The effect of CEO psychopathy on firm innovation performance

(1)

The effect of CEO psychopathy on firm innovation performance

Master Thesis

MSc Business Administration – Strategic Innovation Management

Faculty of Economics and Business

University of Groningen

Jasper Mulders

S2769662

j.f.mulders@student.rug.nl

Thesis Supervisor: Pedro de Faria

(2)

2

Abstract

This study builds on insights from both the upper-echelons theory and the dark triad of personality as theoretical foundations, to theorize on the effect of CEO psychopathy on firm innovation performance. Drawing on prior literature, I predict that CEO psychopathy has a negative effect on firm innovation performance, an effect that is mitigated for firms in high dynamic industries. These hypotheses are tested on a panel dataset that covers a time period from 2008 to 2016, using a sample consisting of firms in the S&P 100. I use the count of patents as innovation performance measure and developed a new psychopathy dictionary for the text analysis software Linguistic Inquiry and Word Count to measure CEO psychopathy. The hypotheses are not supported by the results of my study, which gives the indication that the level of psychopathy does not influence firm innovation performance. This finding can be explained by the fact that individuals with high levels of psychopathy do not reach CEO positions because of the selection procedures of firms. My research thereby contributes to the understanding of the psychopathy personality on CEO levels and suggests that future research should focus on the other elements of the Dark Triad of personality.

Key words: Innovation · Innovation Performance · CEO psychopathy · Dark triad · Upper-echelons

(3)

3

List of abbreviations and acronyms

CEO : Chief Executive Officer CP : Corporate Psychopath DT : Dark Triad

LIWC : Linguistic Inquiry and Word Count R&D : Research and Development ROA : Return On Assets

SIC : Standard Industrial Classification S&P : Standard and Poor’s

(5)

5

1. Introduction

Innovation performance is arguably one of the most important organizational outcomes for a firm. It has proven to be of high importance for organizations and businesses to stay competitive in their industry and outperform other firms (Brown and Eisenhardt, 1997). Innovation performance is an organizational outcome that is largely dependent on the decisions of the chief executive officer (CEO) (Mackey, 2008). A CEO is the central figure in the organization and the primary strategic decision-maker for a firm (Busenbark et al., 2016). The CEO is the center of external attention and firm governance, and thereby able to determine the strategic direction of an organization via his or her vision, decisions and policy (Finkelstein et al., 2009). The upper-echelons theory (UET) argues that organizational outcomes are a reflection of the values and cognitive bases of powerful actors in the organization (Hambrick and Mason, 1984). This theory argues that decisions of a CEO are largely affected by personal characteristics, values, and cognitive biases. A recent stream of literature (Liu et al., 2018; Tang et al., 2011) extents the UET and includes the so-called ‘Dark Triad’ (DT) of personality (Paulhus and Wiliams, 2002) into the relationship between CEO attributes and organizational outcomes. This perspective argues that the personality of a CEO and the corresponding psychological traits are significant antecedents of organizational outcomes.

(6)

6

are pro-active towards innovation (Kashmiri et al., 2017). Although researchers have described the psychopathy personality as the most malevolent of the three personalities (Rauthman, 2012), has the psychopathy personality received fewer attention by prior research than the other DT personalities, especially on CEO levels.

A reason for the small amount of research towards the psychopathy personality on CEO levels is that this personality type contains behavioral characteristics that were long been seen as relatively unsuccessful and non-present in the business world. Psychopathy is classified as a mental disorder characterized by disagreeableness, shallow affect, pathological lying, a parasitic lifestyle, callousness and a lack of conscience, regret and remorse (Cleckley, 1988; Hare, 1994). High levels of psychopathy are commonly found by criminals and prior research claims that twenty percent of the prison population in the United States (US) is considered a psychopath (Hare, 1999). Several researchers have extended this extreme view on psychopathy and argue that there are certain individuals who possess lower levels, or subclinical levels of psychopathy (Babiak, 1995; Babiak and Hare, 2006; Boddy, 2006). Individuals with lower levels of psychopathy have an ability to mask some of their undesirable personality and the ones who can function in work environments have been called corporate psychopaths (CP) (Babiak, 1995; Boddy, 2006; Clarke 2005). The presence of such an individual in an organization leads to detrimental results for the performance of an organization according to several researchers (Babiak and Hare, 2006; Boddy, 2010, 2017; Mathieu and Babiak, 2016a). There is an increased chance of unethical behavior (Harrison et al., 2018), bullying (Boddy, 2011), and workplace deviance (Dunlop and Lee, 2004). Despite these negative aspects that are perceived with CPs, researchers argue that almost every large company has such an individual working for them (Clarke, 2005).

(7)

7

personality, and ruthless leadership style, I argue that CEOs with psychopathic traits possess particular behavior traits that are detrimental to innovation performance. However, no clear empirical evidence for this relationship is provided by the literature yet. Only recently, a case study argues that the presence of a psychopathic CEO results in a decline of organizational innovativeness (Boddy, 2017). Drawing on this gap in research, I aim to address the negative effect of psychopathy on firm innovation performance. My research is guided by the following research question:

Research Question: What is the effect of CEO levels of psychopathy on firm innovation performance?

To provide a complete understanding of this relationship, this research will explore the moderating role of industry dynamism. High dynamic industries are characterized as uncertain and turbulent environments to operate in as a CEO, and can be operationalized by making the distinction between high-technology industries and other industries (Henderson et al., 2006; Thornhill, 2006). Prior research has emphasized that individuals with psychopathic tendencies show behavior traits that are seen as abilities that could thrive in uncertain and turbulent environments. These individuals have an ability to stay calm and unemotional under high pressured circumstances (Lilienfeld et al., 2012), and show behavior that includes decisiveness and willingness to take risks (Babiak and Hare 2006; Babiak et al. 2010). Following this reasoning, I assume that the industry dynamism will moderate the relation between CEO psychopathy and firm innovation performance. Implicating that the negative influence of CEO psychopathy on firm innovation performance will decrease for firms in high dynamic industries.

(8)

8

Linguistic Inquiry and Word Count (LIWC) (2015). I do not use existing psychopathy assessment tools like the Psychopathy Checklist-Revised (PCL-R) (Hare, 1991), its revised version the Psychopathy Checklist: Screening Version (PCL: SV) (Hart et al., 1995), the Psychopathic Personality Inventory (PPI) (Lilienfeld and Andrews, 1996) or the 113-item Business Scan 360 (B-Scan) (Babiak and Hare, 2016b) since these tools encounter some measurement difficulties. They rely on self-reports and reports from direct colleagues or other employees about the behavior of specific individuals. This makes the reports very much bounded to individual thoughts and preferences, and they are difficult to collect since CEOs are reluctant to participate in particular studies that could reveal detrimental results about themselves. Therefore, I applied the LIWC software employing my psychopathy dictionary on earnings conference calls and annual reports. Through these reports, which I collected with my colleagues, I am able to detect the different levels of psychopathy among CEOs.

The results of my analysis do not support my hypotheses and give the indication that the level of psychopathy does not influence firm innovation performance. This finding can be explained by the fact that individuals with high levels of psychopathy do not reach CEO positions because of the selection procedures of firms. My research thereby contributes to the existing literature in twofold. First, it draws on the limited amount of research towards the psychopathy personality on CEO levels, and examines its effect on innovation performance. Moreover, it contributes by adding a moderating variable to this relationship which allows for a better reflection of the relationship, namely industry dynamism. The results provide a better understanding of the psychopathy personality on CEO levels and suggests that future research should focus on the other elements of the DT of personality. Secondly, this study serves as a methodological contribution by providing a psychopathy dictionary for LIWC that enables researchers to detect different levels of psychopathy at individuals through their linguistic output, instead of reports from others.

(9)

9

(10)

10

2. Literature review

This literature review provides an overview of the existing literature and theories that are used in this thesis. First, literature on the three personalities of the DT will be discussed and followed by a review on the concept of corporate psychopathy. After this, the components and antecedents of innovation performance will be explained. This section ends with the development of hypotheses concerning the relationship between the constructs of psychopathy and innovation performance based on underlying mechanisms.

2.1 The Dark Triad

The DT is composed out of three personality traits that contain a degree of malevolence that directly affects interpersonal behavior (O’Boyle et al., 2012). The three traits are labeled dark based on their level of social averseness and the sharing of a tendency to be callous, selfish and malevolent in their interpersonal dealings (Paulhus and Wiliams, 2002). Together, the DT elements act as powerful psychological antecedents to fraudulent behavior (Harrison et al., 2018) and risky financial endeavors (Jones, 2014), since these individuals are more prone to participate in selfish and callous behavior.

(11)

11

charismatic leadership if it benefits themselves or promotes their own interest. Their desire to influence other people leads to an ability to impact and direct followers with charismatic leadership, which is linked to the behavior of successful communicators and leaders (Deluga, 2001). This leadership behavior, characterized by a desire to control and self-interest, leads to high competitiveness and mistrust among employees on one side, and low openness, trust, and risk-taking on the other side. This suggests an environment in which innovation becomes less important and receives less attention, and thus, becomes obsolete in the organization.

The narcissistic personality is characterized by a high level of aggrandizement and self-admiration. Narcissistic individuals are perceived as persons who show arrogance and feelings of superiority and grandiosity; who are very prone towards power, prestige, status, and leadership; and are constantly seeking attention and admiration (Raskin and Hall, 1979; Morf and Rhodewalt, 1993). They tend to exaggerate their own achievements, refuse to compromise, and block criticism that is coming their way. In interpersonal and social relationships they are regularly perceived as abrasive and insensitive to their impact on others (Judge et al., 2009). Nonetheless, their need to maintain an unrealistic and inflated esteem pushes them to external self-affirmation through social interaction. Narcissistic persons are often associated with leadership roles and prior research has shown that many CEOs with narcissistic traits lead highly successful companies (Maccoby, 2000). Leadership roles contain many facets that help such an individual to obtain his or her self-affirmation, seek for status and prestige, and need for a role with the greatest authority (Wales et al., 2013). Chatterjee and Hambrick (2007) argue that a narcissistic CEO leads to more extreme firm performance, both higher benefits and bigger losses. Narcissistic CEOs are associated with higher risk-taking, especially when they are explicitly incentivized towards risk-taking through stock options (Buyl et al., 2019). Moreover, CEO narcissism is associated with an orientation towards organizational growth, which encompasses increased risk-taking and proactiveness and ultimately leads to better innovation performance (Wales et al., 2013). Therefore, in particular circumstances, a CEO with narcissistic tendencies can be beneficial for the innovation performance of a firm.

(12)

12

to obtain what they desire like Machiavellians. Besides this overlap, the psychopathy personality is characterized by dishonesty, impulsivity, lack of empathy, callousness, remorselessness, lack of forethought and an inability to establish close interpersonal relationships (Cleckley, 1988; Hare 1994, 1999). Individuals with high levels of psychopathy are called “clinical psychopaths”, and are seen as people with a social disorder, possessed by less than one percent of the normal population. These individuals have no conscience and demonstrate an egoistic and ruthless way of living (Hare, 1994). Most of the clinical psychopaths are found in prison. However, researchers estimate that roughly 10% of the people possess moderate levels of psychopathic tendencies (LeBreton et al., 2006). These so-called “subclinical psychopaths” are able to participate in the normal society and evade contact with legal authorities. Subclinical psychopaths have high levels of self-confidence and are able to successfully influence others due to personality traits that make them charming, outgoing, cool under pressure and willing to take risks (Hare, 1999). Often, their high self-confidence leads to an arrogant and grandiose display of self-promotion (LeBreton et al., 2006), and their risk-taking is associated with highly impulsive, short-term decisions (Babiak and Hare, 2006). Subclinical psychopaths have an inability to form very meaningful relationships due to a disregard of social exchange (O’Boyle et al., 2012), and do not experience any guilt, shame, remorse or regret when their decisions hurt others (Hare, 1994). These characteristics lead to exhibited patterns of anti-social behavior based on judgments concerning one’s own well-being and desires while minimalizing that of others (Levenson, 1992). They are comfortable with exploiting and dominating others while disregarding societal norms and showing anti-social behavior (Hare, 1999).

2.2 Corporate psychopaths

Generally, subclinical psychopaths are believed to possess aversive behavioral traits that are detrimental to successful functioning in work environments (O’Boyle et al., 2012). However, there are some traits of people with psychopathic tendencies that could be beneficial in particular environments. Therefore, this section will delineate a profile about subclinical psychopaths concerning their negative and positive attributes in corporate settings, the CP.

(13)

13

(14)

14

makes them able to exploit others in order to attain their own goal to reach top positions. Their desire towards prestige and status gravitates them towards professions and organizations where they can gain power, money, and prestige (Boddy et al., 2015). Moreover, individuals with psychopathic tendencies who attain leadership positions are being positively associated with creativity, strategic thinking and communication skills shortly after they start the job (Babiak et al., 2010). CP leaders present themselves as calm and reliable people, which make them likable at first sight and able to lead employees and stakeholders to trust and follow them (Boddy et al., 2010a). The ability to move individuals within the organization in combination with their self-confidence and lack of empathy can be seen as a good leadership abilities when an organization needs a new direction in difficult times. For instance, Albert Dunlap, who was a specialist in turnaround management and hired as a CEO for many companies in financial difficulty. He was an acknowledged CP with ruthless tactics to cut costs with mass employee layoffs (Deutschman, 2005). Thus, although most of the behavioral attributes of a CP as a leader are seen detrimental for overall firm performance, some could be beneficial for an organization that faces severe problems.

2.3 Innovation performance

Innovation is defined by Thompson (1965) as: “the generation, acceptance, and implementation

of new ideas, processes, products or services.” Innovation is an important driver for firms in the urge to stay ahead of the competition and it enables firms to change or adapt to shifting environmental circumstances and technological changes. Firms need to renew themselves by exploiting existing competencies and exploring new ones (March, 1991). The starting point for innovation is creativity by individuals and teams within the organization according to Amabile et al. (1996). They define innovation as: “the successful implementation of creative ideas within an organization“. Firms must ensure that several aspects related to innovation and creativity are aligned in a way that they constitute innovation performance. The following section explains the antecedents of innovation performance and elaborates further on the UET to discover what attributes of a CEO drives or hampers innovation performance.

(15)

15

(R&D) and their willingness to take risks. Earlier work of Laverty (1996) already showed that some CEOs became risk aversive regarding innovation to stimulate short term growth. CEOs tend to choose to increase the market value of their firm, while intertemporally discounting on R&D investment. Besides the direct impact that a CEO can have on the innovation performance, there are several underlying mechanisms of innovation performance that can be influenced by CEO attributes. First, innovation requires a long-term vision of the future by an organization and the CEO (Somech and Drach-Zahavy, 2013). Innovative projects are perceived as risky and challenging while taking a long time to reap uncertain results, and CEOs need to demonstrate a thoroughly thought-out visualization of the future and long-term investments to stimulate innovation. Furthermore, the type of leadership style demonstrated by a CEO is an important attribute that has a significant influence on creativity and innovation. Transformational leadership is seen as fundamental to firm innovation because this leadership stimulates people to innovate and encourages them to implement new ideas (Howell and Avolio, 1993). Transformational leaders are able to increase the intrinsic motivation, the desire to take more challenging goals, and the individual creative performance of their employees. While transactional leadership is found to be detrimental to creativity and innovation since it is focused on short-term goals with reward and punishment systems (Amabile et al., 1996). So, innovation performance is more likely to improve when CEOs demonstrate transformational leadership with a clear long-term vision for creativity and innovation.

Additionally, to achieve a higher innovation performance, CEOs need to ensure that they create an innovation culture that stimulates creativity and makes the organization able to employ a

learning orientation. CEOs are found to be the primary influence for creating and modifying

(16)

16

committed to innovation, become less likely to miss emerging opportunities, and therefore outperform competitors (Calantone et al., 2002).

Besides a learning orientation, an innovation culture includes a shared vision, open-mindedness, and intraorganizational knowledge sharing. Shared vision refers to the organization-wide degree of consensus regarding the present and future strategic direction of the firm. A shared vision helps various divisions to overcome cross-functional communication barriers to increase information flows and coordination with other departments and stimulates a common sense for innovation (Damanpour, 1991). Intraorganizational knowledge sharing refers to the organization-wide sharing of information and resources, and the belief that knowledge should be distributed among different units in the organization to achieve a higher innovation performance (Calantone et al., 2002). Lastly, open-mindedness is the state of mind in which the organization and its members are willing to critically evaluate their daily routines and accept new ideas or concepts (Sinkula et al., 1997). Thus, to improve innovation performance, CEOs should create an innovative culture that stimulates creativity and learning within the organization.

2.4 Hypothesis Development

The relationship between CEO psychopathy and innovation performance as an organizational outcome has hardly received any attention in the literature to date. In my literature review, I have provided insights regarding the different constructs in this relationship, and in the following section, I will theorize on the development of my hypotheses.

CEOs with psychopathic traits possess several personality attributes that are considered as detrimental for achieving a high innovation performance. First of all, instead of the required long-term vision, a CP often demonstrates only a very short-minded vision of the future due to their lack of forethought, egoism, and desire for prestige and status (Boddy, 2013). In an attempt to maximize immediate wealth and power, they act in a way that enhances their own short-term interest rather than the long-term interests of their employees, coworkers, the environment and society.

(17)

17

natural tendency to bully other employees as they enjoy hurting people and they are not concerned with any financial or emotional effects of their actions on other people due to their lack of empathy and conscience. They use bullying and punishing practices as a way to distract other normal functioning staff from paying attention to their poor daily activities and fraudulent behavior (Babiak et al, 2010). A transactional leadership style focuses on short-term goals with reward and punishment systems and does not encourage people to innovate or implement new ideas. Moreover, the increased bullying practices and unfair supervision at the workplace result in lower job satisfaction, and employee turnover is expected to be much higher in organizations managed by CPs (Boddy et al., 2015). So, the transactional leadership style demonstrated by a CEO with higher levels of psychopathy focusses on short-term goals, does not encourage innovation, and leads to a high employee turnover with a loss in employee commitment and internal knowledge as a consequence.

Furthermore, the organizational culture and amount of learning orientation are important antecedents of a firm its innovation performance. Employees working under a CP manager felt that the organization lost a clear strategic direction, and there was a lot of confusion about the vision of the company (Boddy et al., 2015). The shared vision that is needed to create an organization-wide consensus about the present and future strategic direction disappeared slowly among the employees. Additionally, co-operation and information sharing among employees and different departments decreases tremendously when working under a CP (Mathieu and Babiak, 2016a). This is a result of CP CEOs trying to hide their fraudulent behavior and poor daily operations by diminishing cross-departmental communication to avoid the possibility that this behavior is coming to light (Boddy et al., 2015). Lastly, organizations with CP CEOs lack a commitment to learning, as these CEOs fail to provide training and information that is needed for their employees (Boddy, 2017). They do not encourage learning and development in the organization since they are only focused on their own well-being. Thus, working under a CEO with higher levels of psychopathy lead to a loss of shared vision, a decrease in intraorganizational knowledge sharing and a lack of learning orientation to constitute an innovative culture.

(18)

18

favor their own well-being on the short-term while neglecting the long-term goals of the organization as a whole. They demonstrate a transactional leadership style that involves bullying and punishments practices, which leads to high employee turnover and loss of employee commitment. Furthermore, the organizational culture under a CP is not innovation-oriented and lacks a clear learning orientation. There is no shared vision and commitment to learning among employees, and intraorganizational knowledge sharing decreases tremendously. Therefore, I propose the following hypothesis:

H1: The higher the level of CEO psychopathy, the lower the level of firm innovation

performance.

Nevertheless, I do not expect CEO psychopathy to have the same negative effect on innovation in every organizational environment. Besides the influence that a person’s personality has on CEO decisions and attributes, are the CEO decisions and attributes regarding strategy also thought to be highly dependent on the firm environment and the corresponding industry characteristics (Hambrick and Finkelstein, 1987). Some of these industry characteristics might therefore mitigate the negative relationship between CEO psychopathy and firm innovation performance.

(19)

19

low-tech firms chooses a CEO who fits the current strategy and follows the already chosen direction. Therefore, high dynamic or high-tech firms might experience different effects from higher levels of CEO psychopathy.

First, prior research has indicated that CPs have an ability to stay calm and unemotional under high pressured circumstances (Lilienfeld et al., 2012), and their decisiveness and willingness to take risks are seen as abilities that thrive in chaotic environments (Babiak and Hare 2006; Babiak et al. 2010). Their decisiveness may probably stimulate their short-minded vision which is negatively related to innovation performance. However, for firms in high dynamic industries, this might be beneficial since these industries require quick-acting and an ability to move the firm in a new direction.

Moreover, high dynamic industries are considered as industries in which the knowledge develops rapidly and a risk-taking culture is required to become the first mover (Thornhill, 2006). CEOs with higher levels of psychopathy are expected to limit the amount of learning orientation in an organizational culture. However, these individuals are also reported to take financial, legal and moral risks that normal individuals would not take (Babiak and Hare, 2006). Since CEOs are thought to shape an organizational culture by endorsing management practices that represent their own values (Tsui et al., 2006), their risk-taking propensity might alleviate some of the negative effects that CEO psychopathy has on the learning orientation in high dynamic industries. Thus, CEOs with psychopathic traits are able to stay calm in highly dynamic environments, able to move the organization into a new direction, and pursue highly risky strategies. Therefore, I expect that the negative effect of CEO psychopathy on firm innovation performance is mitigated when the firm is operating in a highly dynamic industry, and propose the following hypothesis:

H2: The negative effect of higher levels of CEO psychopathy on firm innovation

(20)

20

3. Research Methods

The research towards psychopathy on CEO levels find itself in an early stadium. The literature gap that I found in my literature review regarding CEO psychopathy and innovation performance will be studied by applying the LIWC text analysis software with a new to develop psychopathy dictionary for the independent variable and amount of granted patents as the dependent variable. The development and use of the psychopathy dictionary is explained in the independent variable paragraph. Furthermore, this section elaborates on the data collection, sample characteristics, dependent variable, moderator, and control variables.

3.1 Sample characteristics

The sample of firms in this research includes all firms that are in the Standard and Poor’s 100 (S&P 100) index option contract (OEX), a sub-set of the S&P 500 (us.spindices.com). The S&P 100 is founded in 1983 and designed to measure the performance of large-cap companies in the US. These companies are the largest and most stable companies in the S&P 500 and are the most actively traded stocks in the world. I choose for this sample because of the size of these firms and the availability of information about these firms. Public firms are required to give a comprehensive report on the company’s activities throughout the year to its shareholders and US companies with more than $10 million worth of assets are enforced by the SEC to share a 10-K form. My sample for this study comprises all the firms who were in the S&P 100 over a period of ten years (2008-2018). Due to many price fluctuations of stock options and therefore firm valuations, new firms came into the one-hundred firms’ list, where others left the list. Thus, the total amount of firms that are included in the list over this period was 150. The final sample of firms is slimmed down to an amount of 148 firms, since Lehman Brothers Holding Inc. was in the list but declared bankrupt in 2008, and Wachovia Corp. was acquired by Wells Fargo in 2008, so no data could be gathered.

3.2 Data collection

(21)

21

by the US Patent and Trademark Office (USTPO) to publicly listed firms internationally over the period 1980-2017. Relying on one source gives consistency, comparability, and reliability according to Ahuja (2000). Since this large dataset contains a lot of patent information about other firms, I manually merged the necessary patent data for the specific firms and years in my chosen sample. The initial observation period covered a ten year period between 2008-2018. However, the Darden dataset did not include the year 2018 and I removed the year 2017 because the patent count for every firm in that year strongly decreased in comparison to other years. I assume that the dataset does not cover all patents of the last year (2017) since many patents are awarded in the year after application. Moreover, not all 148 firms are included in the dataset, so the dataset for the dependent variable has a year range of 2008-2016 and includes 140 firms. The independent variable in my thesis is the level of psychopathy among CEOs of firms in the S&P 100 list. To measure this, I use the text analysis software LIWC and apply this software on secondary data that includes CEO sayings. Together with my colleagues, we choose to use annual reports and transcribed company earnings conference calls for the independent variable since these reports include many CEO contributions. Annual reports contain a letter to the shareholder from the CEO and CEOs have a direct influence on the end result (Chatterjee and Hambrick, 2007). The transcribed earnings conference calls are direct conversations between the CEO and its shareholders and include therefore many direct sayings of the CEO. The annual reports for the independent variable are extracted from AnnualReports.com which provides access to annual reports in their actual format. We inquire missing reports at the focal firm’s website or report them as missing in the database. If there are no annual reports available, we use the 10-K forms which are similar to annual reports. The earnings conference calls are collected with the Thomson Reuters Eikon database which provides all the quarterly earnings conference calls in transcripts (Eikon.thomsonreuters.com).

(22)

22

cases are reported as not available. In total, we hand-collected 1661 annual reports and 10-K forms and 5078 earnings calls.

Lastly, I collect the data for all the control variables through the Compustat and Execucomp databases which contain a broad and comprehensive overview of global firm data (e.g. financial and accounting data). Afterward, I merged the several variables in Excel to make further Stata analysis feasible. This merging results in a final sample of 732 observations (n=121) in Stata, which contains all the information on every variable needed for this study.

3.3 Dependent Variable

As a measurement of innovation performance, I use the count of US granted patents of firms in the S&P 100 over the period of 2008-2018. The count of patents is a non-negative count variable and comprises yearly observations. Despite the notion that there are multiple methods for the measurement of innovation performance, I choose for the count of patents because academic literature has acknowledged that patents are a primary measure for innovation performance (Ahuja, 2000). Patents are often a better estimation of innovation than R&D expenditures, since R&D measurements indicate only the budgeted resources allocated to innovation input, but not the actual innovation output (Hagedoorn and Cloodt, 2003). Moreover, patents are recognized as a rich and fruitful source of data for studies of innovation as they contain highly detailed information about the innovation and the inventors, there are a very large number of patents, and they include citations to previous patents and scientific literature (Hall et al., 2001; Balkin et al., 2000).

3.4 Independent Variable

(23)

23

in a written discourse (Pennebaker et al., 2007). LIWC calculates the percentage of words from written texts that fall in predefined categories of a dictionary. By combining several specific categories of the dictionary, one is able to determine an individual’s personality based on textual output. For instance, the narcissistic personality has effectively been captured by some researchers through LIWC categories (Chatterjee and Hambrick, 2007, 2011; Buyl et al., 2019). Until now, there was no combination of categories readily available that defines the psychopathy personality. In my research, I have therefore used the original LIWC2015 dictionary to develop a new variable that captures the psychopathy personality (Appendix 1). This variable is the is constructed with the following formula:

𝐿𝑒𝑣𝑒𝑙 𝑜𝑓 𝑝𝑠𝑦𝑐ℎ𝑜𝑝𝑎𝑡ℎ𝑦

= ∑ 𝐻𝑖𝑔ℎ 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑡𝑙𝑦 𝑢𝑠𝑒𝑑 𝑤𝑜𝑟𝑑 𝑐𝑎𝑡𝑒𝑔𝑜𝑟𝑖𝑒𝑠

∑ 𝐻𝑖𝑔ℎ 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑡𝑙𝑦 𝑢𝑠𝑒𝑑 𝑤𝑜𝑟𝑑 𝑐𝑎𝑡𝑒𝑔𝑜𝑟𝑖𝑒𝑠 + ∑ 𝐿𝑜𝑤 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑡𝑙𝑦 𝑢𝑠𝑒𝑑 𝑤𝑜𝑟𝑑 𝑐𝑎𝑡𝑒𝑔𝑜𝑟𝑖𝑒𝑠 This equitation results in a continuous variable between 0 and 1. This process is executed for both earnings conference calls and annual reports and results in two independent variables. Since the earnings conference calls contain four data points per year, the year average is taken to create consistency in our dataset. Afterward, the average level of psychopathy between both measures is calculated to create a third variable. The psychopathy variable measured with earnings conference calls is used as the primary variable to test the relationship since the voice of the CEO is the most visible through this source. The earnings conference calls contains the highest percentage of spoken words by the CEO. Moreover, the Cronbach’s alpha indicated that the annual reports and earnings conference calls are not highly correlated (α = 0.30). This means that the average between both is not a very reliable source to use as the independent variable since the alpha was lower than 0.60. For this reason, the annual reports variable and average variable are used in the additional checks to examine if there are measurement errors in my main model. The process of developing the psychopathy dictionary is explained in the subsequent paragraph.

3.4.1 Development of psychopathy dictionary

(24)

24

tendencies (Hare, 1991; Lilienfeld and Andrews, 1996; Babiak and Hare; 2012). Secondly, I have closely examined and analyzed existing literature to find behavior traits that delineates a profile of such an individual and distinguishes them from other individuals. Hereafter, I have conceptualized these behavior traits into LIWC categories to measure the level of psychopathy among CEOs through textual output. The main component of language to identify an individuals’ personality are function words (Fiedler, 2007). Function words include articles, auxiliary verbs, conjunctions, prepositions and pronouns (e.g. I, you, the, and, to, a). These words are used to connect other words and form sentences. They are very hard to control for and the way people use function words reflects their linguistic style.

Firstly, psychopaths are believed to view the world and others instrumentally and theirs for the taking (Porter & Woodworth, 2007). This instrumental view is reflected in their linguistic output through more explanatory and causally framed language. They are thought to use more causation and more subordinating conjunctions (e.g. because, since). In combination with their lack of emotion, subclinical psychopaths are also thought to display an increased psychological

distancing in their written documents and language (Hancock et al., 2013; Sumner et al., 2012).

Their emotional deficit towards others and their callousness makes them able to distance themselves psychologically. Following prior research, individuals with psychopathic traits are expected to use more past tense words, less present tense words and a higher rate of articles (Hancock et al. 2013).

(25)

25

Hancock et al. (2013), also argue that psychopaths are more concerned with basic physiological

needs such as food, money and shelter, and less with higher interpersonal needs like family and

love. For subclinical levels of psychopathy, I expect there for that these individuals are less focused on meaningful relationships and more on achieving their goals and basic needs. The basic needs are captured by the following sets of words in the LIWC dictionary: achievement, biology, health, leisure, money, sex, and work. Where higher interpersonal needs include: family, social, friends, religion, positive emotion, and death (Hancock et al., 2013; 2018).

So, the final independent variable is constructed out of the following behavioral traits and corresponding word categories: instrumental view, psychological distancing, narcissism, and

personal and physiological needs. These traits characterize the psychopathy personality and make

researchers capable to capture this personality through textual output with the LIWC software. Concluding, a higher outcome of this variable indicates a higher tendency towards psychopathy for the specific CEO. The independent variable will be lagged with one year (t-1), as I assume that the effect of CEO psychopathy will affect the innovation performance in the subsequent year. 3.5 Moderating variable

(26)

26 3.6 Control variables

In this study, I control for several individual, firm, and industry-specific control variables. To avoid any confounding variables, these controls are based on prior research towards psychopathy and mainly CEO narcissism.

CEO age and CEO tenure. Building on prior research of Chatterjee and Hambrick (2007), I

control for CEO age and tenure. Younger CEOs are expected to spend more on R&D (Barker and Muller, 2002), while older CEOs tend to follow more conservative and low-growth strategies (Hambrick and Mason, 1984). CEO age is measured as the age of the CEO in the year of observation. Longer-tenured CEOs are expected to prefer a focus on stability and efficiency and show little interest in pursuing risky innovation strategies (Musteen et al., 2010). Moreover, longer-tenured CEOs are able to hire employees or managers who share their views on innovation. CEO tenure is measured years spent in the CEO position.

CEO duality. Following Buyl et al. (2019), I control for CEO duality, which controls for the

situation that a CEO is the chairman of the board as well. CEO duality indicates how much power a CEO has in his organization. CEO duality is measured as a binary variable equaling 0 if the CEO is not chairman and 1 if the CEO is chairman as well.

Firm size. Larger firms are believed to have more (financial) resources to invest in

innovation on the one hand (Teece, 1986), but result in organizational inertia and less flexibility and creativity on the other hand (Cohen and Levin, 1989). Firm size is measured as the natural logarithm of the number of employees, as firms that have more employees are likely to have more resources to spend on innovation.

Firm age. Firm age is to a certain level related to the level of experience and competencies

of the managers and employees, which will positively contribute to the innovative performance (Huergo and Jaumandreu, 2004). Also, it could be argued that older firms have more experience with innovation (Calantone et al., 2002). Firm age is measured in the number of years since the firms is founded.

Firm R&D intensity. Firms that have a higher R&D intensity are proven to be better able to

(27)

27

that are more active in R&D and spent more resources in this field, are more likely to have a higher patenting frequency. Firm R&D intensity is measured as a ratio of R&D investment to the total amount of sales per firm and lagged with one year (t-1), as I assume that the effect of R&D will affect innovation performance in the subsequent year.

Firm performance. Firm performance effects are controlled for since differences in

financial performance can affect the strategic directions of firms. Following prior research (Balking et al., 2000; Chatterjee and Hambrick, 2007), I use the Return on assets (ROA) measure based on the net income of firms, which shows the efficiency of a firm utilizes its current assets. Also, this control for firm performance is lagged with one year (t-1) since I assume that that firm performance affects innovation performance in the subsequent year. Finally, I controlled for year variances by adding a year dummy.

3.7 Analytical Method

To test the relationship of this research, I analyze the data using the statistical software program Stata. In this study, I make use of panel data since I want to observe the influence of CEO psychopathy on firm innovation performance of multiple firms over time. Using panel data helps in eliminating omitted variable biases and heterogeneity (Park, 2011). First, descriptive statistics are conducted to display means, standard deviations, minimum values, maximum values and correlations between the constructs.

(28)

28

with Poisson regression but includes an extra parameter to control for overdispersion (Allison and Waterman, 2002).

(29)

29

4. Results

The following section presents the results of my statistical analysis. First, the descriptive statistics and correlations are described. Then, the regression results regarding my hypotheses are shown and followed by the results of the additional checks.

4.1 Descriptive analysis

Table 1 presents the summary statistics and correlations for all the variables in my model. This

table shows the mean, standard deviation, minimum and maximum values per variable for a sample of 732 observations that are included in the regressions. First, I observe that the sample has a mean of 249.13 granted patents per firm with a maximum of 3210 patents and a standard deviation of 481.35. The count of granted patents has a distribution that is skewed to the left, which indicates that most of the firms only obtain a small number of patents (Figure 1). The statistics show that the level of psychopathy among CEOs has a mean of 0.55 for earnings conference calls, 0.73 for annual reports, and 0.64 for the average between earnings conference calls and annual reports. Further, CEOs are on average 57 years old with a tenure of 6.42 years, firms have an average ROA of 13.41 and spent on average 4 percent of their annual sales on R&D.

Figure 1: Distribution of granted patents

(30)

30

Additionally, I observe some small negative correlations between innovation performance and the level of psychopathy: earnings calls (-0.12), annual reports (-0.13) and the average between annual reports and earnings calls (-0.16). There is a moderate correlation between innovation performance and industry dynamism (0.40), and a relatively high correlation between industry dynamism and R&D intensity per firm (0.68). Lastly, I checked for multicollinearity within my sample with the Variance Inflection Factors test (VIF) (Appendix 3). The values of the VIF test are all below 10.00 and the tolerance numbers are above 0.10, which indicates that I can exclude multicollinearity.

4.2 Regression results

The overview of the results of the random-effects binomial regression is displayed in table 2. This regression is performed to test the two hypotheses in this thesis. To test the hypotheses, I constructed three different models. Model 1 displays the results for the baseline model which includes the innovation performance, control variables, dummy variable, and the industry moderator as a normal control variable. Model 2 adds the level of psychopathy (earnings calls) to the other variables and tests the first hypothesis. The third and last model tests the second hypothesis in which the interaction effect regarding the industry dynamism (high-tech and other industries) is added.

(31)

31 4.3 Additional analyses

In order to check if the non-significant results are not the consequence of a measurement problem or a methodological problem, I performed some additional analyses. First, I ran four additional negative binomial regressions (models 4-7) using CEO psychopathy measured through annual reports and the average between earnings conference calls and annual reports (Appendix 4A). This is done to check if a different measurement for the independent variable results in different outcomes than the main model. Secondly, to check whether I used the right method for my analysis, I ran two additional models (models 8-9) using a different regression model. I performed a random-effects GLS regression with the same variables as the main model (Appendix 4B).

In model 4, I conducted a negative binomial regression between firm innovation performance and CEO psychopathy measured through annual reports and including all control variables. Model 6 runs the same regression, but includes the interaction effect with industry dynamism. Both models are repeated for model 5 and 7 using the average level of psychopathy between earnings conference calls and annual reports as the independent variable. All four models display similar results as the main model in this thesis. Hypothesis 1 is not accepted in model 4 (β = -0.144, p > 0.1) and model 5 (β = 0.297, p > 0.1), and hypothesis 2 gets rejected in model 7 (β = -3.631, p > 0.1). Furthermore, the same control variables stay significant. The only thing that stands out is model 6, which shows a significant opposite result of what I expected in hypothesis 2 (β = -4.974, p ≤ 0.05). So, when using the annual reports for CEO psychopathy, the relationship between CEO psychopathy and innovation performance gets negatively moderated in high dynamic industries. But, as mentioned earlier, the annual reports contain a lot less spoken words by the CEO and are therefore a less reliable source to use for the independent variable.

(32)

32

show much higher coefficients and standard errors compared to the main model. The higher coefficients indicate that the level of psychopathy could explain more of the innovation performance. Moreover, the standard error increases as well and measures how accurate the sample is an estimate of the population (McHugh, 2008). A smaller standard error indicates that the sample is to a higher degree representative for the overall population. So, the high standard error in this model provides little accuracy about the CEO population because it is not a good estimate of the population parameter. Moreover, most of the included control variables become non-significant in the additional regressions. Only the industry control and R&D intensity remain significant in the additional checks.

(33)

Table 1: Descriptive statistics and correlation matrix

Variables Mean Std. Dev. Min Max 1 2 3 4 5 6 7 8 9 10 11 12

1 Innovation performance 249.13 481.35 0.00 3210.00 1.00

2 Level of psychopathy (Earnings calls) 0.55 0.02 0.48 0.72 -0.12*** 1.00

3 Level of psychopathy (Annual reports) 0.73 0.04 0.56 0.86 -0.13*** 0.24*** 1.00

4 Level of psychopathy (Average) 0.64 0.02 0.55 0.72 -0.16*** 0.67*** 0.88*** 1.00

5 Industry dynamism 0.27 0.45 0.00 1.00 0.40*** -0.11*** -0.06 -0.10*** 1.00

6 CEO age 57.27 5.58 29.00 75.00 -0.19*** 0.06 0.15*** 0.14*** -0.17*** 1.00

7 CEO tenure 6.42 5.14 1.00 36.00 0.04 -0.12*** -0.11*** -0.14*** 0.10*** 0.24*** 1.00

8 CEO duality 0.67 0.47 0.00 1.00 -0.09*** -0.01 0.05 0.04 -0.10*** 0.10*** 0.15*** 1.00

9 Firm size (log) 10.75 1.21 6.91 13.38 0.02 0.05 0.03 0.04 0.09 -0.01 0.01 0.00 1.00

10 RD intensity 0.04 0.07 0.00 0.43 0.30*** -0.12*** -0.17*** -0.19*** 0.68*** -0.13*** 0.10*** -0.12*** 0.03 1.00

11 Return on assets (ROA) 13.41 10.38 -116.30 47.10 0.11*** -0.04 -0.09** -0.08** 0.25*** 0.07* 0.00 0.00 0.07* 0.20*** 1.00

12 Firm age 74.29 51.26 1.00 223.00 -0.05 0.11*** 0.15** 0.17*** -0.12*** 0.07* -0.15*** 0.15*** -0.13*** -0.08** 0.04 1.00

Note: Observations = 732. N = 121. Year dummy not reported. *** p<0.01, ** p<0.05, * p<0.1

Table 2: Random Effects Negative Bionominal Regression results Variables

Coeffecient Std. Error Coeffecient Std. Error Coeffecient Std. Error Independent variable

Level of psychopathy (Earnings calls) 0.676 1.273 -0.830 1.657

Moderating variable

Industry dynamism 0.432** 0.188 0.435** 0.187499 -1.608 1.393

Interaction effects

Level of psychopathy (EC) x Industry dynamism 3.678 2.479

Control variables

CEO age 0.013** 0.006 0.012** 0.006 0.013** 0.006

CEO tenure -0.009 0.006 0.009 0.006 -0.010 0.006

CEO duality -0.045 0.062 0.045 0.062 -0.052 0.062

Firm size (log) -0.058** 0.027 -0.060** 0.027 -0.061** 0.027 RD intensity 3.115*** 0.697 3.150*** 0.696 3.276*** 0.684 Return on assets (ROA) 0.023*** 0.004 0.023*** 0.004 0.023*** 0.004 Firm age 0.009*** 0.002 0.009*** 0.002 0.009*** 0.002 Intercept (constant) 0.577 0.478 0.235 0.802 1.040 0.974

Wald X2 262.950 262.670 263.700

Log Likelihood -3287.765 -3287.6247 -3286.5287

Observations 732 732 732

Note: Year dummy is included but not reported. *** p<0.01, ** p<0.05, * p<0.1

Model 1 Model 2 Model 3

(34)

5. Discussion

The main objective of this research is to examine how CEO psychopathy affects firm innovation performance. Building on prior research, I expected CEO psychopathy to have a negative relationship with innovation performance (H1). Additionally, I expected this relationship to be less negative for firms in high dynamic (high-tech) industries (H2). The results of my main analysis do not support these hypotheses, and the additional checks display similar results. Moreover, the additional results do not indicate any measurement or methodological problem. The following section discusses the theoretical and managerial implications of my analysis. Then to conclude this thesis, I provide the limitations and suggestions for future research that derive from my research.

The results of my analysis do not support the first hypothesis. This is a surprising result, as it was expected that CEO psychopathy would be negatively related to firm innovation performance building on prior research. A case study found that an overall decline in innovativeness and creativity was felt under a psychopathic CEO (Boddy, 2017), and the literature review illustrates that several behavioral characteristics of individuals with higher levels of psychopathy are thought to be detrimental for firm innovation performance. Also, the second hypothesis is not supported by the results of my analysis. There is no evidence that the negative relationship between CEO psychopathy and firm innovation performance is mitigated in high dynamic industries. Prior research has not directly examined this moderating factor, but my results do not provide support for the thoughts of other researchers that there are positive attributes residing in the psychopathy personality (Lilenfeld et al., 2012; Boddy et al., 2010a). Besides the evidence that these findings give a call for more research towards the psychopathy personality on CEO levels, I will try to discover possible explanations for my results.

(35)

35

with psychopathic traits are more often found in senior levels of the organization (Boddy et al., 2010a), with approximately 3.5% opposed to 1% in junior levels. However, these percentages are relatively low and the variations in the level of psychopathy between the CEOs in this study are also relatively small. It might be that the negative traits are less profound by individuals who reach the top positions in organizations because there is a selection process inside organizations that obstructs the individuals with high levels of psychopathy to reach the CEO position. Therefore, I do not find significant results in my analysis because the ones who would have a strong negative effect on innovation performance, do not reach the CEO position. It is not that high levels of CEO psychopathy is not related to a lower innovation performance, but the selection process for the CEO position prevents individuals with higher psychopathic tendencies to attain these positions. Second, it could be that there is a methodological explanation for the non-significant results in this study. The measurement that I use to determine the level of psychopathy among CEOs is quite innovative. There is a lot of discussion in the literature on how to measure psychopathy at individuals in corporate settings (Babiak et al., 2010; Mathieu and Babiak, 2016b), since the current measurements suffer from self-report biases and are very much bounded to perceptions of others. Therefore, I have developed a new way to measure psychopathy, to provide a better understanding of psychopathy in corporate settings. The psychopathy dictionary that I constructed for the LIWC software enables me to measure the level of psychopathy through written discourses. However, it could be that this measurement needs to be refined with other kinds of texts or information that reveal more about the personality of CEOs. The earnings conference calls that I use in this study contain a higher amount of spoken words by the CEO opposed to the annual reports, and existing research argues that it is a reliable source to identify certain CEO features through linguistic output (Larcker and Zakolyukina, 2012). Nevertheless, the texts in the earnings conference calls are not 100 percent originating from the CEO and include multiple texts from other individuals. By refining the measurement tool, the psychopathy dictionary could become a more valid measurement for CEO psychopathy and eventually culminate in significant results.

(36)

36

tendencies are more attracted towards firms in high dynamic industries, or that companies appoint CEOs with psychopathic tendencies in periods of successful innovation. Research on the other DT personalities show that CEOs with grandiose narcissism prefer dynamic strategies, extremely risky investments, or rapid changes in strategic initiatives (Chatterjee and Hambrick, 2007; Zhang et al., 2017), which are key characteristics for dynamic environments. Thus, it could be that CEOs with psychopathic traits self-select themselves towards firms in high dynamic industries, or firms tend to choose for CEOs with these characteristics.

5.1 Theoretical implications

This study contributes to the growing body of literature on the psychopathy personality. Particularly, in response to call for additional research towards the DT personalities and leadership (Judge et al., 2009; Boddy et al., 2010a). I contribute to the limited body of research that is available on the psychopathy personality on CEO levels, and examine its effect on firm innovation performance. Moreover, I contribute to the literature by adding an additional variable that allows me to reflect on this relationship, namely industry dynamism. The expected negative effect of CEO psychopathy on innovation performance is not supported with significant results, and the interaction effect of industry dynamism on this relationship lacks significance as well. These results are not in line with prior findings of a case study in the UK, which argues that CEO psychopathy leads to lower firm innovation performance (Boddy, 2017). My results imply that there is no definitive answer that high levels of CEO psychopathy lead to a lower firm innovation performance.

(37)

37

(2010a), who argue that psychopathic individuals are more likely to be found in senior positions of the organization.

Lastly, this study contributes to the literature in a methodological way by providing a new psychopathy dictionary that can be used in the LIWC software. The dictionary serves as an answer in the discussion in psychopathy literature on which assessment tool should be applied to measure psychopathy in corporate settings (Babiak et al., 2010; Mathieu and Babiak, 2016b). After being subject to some refinements, the dictionary can serve as a new methodological foundation for future research to measure psychopathy among individuals.

5.2 Managerial implications

With the results deriving from this research, I aim to provide some valuable insights for managers and executives. First, based on my literature review, prior research indicates that policymakers need to clearly assess the new to be chosen CEO on their level of psychopathy, considering all the negative behavioral attributes that come with individuals with psychopathic tendencies. Many of the daily operations in the workplace could encounter severe difficulties when a CEO with psychopathic traits is in charge. However, my results do not support these assumptions and the psychopathic behavioral attributes are not present in a degree that they lower firm innovation performance. Individuals with high levels of psychopathy are apparently not able to reach the top positions in organizations. Therefore, I advise policymakers and managers to focus on other personality traits and behavioral characteristics that could influence firm innovation performance, instead of the psychopathy personality.

(38)

38 5.3 Limitations and future research:

There are several limitations in this study that limit the generalizability of my findings and should be taken into account for future research. First, the context of the study limits the generalizability since I have only focused on the S&P 100 firms. These firms are considered as the largest firms in the world and are mainly US-based firms. The CEO might be of larger influence when he or she is the leader of a smaller sized company or at the head of a family business. Therefore, future research should examine a sample with firms of different sizes and extent towards other geographical locations.

(39)

39

(40)

40

6. References

Ahuja, G. (2000). Collaboration networks, structural holes, and innovation: A longitudinal study.

Administrative science quarterly, 45(3), 425-455.

Allison, P. D., & Waterman, R. P. (2002). Fixed–effects negative binomial regression models. Sociological methodology, 32(1), 247-265.

Amabile, T. M., Conti, R., Coon, H., Lazenby, J., & Herron, M. (1996). Assessing the work environment for creativity. Academy of management journal, 39(5), 1154-1184. AnnualReports.com. (n.d.). Retrieved from

http://annualreports.com/

Babiak, P. (1995). When psychopaths go to work: A case study of an industrial psychopath.

Applied Psychology: An International Review, 44(2), 171-188.

Babiak, P., & Hare, R. D. (2006). Snakes in suits: When psychopaths go to work. New York, NY: Regan Books.

Babiak, P., Neumann, C. S., & Hare, R. D. (2010). Corporate psychopathy: Talking the walk.

Behavioral Sciences and the Law, 28, 174 –193.

Balkin, D. B., Markman, G. D., & Gomez-Mejia, L. R. (2000). Is CEO pay in high-technology firms related to innovation? Academy of management journal, 43(6), 1118-1129.

Barker, V. L., & Mueller, G. C. (2002). CEO Characteristics and Firm R&D Spending. Management

Science, 48(6), 782-801.

Bena, J., Ferreira, M. A., Matos, P., & Pires, P. (2017). Are foreign investors locusts? The long-term effects of foreign institutional ownership. Journal of Financial Economics, 126(1), 122-146.

Boddy, C. R. P. (2006). The dark side of management decisions: Organisational psychopaths. Management decision, 44(10), 1461-1475.

Boddy, C. R. P. (2011). Corporate psychopaths, bullying and unfair supervision in the workplace. Journal of business ethics, 100(3), 367-379.

Boddy, C. R. P. (2013). Corporate psychopaths: Uncaring citizens, irresponsible leaders. Journal

of Corporate Citizenship, 49, 8-16.

Boddy, C. R. P. (2017). Psychopathic Leadership A Case Study of a Corporate Psychopath CEO.

Journal of Business Ethics, 145(1), 141–156.

(41)

41

Boddy, C. R. P., Ladyshewsky, R. K., & Galvin, P. (2010b). The influence of corporate psychopaths on corporate social responsibility and organizational commitment to employees. Journal

of Business Ethics, 97(1), 1-19.

Boddy, C., Miles, D., Sanyal, C., & Hartog, M. (2015). Extreme managers, extreme workplaces: Capitalism, organizations and corporate psychopaths. Organization, 22(4), 530-551. Brown, S. L., & Eisenhardt, K. M. (1997). The art of continuous change: linking complexity theory

and time-paced evolution in relentlessly shifting organizations. Administrative Science

Quarterly, 42(1), 1-34.

Busenbark, J. R., Krause, R., Boivie, S., & Graffin, S. D. (2016). Toward a configurational

perspective on the CEO: A review and synthesis of the management literature. Journal of

Management, 42(1), 234-268.

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.

Calantone, R. J., Cavusgil, S. T., & Zhao, Y. (2002). Learning orientation, firm innovation capability, and firm performance. Industrial marketing management, 31(6), 515-524. 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.

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.

Chiaburu, D. S., Muñoz, G. J., & Gardner, R. G. (2013). How to spot a careerist early on: Psychopathy and exchange ideology as predictors of careerism. Journal of Business

Ethics, 118(3), 473–486.

Christie, R., & Geis, F. L. (1970). Studies in Machiavellianism. New York, NY: Academic Press. Chung, C. K., & Pennebaker, J. W. (2007). The psychological function of function words. In K. Fiedler (Ed.), Social communication: Frontiers of social psychology (pp. 343–359). New York, NY: Psychology Press.

Clarke, J. (2005). Working with Monsters: How to Identify and Protect Yourself from the

Workplace Psychopath. Random House, Sydney.

Cleckley, H. M. (1988). The mask of sanity. St. Louis, MO: Mosby.

Cohen, W. M., & Levin, R. C. (1989). Empirical studies of innovation and market structure.

Handbook of industrial organization, 2, 1059–1107. New York: Elsevier Science

(42)

42

Cohen, W. M., & Levinthal, D. A. (1990). Absorptive capacity: A new perspective on learning and innovation. Administrative Science Quarterly, 35(1), 128-152.

Crossland, C., & Hambrick, D. C. (2007). How national systems differ in their constraints on corporate executives: A study of CEO effects in three countries. Strategic Management

Journal, 28(8), 767-789.

Dahling, J. J., Whitaker, B. G., & Levy, P. E. (2009). The development and validation of a new Machiavellianism scale. Journal of Management, 35(2), 219–257.

Damanpour, F. (1991). Organizational innovation: A meta-analysis of effects of determinants and moderators. Academy of management journal, 34(3), 555-590.

Deluga, R. J. (2001). American presidential Machiavellianism: Implications for charismatic leadership and rated performance. The Leadership Quarterly, 12(3), 339–363. Deutschman, A. (2005). Is your boss a psychopath. Fast Company, 96(7), 44-52.

Dunlop, P. D., & Lee, K. (2004). Workplace deviance, organizational citizenship behavior, and business unit performance: The bad apples do spoil the whole barrel. Journal of

Organizational Behavior, 25(1), 67–80.

Dutton, K. (2012). The wisdom of psychopaths: What saints, spies, and serial killers can teach us. New York, NY: Farrar, Straus & Giroux.

Finkelstein, S., Cannella, S. F. B., Hambrick, D. C., & Cannella, A. A. (2009). Strategic leadership:

Theory and research on executives, top management teams, and boards. Oxford

University Press, USA.

Francis, J., & Schipper, K. (1999). Have financial statements lost their relevance? Journal of

accounting Research, 37(2), 319-352.

Furnham, A., & Crump, J. (2005). Personality traits, types, and disorders: an examination of the relationship between three self‐report measures. European Journal of Personality, 19(3), 167-184.

Galasso, A., & Simcoe, T. S. (2011). CEO overconfidence and innovation. Management Science, 57(8), 1469–1484.

Hagedoorn, J., & Cloodt, M. (2003). Measuring innovative performance: is there an advantage in using multiple indicators? Research policy, 32(8), 1365-1379.

Hall, B. H., Jaffe, A. B. & Trajtenberg, M. (2001). The NBER patent citation data file: Lessons, insights and methodological tools. NBER Working Paper 8498.

The effect of CEO psychopathy on firm innovation performance