Risky business: How narcissistic CEOs induce information asymmetry and misalignment within their board and how board financial expertise mitigates that

(1)

Risky business: How narcissistic CEOs induce

information asymmetry and misalignment within their

board and how board financial expertise mitigates that

Abstract:

Taking on the Upper Echelon (UE) theory perspective, this paper analyses the effect of Chief Executive Officer (CEO) narcissism on organizational risk. This relationship is mediated by the share of independent directors appointed by the CEO. This paper argues that a highly narcissistic CEO will surround themselves with independent directors who adhere to the narcissistic tendencies of the CEO. Consequently, this leads to information asymmetry and misalignment (i.e., agency conflicts) which leads to an increase in organizational risk. Moreover, the moderating effect of board financial expertise is studied, arguing that a higher level of board financial expertise reduces the information asymmetry and misalignment and therefore reduces the organizational risk. The hypotheses are tested through a multi-level regression with a random effect on 496 firms of the S&P 500 list between the period of 2011 to 2017. This paper finds empirical support that CEO narcissism has a small positive relationship with the share of independent directors appointed by the CEO. With this finding, this paper contributes to the combined literature of UE and agency theory. Specifically, it contributes to a deeper understanding of how personality attributes of the CEO affect organizational outcomes, while taking board financial expertise into account.

June 18, 2020 Mitra Shahbaz | S3772926 m.shahbaz.1@student.rug.nl

MSc BA: Strategic Innovation Management

Faculty of Economics and Business, University of Groningen Supervised by Michelle Weck

Co-assessed by Marvin Hanisch

Word count

12.902 words

Key words

(2)

1 – INTRODUCTION

“Why is one firm more successful than another?” Scholars have repeatedly considered different

angles and perspectives to find this answer, one angle which has received increasing attention is the role of leadership (Tamer, 2011), particularly that of the Chief Executive Officer (CEO). It is considered that brave vision, high self-confidence and winner-takes-all mentality are CEO attributes that shift these leaders to success (Savchuk, 2019). CEOs of successful companies, such as Apple’s late Steve Jobs and Tesla’s Elon Musk are both portrayed to be leaders with those attributes (Savchuk, 2019). As CEOs hold the ability to decide on strategic decisions which influence the firm (Agnihotri & Bhattacharya, 2019; Chatterjee & Hambrick, 2007; Cragun, Olsen & Wright., 2019; Picone et al., 2014) analysing the attributes becomes significantly important.

A particular stream of literature that has been used to analysed CEO attributes is the Upper Echelon (UE) theory (Hambrick & Mason, 1984). When confronted with strategic decision-making situations, UE theory argues that those situations show the CEO’s behaviour and leadership style. UE theory argues that executives make decisions based on their personal interpretations. These personal interpretations are based on previous experience and executive personality, as well as demographic attributes (Hambrick & Mason, 1984). While the demographic attributes (i.e., experiences, values and personalities) have received plenty of attention (Garg & Eisenhardt, 2017; Gerstner, König, Enders, & Hambrick, 2013; Hambrick, 2007; Hambrick & Chen, 1996; Hiller & Hambrick, 2005; Liu, Fisher & Chen, 2018), the personality attributes (i.e., race, gender or personality) of a CEO have remained underexplored (Chatterjee & Hambrick, 2011). Considering the power and influence a CEO has (Haynes et al., 2019; Ingersoll, Glass, Cook & Olsen, 2019; Urban, 2019), taking the personality (i.e., psychological) attributes of a CEO provides great potential to expand the scope of UE literature. One particular CEO personality attribute that has received particular attention is narcissism (Chatterjee & Pollock, 2017; Gerstner et al., 2013; Judge, LePine & Rich, 2006; O’Reilly, Doerr & Chatman., 2018; Peterson, Galvin & Lange, 2012; Rosenthal & Pittinsky, 2006; D. H. Zhu & Chen, 2015). Research has shown that not only do narcissists have a higher likelihood to emerge as leaders (Blair, Hoffman & Helland, 2008; O’Reilly et al., 2018), they also show potential to stay longer in a leadership position (Martin & Combs, 2011). This is perfectly portrayed by the previously mentioned successful CEOs, Steve Jobs and Elon Musk, as they are also described as narcissists (Savchuk, 2019). With successful narcissists, people tend to only see the successful 10% and ignore the irreparable damage which is created that makes up the other 90% (Savchuk, 2019). Considering the power and influence of a CEO on strategic decision-making, the goal of this paper is to consider the unexplored 90% of narcissism and what the effects are on organizational risk.

(4)

cognitive element of narcissism, while the constant reinforcement of that self-view stems from the motivational element. Both cognitive and motivational elements of narcissism have shown the tendency to drive a narcissistic individual towards risky decisions. With the mention of a narcissistic individual, this paper considers the high degree of narcissistic traits within that individual.

Based on the UE theory, this paper argues that CEO narcissism is positively related to organizational risk (Buyl, Boone & Wade, 2019; Chatterjee & Hambrick, 2007; Liu et al., 2018; O’Reilly et al., 2018; D. H. Zhu & Chen, 2015). This argumentation is based on a narcissists' cognitive and motivational elements. Considering their highly unrealistic perception of success and failure, narcissists are positively biased to think their strategies will result in positive outcomes (Buyl et al., 2019; O’Reilly et al., 2018). Therefore, the risk-intensity of a strategy would not be a matter of considering. Moreover, narcissistic CEOs will take on bold and challenging tasks which will bring them – in their opinion – the highest praise, attention and fame (Chatterjee & Hambrick, 2007, 2011). Additionally, this paper argues that the share of independent directors appointed by the CEO, will mediate the relationship between CEO narcissism and organizational risk (Bruni-Bossio & Sheehan, 2013; Cohen, Frazzini & Malloy, 2012). With the argumentation of the mediating effect, another theory comes into perspective; the agency theory. Even though narcissistic CEOs want high-status directors in their presence as that validates their own belief of their intelligence and ability (Chatterjee & Pollock, 2017), narcissistic CEOs also see everything as a threat (Maccoby, 2017). Based on that and the fact that narcissists want to have the highest praise and attention (Buchholz, Jaeschke, Lopatta & Maas, 2018; Buyl et al., 2019; Chatterjee & Hambrick, 2007, 2011; Chatterjee & Pollock, 2017; Kashmiri, Nicol & Aurora, 2017), information asymmetry and misalignment within the board can be induced by the narcissistic CEO. Moreover, combined with UE, narcissism has an influence on experience and information processing which further shows that agency conflicts can occur within a board, which result in a higher organizational risk. Finally, this paper argues that board financial expertise moderates the relationship between the share of independent directors appointed by the CEO and organizational risk. Prior literature shows how financial expertise provides the ability to find information faster and cheaper due to their experience (Minton, Taillard & Williamson, 2014; Sarwar, Xiao, Husnain & Naheed, 2018). This suggests that the more financial experts on the board, the better the strategies from a narcissistic CEO can be analysed to avoid high risk strategies.

(5)

Ou, Tsui & Wang., 2017). However, as an R&D-intensive approach, explorative innovation is also very costly and time-consuming (Ahuja & Lampert, 2001; Jansen et al., 2006), especially when it is constantly repeated. This in turn, could lead to resource depletion which could then result in a survival threat for the firm (Zhang et al., 2017). Innovation is one of the fields in which narcissism shows the potential to increase organizational risk up to a point where it jeopardizes firm survival.

The hypotheses in this paper are tested based by a multi-level regression with a random effect on a panel dataset of 496 S&P500 firms over a period from 2011 until 2017. The CEO narcissism variable was measured by using two out of the four points of Chatterjee & Hambrick’s (2007 & 2011) four-point factors of CEO narcissism. Organizational risk was measured by dividing standard deviation of the return on equity by the mean of return on equity over three years. The share of independent directors appointed by the CEO was measured as the percentage of independent directors on the board whom were appointed by the CEO. Lastly, the data collection for the moderating variable was conducted manually and was measured as the percentage of the board with financial expertise.

The findings of this paper suggest that CEO narcissism has a small influence on the share of independent directors appointed by the CEO, however this relationship does not result in a higher organizational risk. The findings also suggest that a higher share of board financial expertise does not result in lower organizational risk. Based on the findings, the contributions of this paper are as follows. Firstly, it adds to UE literature by showing the influence of CEO attributes on organizational risk. Moreover, the consideration of the share of independent directors appointed by the CEO adds to CEO narcissism literature as it showed a small positive significant relationship. As previous literature mainly takes on the effect on top management teams or managers in regards to CEO narcissism (Buyl et al., 2019; Chatterjee & Pollock, 2017; Kashmiri et al., 2017; Liu et al., 2018), this paper aimed to contribute to the UE theory and agency theory literature by involving the independent directors. Additionally, through the specific measurement of the share of financial board expertise, this paper adds to the UE and agency theory literature. Other than the previous positive relationship found, the positive relationships that this paper argued were not supported, contrary to prior literature. By taking these findings, academics can further elaborate on the UE and agency theory literature. Managers can take these insights and potentially adjust their policies to account for narcissism within the firm.

(6)

2 – THEORETICAL FRAMEWORK & HYPOTHESES

DEVELOPMENT

2.1 – Theoretical framework

According to prior literature, UE theory comprises two interconnected parts (Buchholz et al., 2018; Dubey, Gunasekaran, Childe, Papadopoulos, Hazen, & Roubaud, 2018; Gerstner et al., 2013; Hambrick, 2007). First, faced with strategic decision-making, it is argued that executives act based on their personalized interpretations. Second, these personalized interpretations (i.e., personalized construals) are in turn based on the executives’ personalities, values and experiences (Hambrick, 2007; Hambrick & Mason, 1984)). This UE perspective, initiated by Hambrick and Mason (1984), sparked an academic interest towards the relationship between executive attributes and firm outcomes (Liu et al., 2018). Since then, UE scholars have examined the effects of different executive attributes on organizational outcomes (Gerstner et al., 2013), which resulted in three different literature streams (Liu et al., 2018). The first stream takes into examination the attributes of the top management team (TMT) which affect firm performance. Various studies in this stream argue that organizational outcomes are often better explained when the attention is focused on executive groups rather than executive individuals (Hambrick, 2007). The second stream is focused on the attributes of the individual CEO, which affect firm strategy. Obtaining psychometric data of executives is notoriously difficult which has resulted in the use of mainly demographic indicators such as education and industry background (Gerstner et al., 2013; Hambrick, 2007) to draw implications on personal attributes. Lastly, the third stream combines the previous two streams and examines the individual CEOs’ attributes and the impact of those on TMT processes (Liu et al., 2018). Accordingly, all three streams have a different focus.

This paper takes into consideration the impact of the individual CEOs’ attribute on the share of independent directors who are appointed by that CEO, which in turn influences organizational risk. Notwithstanding the indirect nature of this effect, this paper can be categorized in the second UE theory stream. The share of independent directors refers to the percentage of the board who are non-executive directors who do not have any kind of relationship with the firm to affect the independence of their judgment (Akbar, Kharabsheh, Poletti-Hughes & Shah, 2017).

(7)

particularly attracted to the high-status (D. H. Zhu & Chen, 2015) and the fame (Agnihotri & Bhattacharya, 2019; Al-Shammari et al., 2019; Campbell, Goodie & Foster, 2004) that comes with being a CEO, is narcissism. Narcissistic CEOs, compared to their non-narcissistic peers, have a different reaction to fame and celebrity status (Agnihotri & Bhattacharya, 2019). Therefore, predominantly driven by the fame (Vaknin, 2008), narcissists perform better when they find an opportunity (i.e., the CEO position) that is accompanied by said attention and praise (Buchholz et al., 2018). In fact, Martin and Combs (2011) state that the reason why some CEOs survive longer than others, might be explained through certain managerial constructs, such as narcissism. Thus, narcissism as a CEO personality attribute influences the type of information that is processed and bears implications on strategic decision making, consequently increasing organizational risk.

Considered to be a personality dimension, narcissism is a factor that influences CEOs decisions. Narcissism has been researched thoroughly throughout history (Chatterjee & Pollock, 2017). Introduced in psychology research (1898) and later adopted in the management field (1987), narcissism in this paper defines as “the degree to which an individual has an inflated sense of self and is preoccupied

with having that self-view continually reinforced” (Al-Abrrow et al., 2019; Chatterjee & Hambrick,

2011, p. 204; D. H. Zhu & Chen, 2015). Researchers have had a particular interest in executives’ personalities, specifically Chief Executive Officers (CEOs), as they have a significant influence on the forms and fate of the companies they work for (Chatterjee & Hambrick, 2007; Gerstner et al., 2013). Throughout the years researchers have taken on different approaches when analysing CEOs and their personalities, such as psychoanalytic tradition, psychometric surveys for specific personality dimensions, and content analyses of biographies (Chatterjee & Hambrick, 2007, 2011; D. H. Zhu & Chen, 2015). These analyses show how narcissism shifted from being viewed as a personality disorder in earlier studies, to being re-conceptualization as a personality dimension in later studies (Chatterjee & Hambrick, 2007; D. H. Zhu & Chen, 2015), thereby transcending its former view as a clinical personality disorder (Campbell et al., 2004). It is argued that narcissism comprises cognitive as well as motivational elements. The cognitive side mainly portrays the inflated self-views of narcissists, who believe that they are extremely talented and have superior qualities such as leadership, intelligence, competence and innovativeness (D. H. Zhu & Chen, 2015). Specifically, these inflated self-views typically increase a narcissists’ confidence in their abilities and judgement in task domains. The motivational side depicts the continuous need of narcissistic individuals to have their inflated self-views reaffirmed through various types of behaviour that invites applause and admiration. In order to achieve this narcissistic individuals tend to favour the extreme, bold and challenging tasks especially those visible to highly respected audience (Chatterjee & Hambrick, 2007; D. H. Zhu & Chen, 2015).

(8)

potential lack of alignment of goals, preferences, and actions between the two parties (Nyberg et al., 2010) UE theory combined with agency theory shows that CEOs can interpret and process information however they perceive it, based on previous experiences, own goals or preferences, which consequently can result in misalignment and information asymmetry. This can only be increased with a narcissistic CEO who has highly skewed personalized interpretations (i.e., inflated self-view and self-interest) and could inflict managerial mischief and misalignment for their own benefit. This in turn, potentially results in an increase of organizational risk.

Finally, this paper argues that the agency conflicts induced by a narcissistic CEO within the board could be resolved through an increase of board financial expertise. Through experience and knowledge (Sarwar et al., 2018), it is argued that when the share board financial expertise is higher, the likelihood of organizational risk decreases.

2.2 – Hypotheses development

2.2.1 – CEO narcissism and organizational risk

Values and experiences of narcissistic CEOs significantly differ from less or non-narcissistic CEOs (O’Reilly et al., 2018), as their main concern is usually themselves (Agnihotri & Bhattacharya, 2019). Prior UE literature argues that the lessons CEO’s learn from their previous experiences on boards are likely to be influenced by narcissism (D. H. Zhu & Chen, 2015). Their cognitive side suggests unlimited talent and competence, which suggests that narcissistic CEOs believe they, compared to their non-narcissistic peers, learn more from the same learning opportunity (D. H. Zhu & Chen, 2015). Based on that, it can be stated that narcissism negatively influences a CEOs learning and information processing (Liu et al., 2018). UE theory takes prior experience and the ability to process information to make strategic decisions as a starting point to explain how CEO attributes affect firm outcome. Thus, based on UE theory, it can be argued that a CEO attribute such as narcissism can have a significant role in affecting outcomes on firm-level.

(9)

they state that narcissists’ perception of the probability of success and failure is, which consequently leads to bold and risky actions. Secondly, based on their self-interest and self-centred behaviour, narcissistic CEOs are highly focused on how their decisions affect their own personal rewards rather than what the effect of their decisions are on the fates of others (Buyl et al., 2019). Therefore, acting as a careful steward of organizational resources and avoiding high-risk strategies is unlikely for a narcissistic CEO; especially considering that narcissists have been characterized with a weak risk avoidance motivation (Buyl et al., 2019). Lastly, based on the motivational side of narcissism, it is argued that narcissists are devoted to fame (Vaknin, 2008); they are characterized by strong desire to attract the admiration and attention of the audience (Chatterjee & Hambrick, 2007, 2011). Moreover, this desire to stand in the spotlight drives narcissists to pursue strategies and policies that, in their opinion, warrant that praise (Chatterjee & Hambrick, 2007, 2011). Hence why narcissistic CEOs pursue high-risk strategies, as these strategies are usually those that draw the most attention and applause when executed successfully. As Buyl et al. (2019) argue, using high-risk high-return strategies can result in the depletion of resources, which eventually makes it more difficult to recover from when the strategies are unsuccessful.

Moreover, Agnihotri and Bhattacharya (2019) state that risk-taking propensity in extrovert and achievement-oriented CEOs was high. For narcissistic CEOs however, this risk-taking propensity “was

so high that it even resulted in fraudulent activities” (Agnihotri & Bhattacharya, 2019, p. 893). Based

on the preceding arguments, the following hypothesis is formulated:

Hypothesis 1: There is a positive linear relationship between CEO narcissism and organizational risk.

2.2.2 – CEO narcissism and the share of independent directors appointed by the CEO

(10)

credibility and attention for their contributions, and whose perspectives are sought by others as well (Chatterjee & Pollock, 2017). This consequently leads to those desirable directors to ask more questions in regards to the strategic decisions of the narcissistic CEO. From the narcissistic CEOs perspective, these are all aspects that would be perceived as negative as it takes away the spotlight and questions the intelligence and credibility of the CEO. Therefore, it can be argued that as an attempt to keep the fame, praise and attention narcissistic CEOs crave, they induce misalignment and information asymmetry within the board. In 2012, in 59% of S&P500 firms the CEO was the only “insider” (Bruni-Bossio & Sheehan, 2013, p. 4). This particular fact makes inducing agency conflicts especially easy for CEOs as they hold the inside knowledge of the firm’s business models and goals. Based on UE as well, narcissistic CEOs process information and their past experiences differently (Al-Shammari et al., 2019; D. H. Zhu & Chen, 2015), which only increases the potential for agency conflicts.

Moreover, Chatterjee & Pollock (2017) highlight the importance of loyal top management teams (TMT) for a narcissistic CEO rather than the high-status desirable directors. According to Chatterjee and Pollock (2017, p. 711), narcissistic CEOs will surround themselves with loyal

“lieutenants” who will defend and protect their decisions when needed. Exerting this power over the

whole firm is easier in smaller firms, such as start-ups, than for larger firms as the latter spans more layers of management and control (Chatterjee & Pollock, 2017). Therefore, it is argued that:

Hypothesis 2: There is a positive linear relationship between CEO narcissism and share of independent

directors appointed by CEO.

2.2.3 – Share of independent directors appointed by the CEO and organizational risk

Building on the agency theory, Chu, Mathieu and Mbagwu (2019) argue that when the interest of the CEO and the independent directors is aligned it could be that riskier projects which are more beneficial for shareholders are favoured. This implies that when the share of independent directors appointed by the CEO within the board is higher, the likelihood for managerial mischief increases. This in turn, could result in a higher organizational risk. Additionally, it is argued that CEOs and directors could use their existing power to develop a coalition, to shift the balance of power further towards their own direction (Krause, Withers & Semandeni., 2017). When the CEO appoints the biggest share of the independent directors, it is unlikely that they will challenge the power of the CEO. As stated by Krause et al. (2017), when a firm does not have a powerful board, it is unlikely to challenge the power of the CEO. Consequently, this suggests that the high-risk high-return strategies presented to the independent directors by the CEO, would not be turned down and consequently, the organizational risk could increase.

(11)

the firm’s business model they show the inability to act on an informed basis (Bruni-Bossio & Sheehan, 2013; Kim, Mauldin & Patro, 2014). Before the SOX Act, CEOs would have resolved this challenge by hiring individuals with increasing knowledge of the industry and focal firm. However, after the enactment of the SOX Act, this restrains the opportunity of experienced inside directors to serve on the board. When the independent directors do not have the right information to make informed decisions, due to their lack of knowledge on the focal firm, the organizational risk increases.

Hypothesis 3: There is a positive linear relationship between share of independent directors

appointed by CEO and organizational risk.

In sum, I argue that the share of independent director mediates the relationship between CEO narcissism and organizational risk for two reasons. The first reasoning continues to build on the agency theory, however considering CEO narcissism, it takes another perspective. The CEO is usually one of the few members on the board who serves as an “insider” while for example, 84% of directors on S&P 500 companies in 2012 were independent (i.e., an outsider) (Bruni-Bossio & Sheehan, 2013, p. 4). This means that the information in regards to the firm’s business model and strategic decisions can be manipulated by a narcissistic CEO, inducing a misalignment and information asymmetry between the CEO and the board. A narcissistic CEO, based on their inflated self-views and self-interest (Al-Abrrow et al., 2019; Chatterjee & Hambrick, 2007; D. H. Zhu & Chen, 2015), have more incentive to withhold valuable information from their board. As argued by Maccoby (2014), narcissistic leaders see everything as a threat therefore, even though there could be alignment between the CEO and the board, the narcissistic CEO could still withhold information in order to eventually be the one receiving the praise. Consequently, due to the information asymmetry created through the withheld information, the independent directors could increase the organizational risk. Secondly, the CEO still holds power in the SOX Act regulations to appoint the independent board members. As stated by Cohen et al. (2012, p. 1043), boards (and therefore also CEOs) could still engage in “window dressing” and appoint independent directors that would be more towards their own favour. Thus, it is argued that both these reasons are amplified when the firm has to deal with a narcissistic CEO rather than a less or non-narcissistic CEO.

Hypothesis 4: Share of independent directors appointed by CEO will mediate the positive linear

relationship between CEO narcissism and organizational risk.

2.2.4 – The moderating effect of share of board financial expertise

(12)

situations (i.e., bounded rationality). To be able to interpret these complex and uncertain situations, the prior experience and expertise of board members is highly significant. Having financial experts on corporate boards is generally preferred, however, following the twofold premise of the SOX Act this preference turned into an obligation. Thus, as defined by section 407 in the SOX Act, a financial expert would be “a person who has experience in accounting or finance or has supervisory expertise” (Sarwar et al., 2018, p. 1841).

The role of board financial expertise has not only improved different firm outcomes, such as corporate performance and firm performance, it also improved board efficiency (Sarwar et al., 2018). Moreover, the boards’ decisions are ultimately influenced by the presence of financial expertise within the board (Sarwar et al., 2018). Furthermore, due to their experience, financial experts have an easier time collecting information on complex financial transactions and their associated risks. Consequently, they are more capable and efficient at monitoring senior management (Minton et al., 2014). With this knowledge, the greater share of financial experts, the better boards are equipped to recognize risky strategies that would be harmful for the firm’s financial stability (M. Adams & Jiang, 2020; Minton et al., 2014). Based on the prior argumentation, it is argued that with a higher share of board members with financial expertise, independent board of directors can reduce the information asymmetry (M. Adams & Jiang, 2020) and misalignment induced by a narcissistic CEO. Finally, as shown in the conceptual model (Figure 1), this paper hypothesizes:

Hypothesis 5: The relationship between share of independent directors appointed by CEO and

organizational risk is moderated by the share of board financial expertise. This positive relationship is accentuated when the share of board financial expertise is lower and attenuated when the share of board financial expertise is higher.

(13)

3 – DATA AND METHOD

3.1 – Sample

This study examined firms from Standard & Poor’s (S&P) 500 Index for the period 2011-2017, which implies 3500 firm years. Primary data on proxy statements (DEF 14A) were gathered by using EDGAR (Electronic Data Gathering, Analysis and Retrieval), which is a system used at the U.S. Security and Exchange Commission (SEC). The financial data was gathered from the Datastream database and board data was gathered from the BoardEx database. To compose the CEO narcissism variable, annual compensation data was used which was collected from the Compustat – Capital IQ database. To gather the data for the moderating effect, share of board financial expertise, it needed to be transferred to an individual level. This resulted in a number of 68,598 observations over the period 2011 to 2018. As this paper examined the data from the period 2011 to 2017, the year 2018 needed to be subtracted from the observations. This resulted in a number of 60,192 observations for the period 2011 to 2017, which is the goal sample which was aimed for. The data gathered from EDGAR included the actual collected data of the share of board financial expertise. This was a number of 59,808 observations, which also included the year 2018. Once the observations for year 2018 were deducted, there were 54,571 observations. When dividing the actual observations with the observations of the sample goal, 90,6% of the sample goal was reached. This sample goal was not fully reached due to missing financial data of some firms and also missing data from EDGAR. Additionally, the previously mentioned observations were on individual level however, this paper takes on a board/firm level which is why the observations had to be adjusted accordingly. Once the observations were adjusted to board/firm level, the data was merged with the Compustat dataset which resulted in the final sample of 2889 firm years. As previously state the sample goal was 3500 firm years, the actual sample was 2889 firm years, which means that 82,5% of the sample goal was reached. Due to missing Compustat data the sample goal was not fully reached.

3.2 – Measures

3.2.1 – Dependent variable

The dependent variable organizational risk was calculated by variables obtained from the Datastream database. The dependent variable was calculated by dividing standard deviation of the return on equity by the mean of return on equity over three years. As it seemed more fitting to the period of this research (2011-2017) and in regards to potential CEO changes, the dependent variable was chosen for three years rather than five years.

3.2.2 – Independent variable

(14)

based on four points: (1) The prominence of the CEO’s photograph in the company’s annual report, (2) CEO prominence in the company’s press releases, (3) CEO’s relative cash compensation and (4) CEO’s relative non-cash compensation. Due to data availability, only the relative cash and relative non-cash compensation were calculated. Both points were calculated by using the annual compensation data from Compustat – Capital IQ. The relative cash compensation was measured as the total of the CEO’s salary and bonus divided by that of the second highest paid executive. In this research, the second highest paid executive is considered to be the Chief Financial Officer (CFO). The relative non-cash compensation was measured as the total of the CEO’s monetary value of deferred income, stock grants, and stock options divided by that of the next highest paid executive (CFO). The CEO narcissism variable was found by using the mean of the relative cash and non-cash compensation. The items relative cash and non-cash compensation explained approximately 8.9% of the variance, Cronbach’s alpha was 0.41.

3.2.3 – Mediating variable

The mediating variable share of independent directors appointed by the CEO was calculated by finding the share of independent director who were appointed after the CEO joined the firm (%). This data was available from the BoardEx database.

3.2.4 – Moderator variable

The moderating variable share of board financial expertise was found through primary data. This data was gathered by finding the proxy statements (DEF 14A) of S&P500 firms, by using the SEC database. Financial expertise included looking for key words in the proxy statements such as “experience in banking, finance, accounting, or economics related activities” (R. B. Adams et al., 2018, p. 645). Share of board financial expertise was then measured by the share of directors on the board with financial expertise (%).

3.2.5 – Control variables

(15)

difference in risk-aversion and narcissistic tendency is controlled for by checking for female CEOs. Lastly, on CEO-level, this paper controlled for CEO nationality as cultural dimensions such as power distance, uncertainty avoidance and masculinity had a significant result on performance (Urban, 2019; H. Zhu et al., 2020). As the S&P500 firms are mainly American firms, the CEO nationality received 1 when “American” and 0 for “Other”.

Then, on the board-level, board size and share of women on the board were controlled for. Firstly, based on agency theory, larger boards can experience coordination and communication problems (Gonzalez & André, 2014) which result in slow decision-making. Consequently, reaching consensus within the board can be very time-consuming (Akbar et al., 2017) which reduces the risk-taking propensity of the board. Then, the final control variable on board-level is the share of women on the board. This variable was controlled for based on what is noted above; women are more risk-averse than men (Ingersoll et al., 2019; Zhang et al., 2017; H. Zhu et al., 2020). Since this paper uses the mediating variable share of independent directors appointed by the CEO, checking for the share of women on the board becomes important. Thus, it could be argued that the higher the share of independent directors appointed by the CEO consists of women, the higher likelihood that risky-strategies are avoided.

Furthermore, on firm-level this paper checked for firm size, which was measured as the natural log transformed number of employees (Agnihotri & Bhattacharya, 2019; Al-Shammari et al., 2019; Kashmiri et al., 2017). Often larger firms considered themselves “too-big-to-fail”, however it has become apparent that firm size drives many common systematic risk indicators (Rossoni & Mendes-Da-Silva, 2019; Varotto & Zhao, 2018, p. 45). Therefore, firm size is taken into consideration.

Finally, on industry-level, this paper checked for industry dummies and year dummies. As shown by prior literature, the level of risk-taking behaviour differs per industry (Chatterjee & Hambrick, 2007; Chatterjee & Hambrick, 2011; Kashmiri et al., 2017). Based on the four-digit SIC codes industry dummies were controlled for. Then, seven year dummies were used to check whether there would be any variance in the years (Agnihotri & Bhattacharya, 2019; Al-Shammari et al., 2019).

3.2.6 – Analytical model

(16)

4 – RESULTS

4.1 – Regression results and hypothesis testing

Table 1 in presents the descriptive statistics and correlations of the variables of the measures. Generally, correlations between 0.5 and 1 are considered high correlations, those between 0.3 and 0.49 are medium correlations and everything below 0.29 is considered a small correlation. Therefore, table 1 only shows small correlations, as the highest correlation is between share of independent directors

appointed by CEO and CEO age (r = 0.2678). The second highest correlation is the correlation between share of women on the board and year which is r = 0.2585. The third highest correlation is the

correlation between board financial expertise and industry (r = 0.2496). Finally, the last correlation is the one between share of independent directors appointed by the CEO and CEO narcissism (r = 0.1524). Due to the weak relationships of the correlations, they become of no concern for this paper.

To test the hypotheses, a multilevel analysis (using the xtreg command) with a random effect was used on the panel dataset. Table 2 in provides the results of the mediation analysis.

Model 1 in Table 2 provides the regression results of only the control variables and the main

effect of organizational risk. Here the overall R2 is 0.1333, which means that 13.3% of the dependent variable organizational risk, is explained by the control variables. Also, this model shows the significance of the control variable CEO age (β = 0.017, p = 0.018).

Model 2 presents the result of Hypothesis 1, which predicts a positive relationship between

CEO narcissism and organizational risk. Here, the overall R2 shows that 13.3% of organizational risk is explained by CEO narcissism and the control variables. However, as the R2 is similar to the R2 of Model 1, it indicates that the variability of the dependent variable organizational risk is not explained by the independent variable, CEO narcissism. Moreover, for Hypothesis 1 this paper argues a positive relationship between CEO narcissism and organizational risk, while the results show a negative insignificant result (β = -.007, p = .511). Therefore, the results show no empirical support for Hypothesis 1. Similar to Model 1, CEO age has a significant result as well in Model 2 (β = 0.019, p = 0.011).

Model 3 shows the relationship between share of independent directors appointed by the CEO

and CEO narcissism. The R2 in this model shows 0.2875, which indicates that 28.3% of share of

independent directors appointed by the CEO is explained by CEO narcissism. Here, as predicted in

Hypothesis 2, results show positive and significant results (β = .001, p = .001). Even though the β -value is low, the results show that there is empirical support for Hypothesis 2. In this model, the results for the control variables differ compared to model 1 and model 2. Here, CEO age (β = 0.021, p = 0.001),

board size (β = 0.008, p = 0.009), share of women on the board (β = 0.519, p = 0.026), and female CEO

(β = 0.162, p = 0.055) have significant results.

Model 4 presents the result of Hypothesis 3, which predicts a positive relationship between the

(17)

the R2 explains 13.3% of the dependent variable. Moreover, as the results is negative and insignificant, there is no empirical support for Hypothesis 3 (β=-.048, p =.746).

Finally, Model 5 presents the joint effect of CEO narcissism and independent director appointed by the CEO on organizational risk. With a small difference to model 1, 2 and 4 the R2 in model 5 shows that the independent variables explain 13.2% of the dependent variable. The interaction effect in this model shows that both independent variables CEO narcissism and share of independent

directors appointed by the CEO do not largely explain the variance of the independent variable organizational risk. The results show that the effect of CEO narcissism (β=-.007, p =.541) as well as

(18)

(19)

Table 2 – Mediating effect Model 1 (DV: Organizational risk) Model 2 (DV: Organizational risk) Model 3 (DV: Independent director appointed by CEO) Model 4 (DV: Organizational risk) Model 5 (DV: Organizational risk) Intercept 0.28(1.33) 0.20(1.38) -4.66(.23) 0.26(1.33) 0.19(1.38) Covariates CEO age 2.37(.01) 2.54(.01) * 20.33(.00) *** 2.32(.01) * 2.52(.01) * Board size -1.28(.02) -1.15(.02) 2.60(.00) ** -1.27(.02) -1.14(.02) Share of women on board 0.34(.51) 0.43(.52) 2.22(.07) * 0.35(.51) 0.44(.52) Female CEO -0.61(.66) -0.57(.66) 1.92(.08) + -0.60(.66) -0.56(.66) Firm size -0.21(.00) -0.12(.00) -0.92(.00) -0.21(.00) -0.13(.00) CEO nationality -1.24(.06) -1.25(.06) 0.33(.01) -1.23(.06) -1.25(.06)

Main predictors

CEO narcissism a -- -0.66(.01) 4.07(.00) *** -- -0.61(0.01) Share of independent directors

appointed by CEO a -- -- -- -0.32(.15) -0.39(.15) Model fit Wald-Chi2 or R2 0.1333 0.1326 0.2875 0.1332 0.1324 (Prob > Chi2) 0.4496 0.7995 0.0000 0.4549 0.8083 N individual 2,607 2,603 2,611 2,607 2,603 N firm level 497 496 498 497 496

a Winsorized at the 1st and 99th percentile levels

(20)

Finally, Table 3 in provides the results of the moderation effect using a multi-level analysis with a random effect. Model 1 in this table shows the results for Hypothesis 5, which predicated that the positive relationship between independent directors appointed by the CEO and organizational risk becomes smaller when the board of director has financial expertise. The R2 shows that the independent variable explains 13,4% of the dependent variable. Moreover, the results show an insignificant interaction effect (β=.437, p =.400) therefore, not supporting Hypothesis 5. Despite the insignificance of the moderation effect, the margins plot depicted in Figure 2 was plotted to obtain a visualization of the moderating effect.

Figure 2 – Margins plot moderating effect.

4.2 – Robustness check

(21)

Table 3 – Moderation effect Model 1 (DV: Organizational risk) Intercept 0.21(1.33) Covariates CEO age 2.25(.01) Board size -1.16(.02) Share of women on board 0.39(.51) Female CEO -0.60(.66) Firm size -0.23(.00) CEO nationality -1.27(.06)

Main predictors

Share of independent directors appointed by CEO a -0.85(.29) Board financial expertise a 0.32(.23) Share of independent directors appointed by CEO * Share

of board financial expertise

0.84(.52) Model fit Wald-Chi2 or R2 0.1339 (Prob > Chi2) 0.4681 N individual 2,607 N firm level 497 a Winsorized at the 1st and 99th percentile levels

(22)

5 – DISCUSSION

The UE literature has sparked an increase in the interest of the effect of executive attributes on firm performance. Particularly the personality attributes of the CEO have received growing attention due to the influence of the CEO on firm decisions and outcomes. One personality attribute specifically has received a growing body of research, which is narcissism. Even though there is a growing body in regards to the effects of narcissism, what the effects are of a narcissistic CEO who induces information asymmetry and misalignment within the board of directors remains to be understood.

In this paper, it is argued that firms with a more narcissistic CEO are more subject to organizational risk as the CEO induces information asymmetry and misalignment within the board if the share of independent directors were mostly appointed by that CEO. Several reasons could explain this argumentation. Firstly, a more narcissistic CEO feels superior compared to others, which results in their own belief that they can take riskier decisions without any negative outcome. Moreover, they crave for the validation, attention and praise of others to confirm their own superior feelings, which they achieve by choosing challenging and risky strategies. When narcissistic CEOs choose their independent directors, they choose those independent directors who would be more sympathic to the CEO (i.e., window dressing). In turn, the share of independent directors appointed by the CEO on the board are manipulated and misinformed by the CEO from the start resulting in lower knowledge in terms of the firms’ business model and goals. This, in turn, results in those independent directors appointed by the CEO, to support the CEO in their risky behaviour and therefore increasing the organizational risk. It is argued that the share of board financial expertise moderates this relationship. When the majority of board members have more financial expertise, they are more likely to understand the consequences of the risky strategies that the CEO is pursuing. Therefore, the higher the share of board financial expertise is, the lower the organizational risk gets.

To empirically test the hypotheses, a random-effects model using a panel dataset from a time period between 2011 to 2017, with a sample of 496 firms and 2,603 observations was used. The data supported the second hypothesis arguing that CEO narcissism positively increases the share of independent directors appointed by the CEO. The other hypotheses, contrary to what was hypothesized, did not receive empirical support. These results can be explained both theoretically and empirically. First, the potential reasons for the insignificant results of the mediating effect are discussed, which includes the first four hypotheses.

(23)

reduce their risk-taking in order to maintain the success they have built. Moreover, it is explained why narcissistic CEOs would tend to bold, challenging and therefore riskier strategies. However, due to their inflated self-confidence and prior experience, narcissistic CEOs are also committed to repeat their own strategies (Hiller & Hambrick, 2005) which are not necessarily riskier, bolder or more successful. Also, as overconfident CEOs tend to engage in fast decision-making (Hambrick & Chen, 1996; Hiller & Hambrick, 2005). Consequently, it could be argued that a narcissistic CEO could also stick to strategies which are faster to execute and not necessarily risky. Based on the results, this paper is aligned with Cragun et al. (2019), as they also did not find any empirical evidence that CEO narcissism influences risk-taking. The argumentation here is that the results could be affected by the measurement and the construct definition of risk.

One hypothesis throughout this paper has a significant result, which was the relationship between CEO narcissism and the share of independent directors appointed by the CEO (hypothesis two). As argued before there could be several reasons for this. As there is still room for “window dressing”, and appointing independent directors whom sympathize more with the board, narcissistic CEOs could influence this appointment process. Mainly, as stated by Krause et al. (2017), when a firm lacks a powerful board, then it is highly unlikely that they will question a powerful CEO. This could explain the significant result for the second hypothesis.

However, the share of independent directors appointed by the CEO did not result in higher organizational risk (hypothesis three and four). Here the main argumentation contrary to this hypothesis, would be that even though independent directors are no “insiders” of the firm, based on previous board experience they could still assess risky strategies. This paper did not consider any demographic or personality attributes of the independent directors, which could explain the insignificance of the third and fourth hypothesis.

(24)

Then, the theoretical and empirical reasoning of the insignificance of the moderating effect. This paper argues that the higher share of financial board expertise moderates the relationship between share of independent directors appointed by the CEO and organizational risk (hypothesis five). This paper only took financial expertise as a measure to reduce organizational risk however, other expertise could have contributed to less organizational risk as well. An example would be the risk and high-cost innovation (i.e., exploration) strategies which could jeopardise firm survival (Jansen et al., 2006). Therefore, the insignificance of the moderator could be that only a higher share of board financial expertise on the board is not enough. Moreover, this paper did not consider the combined effect of board financial expertise and board status, as argued by Badolato, Donelson and Ege (2014). In their paper, Badolato et al. (2014) find that without considering board status, increasing the financial expertise is insufficient. The push for more financial experts, as required by the SOX Act, might not have the expected effect on lowering accounting irregularities as new directors have lower average status compared to existing directors (Badolato et al., 2014). As this paper considered that a narcissistic CEO would appoint lower status independent directors, to fit more to their risky behaviour, taking into consideration board status could have influenced the results. Lastly, this paper argued that independent directors would reduce the risk-taking propensity of a narcissistic CEO if they have more financial expertise. One reason as to why this result was insignificant could have been that independent directors are also considered shareholders of the firm. Using this perspective, where the narcissistic CEO and independent directors are aligned, it could be argued that a higher board financial expertise could result in even more risk-taking as there would be a great potential of benefit for the shareholders (Minton et al., 2014).

Lastly, the potential empirical reason for the insignificant evidence of all the hypotheses; the sample size. Even though using S&P500 firms is fairly common to test hypotheses (Oehmichen, Schrapp & Wolff, 2017; Rijsenbilt & Commandeur, 2017), to extend the sample size S&P1000 or S&P1500 firms could have been used as well. An extension of the sample size could have increased the completeness of the study and perhaps contributed to significant results.

In sum, after careful analysis of prior literature this paper presented the hypotheses previously stated. The results showed that one relationship out of the five relationships, received empirical support. There is a small positive evidence that CEO narcissism has an effect on the share of independent directors appointed by the CEO. However, based on the possible arguments previously stated, the other relationships received no empirical support.

5.1 – Theoretical implications

(25)

focused on as they influence firm outcomes (Chatterjee & Hambrick, 2007). Current UE literature has continuously found evidence that CEO narcissism results in higher organizational risk (Al-Abrrow et al., 2019; Buyl et al., 2019; O’Reilly et al., 2018), which is contrary to the findings of this research. The results of this paper show that a narcissistic CEO does influence the share independent directors appointed by the CEO, which provides a small opening to find further research directions in regards to the combined effect of UE and agency theory attributes. Even though this paper fails to show evidence for the remaining hypotheses, it provides a small step for academics on deepening the understanding and combining effect of UE and agency theory literature.

Furthermore, in an effort to explain higher organizational risk due to CEO narcissism, this paper took the effect of independent directors on the board. This approach differs from the three previously identified streams by UE literature (Buyl et al., 2019; Chatterjee & Pollock, 2017; Kashmiri et al., 2017; Liu et al., 2018), as it considers independent directors rather than TMTs. Since this paper found a small positive relationship between CEO narcissism and the stream of independent directors, it contributes new insights to academics in UE and agency theory literature in regards to independent directors.

Moreover, this paper focused on the moderating effect of share of board financial expertise. This focus aimed to contribute more to the agency theory literature as previous studies have highlighted the importance of financial expertise on the board (M. Adams & Jiang, 2020; Minton et al., 2014; Sarwar et al., 2018) Financial experts on a board have prior experience and knowledge which allows them to analyse the strategic strategies proposed by the narcissistic CEO better, faster and with lower costs (Minton et al., 2014). However, the results in this paper show inconsistency with prior literature. This means that when the share of independent directors is mostly appointed by the narcissistic CEO, a higher share of board financial experts does not result in lower organizational risk. The reason for this could be that this paper took on multiple different industries into consideration, whereas previous literature in that matter focused more on the financial sector.

In sum, academics can take on these findings and further elaborate on the UE and agency theory literature, perhaps by using other measures for the variables or building on other literature streams. In the final section of this chapter, future research directions are further elaborated on.

5.2 – Practical implications

(26)

a narcissistic CEO, it shows the essential need for shareholders to understand when they are dealing with a narcissistic CEO. Though the support found for the second hypothesis was rather small, practitioners are advised to still consider attributes from psychology that effect CEOs in their future studies especially since CEOs holds substantial power over strategic decisions within the firm.

Moreover, in this paper, the sample relies on S&P500 firms which are mostly American firms. Therefore, the small evidence found from mostly American firms that CEO narcissism positively effects the share of independent directors appointed by the CEO, becomes an interesting factor to consider for practitioners inside and outside of America. Even though the support found is relatively small, considering the fact that the SOX Act is enacted only in America, CEO narcissism could have more effect in countries outside of America. Therefore, especially in times where firms are more becoming more international, practitioners should take this into consideration when hiring their next CEO.

5.3 – Limitations and future research

This research has several limitations, which could be taken into consideration for future research. Firstly, this study mainly focuses on the narcissism as a CEO personality attribute. Even though the board of directors is taken as a mediating variable in this paper, narcissism within the board is not measured. As argued by Chatterjee and Hambrick (2017) narcissistic CEOs are more likely to attract high-status board members, as shown from research on status homophily. For a narcissistic CEO, having an affiliation with high-status directors shows the possibility to attract other high-status directors and possibly increase the CEO’s own status (Chatterjee & Hambrick, 2017). Thus, there would be a possibility that a narcissistic CEO attracts narcissistic directors. Future research could take into consideration the personality attributes of independent directors into consideration and potentially find how CEO narcissism and director narcissism combined influence organizational outcomes.

In regards to the more empirical limitations and future research directions, this limitation revolves around the measurement of the independent variable, CEO narcissism. As previously discussed, this paper used two out of the four CEO narcissism factors developed by Chatterjee & Hambrick’s (2007). Due to data availability, this paper solely used relative cash and non-cash compensation as a measurement. It is encouraged for future research to repeat this study while taking all four points of measuring CEO narcissism developed by Chatterjee & Hambrick (2007) into consideration.

(27)

Chatterjee & Hambrick’s (2007 & 2011) CEO narcissism variable as Raskin & Shaw (1988) measures the pronouns in situations where CEOs speak freely. However, assuming that narcissistic CEOs would be aware of this method, the free speech in transcripts could still be influenced through predetermined speeches. Additionally, to the many studies that use this measure before this study, the measure of Chatterjee & Hambrick was used in this paper as it was the most current literature in terms of years. Future research directions could look into conducting the primary approach (Agnihotri & Bhattacharya, 2019) through CEO interviews which could result in better insights.

Moreover, as stated previously, there are three control variables which could have had an influence on the outcomes but were not tested for; CEO duality, CEO financial expertise and board busyness. CEO duality would be that the CEO holds both the position as chairman of the board and the CEO position. In combination with CEO narcissism, holding both positions as chair and CEO, could further enforce CEO power (Al-Shammari et al., 2019) which in turn could result in higher agency conflicts which would be an interesting future research direction. Moreover, my moderator involved the financial expertise of the independent directors, however this paper did not check for the financial expertise of the CEO. Prior literature shows that CEO financial expertise improves financial reporting quality as well as it reduces risk (Oradi, Asiaei & Rezaee, 2020). When combined with CEO narcissism, future research directions could consider this control variable. Lastly, Rubin & Segal (2019) argue that the more boards directors serve in, the less effective they are in monitoring management. In order to achieve their high-risk, high-return strategies, narcissistic CEOs could have pushed more towards independent directors who serve in many different boards. Therefore, board busyness could be considered for future research.

Due to scope limits, this paper took into consideration the S&P500 firms. However, in order to further complete this paper and perhaps gain new insights, future research could consider to extend the sample size by considering S&P1000 firms, or S&P1500 firms. Moreover, the sample size goal of this paper was 3500 firm years. As explained in the data and methods section, the sample size goals were not reached. Therefore, this paper could be subject to selection bias as the 9,4% and 17,5% of the unreached sample size goal could have included statistically significant data. Future research could adjust this bias by using a Heckman-test to find out whether the unused data included statistically significant data.

Also, when a mediating variable is taken into consideration, future research could test the indirect and direct measure. In this paper, the mediating effect was insignificant, however future research could find new insights by testing for the indirect and direct measure.

(28)

example, by using R&D intensity as a measuring variable, future research could incorporate innovation risk.

Finally, other underlying factors, which are not accounted for, could have influenced the relationship between CEO narcissism and organizational risk. This brings up the problems of endogeneity and reverse causality, which according to Minton et al. (2014), are always present in corporate governance studies. For endogeneity, it could be argued that another underlying factor, other than CEO narcissism, would have affected organizational risk. For reverse causality, it could be argued that firms that are more likely to take risks, attract narcissistic CEOs. However, this paper did not test for endogeneity nor reverse causality. To lift the robustness of this paper to a higher level, future research could test for reverse causality and endogeneity by using the instrumental variable approach. Several different instruments show great potential for an interesting research. Firstly, it could be interesting to see how and if narcissists are affected by having a family/relationship or pets. As narcissists generally lack empathy (Al-Abrrow et al., 2019; Gerstner et al., 2013; O’Reilly et al., 2018) and have a high self-interest (Chatterjee & Hambrick, 2007; Marquez-Illescas, Zebedee & Zhou, 2019; D. H. Zhu & Chen, 2015) it would be interesting to see how factors that require them to not think about themselves, influence their behaviour. Another instrument that could be interesting would be social media. Based on their need for attention (Buyl et al., 2019; Chatterjee & Hambrick, 2007; Kashmiri et al., 2017) and fame (Agnihotri & Bhattacharya, 2019; Al-Shammari et al., 2019; Campbell et al., 2004) social media can easily feed into CEOs narcissism. Therefore, it could be interesting to see if social media has a high presence in amplifying the narcissism within narcissistic CEOs.

6 – CONCLUSION

(29)

REFERENCES

Adams, M., & Jiang, W. (2020). Do Financial Experts on the Board Matter? An Empirical Test From the United Kingdom’s Non-Life Insurance Industry. Journal of Accounting, Auditing and

Finance, 35(1), 168–195. https://doi.org/10.1177/0148558X17705201

Adams, R. B., Akyol, A. C., & Verwijmeren, P. (2018). Director skill sets. Journal of Financial

Economics, 130(3), 641–662. https://doi.org/10.1016/j.jfineco.2018.04.010

Agnihotri, A., & Bhattacharya, S. (2019). CEO Narcissism and Internationalization by Indian Firms.

Management International Review, 59(6), 889–918.

https://doi.org/10.1007/s11575-019-00404-8

Ahuja, G., & Lampert, C. M. (2001). Entrepreneurship in the large corporation: A longitudinal study of how established firms create breakthrough inventions. Strategic Management Journal, 22(6– 7), 521–543. https://doi.org/10.1002/smj.176

Akbar, S., Kharabsheh, B., Poletti-Hughes, J., & Shah, S. Z. A. (2017). Board structure and corporate risk taking in the UK financial sector. International Review of Financial Analysis, 50, 101–110. https://doi.org/10.1016/j.irfa.2017.02.001

Al-Abrrow, H., Alnoor, A., & Abbas, S. (2019). The Effect of Organizational Resilience and CEO’s Narcissism on Project Success: Organizational Risk as Mediating Variable. Organization

Management Journal, 16(1), 1–13. https://doi.org/10.1080/15416518.2018.1549468

Al-Shammari, M., Rasheed, A., & Al-Shammari, H. A. (2019). CEO narcissism and corporate social responsibility: Does CEO narcissism affect CSR focus? Journal of Business Research,

104(July), 106–117. https://doi.org/10.1016/j.jbusres.2019.07.005

Badolato, P. G., Donelson, D. C., & Ege, M. (2014). Audit committee financial expertise and earnings management: The role of status. Journal of Accounting and Economics, 58(2–3), 208–230. https://doi.org/10.1016/j.jacceco.2014.08.006

Blair, C. A., Hoffman, B. J., & Helland, K. R. (2008). Narcissism in organizations: A multisource appraisal reflects different perspectives. Human Performance, 21(3), 254–276.

(30)

Bruni-Bossio, V., & Sheehan, N. T. (2013). Leveraging board expertise: Strategy mapping as teaching tool. Journal of Business Strategy, 34(4), 3–10. https://doi.org/10.1108/JBS-11-2012-0070

Buchholz, F., Jaeschke, R., Lopatta, K., & Maas, K. (2018). The use of optimistic tone by narcissistic CEOs. Accounting, Auditing and Accountability Journal, 31(2), 531–562.

https://doi.org/10.1108/AAAJ-11-2015-2292

Buyl, T., Boone, C., & Wade, J. B. (2019). CEO Narcissism, Risk-Taking, and Resilience: An Empirical Analysis in U.S. Commercial Banks. Journal of Management, 45(4), 1372–1400. https://doi.org/10.1177/0149206317699521

Campbell, W. K., Goodie, A. S., & Foster, J. D. (2004). Narcissism , Confidence , and Risk Attitude.

311, 297–311.

Chatterjee, A., & Hambrick, D. C. (2007). It’s all about me: Narcissistic chief executive officers and their effects on company strategy and performance. Administrative Science Quarterly, 52(3), 351–386. https://doi.org/10.2189/asqu.52.3.351

Chatterjee, A., & Hambrick, D. C. (2011). Executive Personality, Capability Cues, and Risk Taking: How Narcissistic CEOs React to Their Successes and Stumbles. Administrative Science

Quarterly, 56(2), 202–237. https://doi.org/10.1177/0001839211427534

Chatterjee, A., & Pollock, T. G. (2017). Master of puppets: How narcissistic ceos construct their professional worlds. Academy of Management Review, 42(4), 703–725.

https://doi.org/10.5465/amr.2015.0224

Chen, G., Crossland, C., & Huang, S. (2010). FEMALE BOARD REPRESENTATION AND CORPORATE ACQUISITION INTENSITY. Business, 313(November 2014), 1–43. https://doi.org/10.1002/smj

Cohen, L., Frazzini, A., & Malloy, C. J. (2012). Hiring cheerleaders: Board appointments of “independent” directors. Management Science, 58(6), 1039–1058.

https://doi.org/10.1287/mnsc.1110.1483

(31)

https://doi.org/10.1177/0149206319892678

Dubey, R., Gunasekaran, A., Childe, S. J., Papadopoulos, T., Hazen, B. T., & Roubaud, D. (2018). Examining top management commitment to TQM diffusion using institutional and upper echelon theories. International Journal of Production Research, 56(8), 2988–3006. https://doi.org/10.1080/00207543.2017.1394590

Garg, S., & Eisenhardt, K. M. (2017). Unpacking the CEO-Board relationship: How strategy making happens in entrepreneurial firms. Academy of Management Journal, 60(5), 1828–1858.

https://doi.org/10.5465/amj.2014.0599

Gerstner, W. C., König, A., Enders, A., & Hambrick, D. C. (2013). CEO Narcissism, Audience Engagement, and Organizational Adoption of Technological Discontinuities. Administrative

Science Quarterly, 58(2), 257–291. https://doi.org/10.1177/0001839213488773

Gonzalez, A., & André, P. (2014). Board Effectiveness and Short Termism. Journal of Business

Finance and Accounting, 41(1–2), 185–209. https://doi.org/10.1111/jbfa.12045

Hambrick, D. C. (2007). Upper echelons theory: An update. Academy of Management Review, 32(2), 334–343. https://doi.org/10.5465/AMR.2007.24345254

Hambrick, D. C., & Chen, M. (1996). Hambrick, DC, Cho, TS & Chen, M-J 1996, “The influence of top management team heterogeneity on firms” competitive moves.pdf. Administrative Science

Quarterly, 41, 659–684. https://doi.org/10.2307/2393871

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

Haynes, K. T., Zattoni, A., Boyd, B. K., & Minichilli, A. (2019). Figureheads or potentates? CEO power and board oversight in the context of Sarbanes Oxley. Corporate Governance: An

International Review, 27(6), 402–426. https://doi.org/10.1111/corg.12293

Hiller, N. J., & Hambrick, D. C. (2005). Conceptualizing executive hubris: The role of (hyper-)core self-evaluations in strategic decision-making. Strategic Management Journal, 26(4), 297–319. https://doi.org/10.1002/smj.455

(32)

Narcissism Manifests Among Women CEOs. Journal of Business Ethics, 158(4), 893–907. https://doi.org/10.1007/s10551-017-3730-0

Jansen, J. J. P., Van Den Bosch, F. A. J., & Volberda, H. W. (2006). Exploratory innovation, exploitative innovation, and performance: Effects of organizational antecedents and environmental moderators. Management Science, 52(11), 1661–1674.

https://doi.org/10.1287/mnsc.1060.0576

Judge, T. A., LePine, J. A., & Rich, B. L. (2006). Loving yourself abundantly: Relationship of the narcissistic personality to self- and other perceptions of workplace deviance, leadership, and task and contextual performance. Journal of Applied Psychology, 91(4), 762–776.

https://doi.org/10.1037/0021-9010.91.4.762

Kashmiri, S., Nicol, C. D., & Arora, S. (2017). Me, myself, and I: influence of CEO narcissism on firms’ innovation strategy and the likelihood of product-harm crises. Journal of the Academy of

Marketing Science, 45(5), 633–656. https://doi.org/10.1007/s11747-017-0535-8

Kim, K., Mauldin, E., & Patro, S. (2014). Outside directors and board advising and monitoring performance. Journal of Accounting and Economics, 57(2–3), 110–131.

https://doi.org/10.1016/j.jacceco.2014.02.001

Krause, R., Withers, M. C., & Semadeni, M. (2017). Compromise on the board: Investigating the antecedents and consequences of lead independent director appointment. Academy of

Management Journal, 60(6), 2239–2265. https://doi.org/10.5465/amj.2015.0852

Liu, D., Fisher, G., & Chen, G. (2018). Ceo Attributes and Firm Performance: A sequential mediation process model. Academy of Management Annals, 12(2), 789–816.

Maccoby, M. (2017). Narcissistic leaders: The incredible pros, the inevitable cons. Leadership

Perspectives, 31–40.

Marquez-Illescas, G., Zebedee, A. A., & Zhou, L. (2019). Hear Me Write: Does CEO Narcissism Affect Disclosure? Journal of Business Ethics, 159(2), 401–417. https://doi.org/10.1007/s10551-018-3796-3

Martin, J. A., & Combs, J. G. (2011). Better sooner than later: What triggers early CEO dismissal?

Risky business: How narcissistic CEOs induce information asymmetry and misalignment within their board and how board financial expertise mitigates that