leader Gender -.131 .119 .136
Tenure Leader with Follower .001 .001 .208
Leader Narcissism .059 .068 .194
Law and Code Climate -.027 .050 .299
Leader Narcissism X Law and Code Climate
.017 .065 .496
40 Hypothesis 4. The fourth hypothesis stated that the moderating effect of law and code climate on the relationship between leader narcissism and follower well-being is explained through abusive leadership. This meant that we had to test for moderated mediation. An indirect conditional relationship is demonstrated when the relationship between leader narcissism and follower well-being via abusive leadership differs in strength across different (high and low) levels of law and code climate. We did not find support for this indirect conditional relationship. Although abusive leadership was significantly related to follower well-being (see Table 3), results did no show a significant the interaction effect of law and code climate on the relationship between leader narcissism and abusive leadership (see Table 4). This resulted in a mediated moderation index of -.003 which was not significant (90% CI [-.0228, .0185]) because the confidence interval included the value of zero (see Table 5). This means that higher or lower levels of law and code climate do not influence the strength of the relationship between leader narcissism and follower well-being via abusive leadership, which is not in line with Hypothesis 4. So, Hypothesis 4 is rejected.
Table 5. Regression results of the indirect conditional effect of law and code climate on the relationship between leader narcissism and follower well-being via abusive leadership
Index SE 90% CI
Indirect conditional effect -.003 .018 -.0228;.0185
41 Hypothesis 5. The last hypothesis stated that leader narcissism is related to follower well-being through the indirect relationship via charismatic leadership and the indirect relationship of the interaction between leader narcissism and law and code climate via abusive leadership. With this hypothesis, we tested the whole model. Firstly, we tested whether the indirect relationship via charismatic leadership between leader narcissism and follower well-being is still significant when the whole model is tested. Results are displayed in Table 6. We found indeed that the indirect relationship via charismatic leadership is still significant; leader narcissism was significantly related to charismatic leadership (b = .156, p < .05), charismatic leadership was significantly related to follower well-being (b = .184, p < .05), and leader narcissism was not directly related to follower well-being (b =- .051, p > .05).
Table 6. Regression results of the indirect relationship via charismatic leadership in the whole model b SE p F R2 Charismatic Leadership Model 1 2.084 .105 Constant 5.731*** .247 .000 Leader Gender -.089 .158 .289
Tenure Leader with Follower -.001 .001 .223
Leader Narcissism .156* .089 .042
Law and Code Climate .144* .067 .017
Leader Narcissism x Law and Code Climate
42 Secondly, we tested whether the indirect conditional effect of leader narcissism and law and code climate via abusive leadership on follower well-being was significant. Results are displayed in Table 7. The test resulted in a mediated moderation index of -.001, which was not significant since the 90% confidence interval includes zero ([-.0272;.0140]).
Follower Well-Being
Model 2 3.332* .157
Constant 4.443*** .600 .000
Leader Gender .174 .120 .075
Tenure Leader with Follower -.002* .001 .028
Charismatic Leadership .184* .084 .015
Abusive Leadership -.127 .116 .139
Leader Narcissism -.051 .068 .229
Note: N = 95, *p<.05, **p<.01, ***p<.001
Table 7. Regression results of the indirect conditional effect of law and code climate via abusive leadership in the whole model
Index SE 90% CI
Indirect conditional effect -.001 .013 -.0272;.0140
43 The results of both tests together resulted in the rejection of Hypothesis 5; although the indirect positive relationship between leader narcissism and follower well-being via charismatic leadership was found, the indirect conditional effect of leader narcissism and law and code climate through abusive leadership on follower well-being is not found.
5. DISCUSSION
Many studies have been conducted that looked at narcissism in organizations. Although most studies focused on the detrimental outcomes of narcissism (i.e., the dark side of narcissism), there are also studies that emphasize the positive side of narcissism (i.e., the bright side of narcissism). There is still no consensus about whether having narcissists in organizations is a good or a bad thing (Grijavla et al., 2015). In the present study, we examined whether leader narcissism could have a bright and dark side at the same time. Further, we examined if the possible dark side of leader narcissism could be diminished in a particular organizational climate. To be specific, we tested whether there was an indirect positive relationship between leader narcissism and follower well-being via charismatic leadership (the bright side) and whether there was an indirect negative relationship between leader narcissism and follower well-being via abusive leadership (the dark side). Besides this, we tested if there was a moderation effect of law and code climate on the relationship between leader narcissism and abusive leadership.
In our study, we found evidence for the relationship between leader narcissism and charismatic leadership, the relationship between charismatic leadership and follower well-being, and the indirect relationship via charismatic leadership. This means that narcissistic leaders are perceived as more charismatic, and as a result, followers are likely to score higher on well-being. This finding is in line with what several scholars already suggested; that narcissistic leaders are
44 charismatic leaders (House & Howell, 1992; Maccoby, 2000; Rosenthal & Pittinsky, 2006). However, these suggestions were mostly conceptual, lacking empirical evidence. We extend the literature by providing this empirical evidence with the results of our study. With regard to the positive relationship between charismatic leadership and follower well-being, this is in line with previous findings. By expressing positive emotions, charismatic leaders may trigger these positive emotions in followers as well (Arnold et al., 2007; Bono & Ilies, 2006); a phenomenon called emotional contagion (Hatfield, Cacioppo, & Rapson, 1993). Earlier findings also suggested that
charismatic leadership leads to lower burnout, which implies a higher well-being (De Hoogh & Den Hartog, 2009). We further extend literature with evidence for the indirect relationship between leader narcissism and follower well-being through charismatic leadership. As far as we know, we were the first that examined this relationship in an empirical setting. Although it is a logical inference based on the arguments presented in the theoretical background, this relationship was never examined in this way before.
Even though we did predict a relationship between leader narcissism and abusive leadership, our results did not prove this. Not finding this relationship is in line with the findings of Wisse and Sleebos (2016). They argued that narcissists only engage in abusive behaviors if their ego is threatened, for example by followers who give them negative feedback. When narcissists are left unprovoked, they are not likely to show aggression (Jones & Neria, 2015; Jones & Paulhus, 2010). Another explanation might be that the perception of abusive leadership depends on followers’ personality. This was suggested by Nevicka, De Hoogh, Den Hartog and Belschak (2018). Nevicka and colleagues (2018) found that followers with a low self-esteem saw leaders as more abusive, while followers with high self-esteem did not see their leader as more abusive. They argued that followers with low self-esteem perceive narcissistic leaders as more abusive, and that narcissistic leaders actually engage more in abusive leadership toward these followers. They also