Writing research brief

Published online EarlyView 14 January 2015 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/smj.2348 Received 11 November 2013; Final revision received 17 October 2014

CORPORATE SOCIAL RESPONSIBILITY OR CEO NARCISSISM? CSR MOTIVATIONS AND ORGANIZATIONAL PERFORMANCE

OLEG V. PETRENKO,1 FEDERICO AIME, 1* JASON RIDGE,2

and AARON HILL1 1 Department of Management, Oklahoma State University, Stillwater, Oklahoma, U.S.A. 2 Department of Management, Clemson University, Clemson, South Carolina, U.S.A.

This study builds on insights from both upper echelons and agency perspectives to examine the effects on corporate social responsibility (CSR) practices of CEO’s narcissism. Drawing on prior theory about CEO narcissism, we argue that CSR can be a response to leaders’ personal needs for attention and image reinforcement and hypothesize that CEO narcissism has positive effects on levels and profile of organizational CSR; additionally, CEO narcissism will reduce the effect of CSR on performance. We find support for our ideas with a sample of Fortune 500 CEOs, operationalizing CEO narcissism with a novel media-based measurement technique that uses third-party ratings of CEO characteristics with validated psychometric scales. Copyright © 2015 John Wiley & Sons, Ltd.

INTRODUCTION

Researchers in strategic management, economics, and finance have examined a variety of explanations for managerial decisions to spend organizational time and effort enhancing corporate social responsi- bility (CSR). A considerable emphasis of this litera- ture has been to adopt a stakeholder lens (Donaldson and Preston, 1995; Freeman, 1984) to either norma- tively evaluate the merits of CSR or to instrumen- tally consider how CSR affects organizational finan- cial performance (e.g., McWilliams and Siegel, 2000; Ramchander, Schwebach, and Staking, 2012; Waddock and Graves, 1997; Wright and Ferris, 1997). More recently, researchers have focused more directly on studying the determinants of CSR.

Keywords: upper echelons; CEO narcissism; corporate social responsibility; organizational performance; strategic decisions *Correspondence to: Federico Aime, 205 Business Building, Stillwater, OK, 74078, U.S.A. E-mail: aime@okstate.edu

These researchers have looked at both external drivers like, for example, the salience of exter- nal stakeholders (Agle, Mitchell, and Sonnenfeld, 1999), stakeholder activism (Clark and Hebb, 2004; David, Bloom, and Hillman, 2007; Marquis, Glynn, and Davis, 2007; Sen, Gurhan-Canli, and Morwitz, 2001), or institutional pressures (Neubaum and Zahra, 2006) and internal drivers like, for example, executive incentives (e.g., Deckop, Merriman, and Gupta, 2006; McGuire, Dow, and Argheyd, 2003), management team commitment to ethics (Muller and Kolk, 2010), and CEO political ideologies (Chin, Hambrick, and Treviño, 2013) as antecedents to CSR. Among this research on determinants of CSRs, the vast majority of the studies have explored the effects of external normative values (e.g., the ethical concerns of particular stakeholder groups or of the institutional environment) on CSRs owing to the fact that these reflect a form of alignment of firm policy to its stakeholder value preferences; fewer have examined the effects of internal values at the organizational or individual levels (e.g., the

This research is focused on the driving forces of CRS - that is both external stakeholders and stake holders activism

the determinant of CRS includes a variety of studies in other to explore the effect of external normatives, ethical concerns in a stakeholder group or institution. For the fact that CRS reflect in the alignment of firm policies and procedures its stakeholders value preferences can be fully examined by the effect of internal value or individual ethical concerns

and also internal driving forces

Corporate Social Responsibility or CEO Narcissism? 263

ethical concerns or the political ideology of top management) on CSR; and very few have stud- ied the effects of psychological characteristics of individuals on CSR (e.g., Aguilera et al., 2007). This emphasis of values over other psychological characteristics as explanations for CSR is proba- bly expected given that CSR is a value loaded con- cept defined as “actions that appear to further some social good, beyond the interests of the firm and that which is required by law” (McWilliams and Siegel, 2001: 117). But the scarcity of research relating executives’ psychological characteristics to CSR is striking given the emphasis of upper echelons work on the effects of executive personality characteris- tics on firm strategic decisions.

The present study shifts the attention more squarely to executive psychology as an explanation for CSR, showing that CSR initiatives may result from leaders’ personal needs for attention and image reinforcement and how such initiatives may be less strategic in terms of financial performance and focus for their organizations. Researchers in management have shown that the characteristics of top executives, and especially CEOs, affect organizational decisions and behaviors (Chatterjee and Hambrick, 2007; Finkelstein and Hambrick, 1996; Hambrick, 2007; Hambrick and Mason, 1984; Sanders, 2001b). Intangible decisions like involvement in CSR present a considerable opportunity for these decisions to be affected by executive’s characteristics because such actions rarely present a situation where probabilities can be easily calculated or outcomes be certain. In line with this logic, Weidenbaum and Jensen (2009: xi) challenge readers to “merely consider the ability of the CEO of a major company to satisfy his whim in terms of the selection of charities and pet causes that the organization will support,” pointing to the risk of CSR choices being more closely tied to personal drivers than to organizational stakeholder logic. We argue in this paper that organizations with CEOs that have a high need for attention and are preoccupied with having their positive self-views reinforced will engage in higher levels of corporate social responsibility.

Specifically we are concerned in this paper with how CEO narcissism may affect organizational CSR because narcissistic CEOs have a high need for attention and praise as well as a strong desire to have their positive self-views reinforced, which has been shown to affect CEO decision making (Chat- terjee and Hambrick, 2007; Gerstner et al., 2013).

Narcissists seek to broadly and constantly generate what Kernberg (1975) called “narcissistic supply”, the reinforcement to self-image derived from per- sonal exhibitionism or from external adulation and flattery (Bogart, Benotsch, and Pavlovic, 2004; Wal- lace and Baumeister, 2002). Since CSR activities offer a difficult to assess opportunity for attracting observers’ attention, gaining praise from internal (e.g., employees) and external (e.g., media) stake- holders, and achieving notoriety, we expect more narcissistic CEOs — defined as CEOs who have inflated views of themselves and who seek to have those positive self-views continuously reinforced (Campbell, Goodie, and Foster, 2004; Chatterjee and Hambrick, 2007) — to engage their organiza- tions in more CSR initiatives as a way to enhance their moral feelings of superiority and to attract attention and praise. In essence, for example, an executive’s decision to spend organizational efforts to enhance CSR can have significant benefits for the executive such as attention in the media and positive praise from employees and the community. Just as celebrities generate attention for themselves by participating in charitable events, chief execu- tives can make themselves noticeable through their firms’ CSR activities. Journalists and other organi- zational stakeholders frequently comment that most organizations make a show of embracing corpo- rate responsibility activities (e.g., Lewis, 2008) and CEOs are often personally positioned at the center of praise or criticism when it comes to the social behaviors of their organizations. Yet, despite their high potential contribution to the CEOs’ narcis- sistic supply, CEOs’ personal needs for attention and image reinforcement have been largely absent from CSR research. In line with this logic, we first hypothesize and find that organizations controlled by narcissistic CEOs will engage in more CSR. In line with this same logic, we also look at spe- cific actions that receive media acclaim and show that narcissistic CEOs are more likely to seek such recognition from the media through their organiza- tions’ philanthropic actions.

Finally, we address the strategic implications of the influence of CEOs’ personal interests on CSR. We specifically show that CEO narcissism will negatively moderate the relationship between CSR and firm performance. This last finding is important for the discussion of the relationship between CSR and firm performance because it highlights a CSR cost that may not even be values driven, like in Friedman’s depiction of CSRs as “spending

The concerns of this article is much concern about some behavior of CEO that might have a negative feedback on CRS. Because since Narcissism are the people that has a strong desire for self praise this might affect the rules of engagement.

Considering the ability of the CEO to make major decisions to satisfy the company in terms of the selection of charities and causes that the company will support will show the risk involved and this can be closely related to personal drive than organistional stakeholders logic. This article revealed that CEOs with high need of attention can be preoccupied with having their positive self views and this will result to higher level of corporate social responsibility.

Since Narcisissist has the tendencies to generate narcissistic supply, with the feel of reinforcing self image derived from personal exhibitionism or flattery. Since CRS can attract praise from both internal and external shakeholders, in other to notoriety, as a result Narcissist CEO tend to see CRS as a way to inflate their views and also have those reinforced views constantly.

this article draws our attention to executive psychology while talking about CRS because CRS initiative results may be yield from leaders, with the need for attention to portray their company in a way to reinforce such initiative. they many have less strategic in financial performance for their organization. and it is clear that the character of top executive can actually determine the organization behavior and decisions

in organisations it is more easy for narcissist CEO to seek recognition from the media through their company philanthropic actions

264 O. V. Petrenko et al.

someone else’s money for a general social interest” (Friedman, 1970: 174), but more explicitly within the realm of residual agency cost. The potential for leaders to make CSR choices to fulfill their personal needs may put at issue the idea that while CSR can certainly be a valuable strategic choice for firms, they can also be nonstrategic when driven by managerial self-interest. Given the complex nature of the relationship between CSR and financial performance, with about half of the studies on the relationship between CSR and financial perfor- mance reporting positive relationships and the rest of the studies suggesting mixed or nonsignificant findings (Chin et al., 2013; Margolis and Walsh, 2003), it is important to examine determinants of CSR that may be not strategic as in the case of managers acting for their own interest.

To test our ideas, we conducted an empirical study Fortune 500 CEOs over a 10-year period. We used the Kinder, Lyndenberg, and Domini (KLD) CSR database for our measurement of corporate social responsibility and COMPUSTAT for our measurement of accounting financial performance and a wide array of control variables. To avoid the limitations of proxy-based measures for assessing psychological constructs (Carpenter, Geletkanycz, and Sanders, 2004), we utilize a new approach to the measurement of CEO characteristics. Specifi- cally, based on a wide array of psychology research showing the value and accuracy of third-party rat- ings (e.g., Arthur et al., 2003; Connelly and Ones, 2010; Oh, Wang, and Mount, 2011), we develop a video-based measurement of CEO characteristics that allows us to evaluate CEOs on narcissism with a validated narcissism scale (Svindseth et al., 2008). We find substantial support for our hypotheses.

This study makes several contributions to research. First, it contributes to upper echelons research on CEO narcissism by showing that CSR is an especially pertinent domain to generate “narcissistic supply”. Second, it contributes to methods in upper echelons research by utilizing a new, theory-based method to measure CEO charac- teristics in an unobtrusive way. As explained later in the manuscript, the video-based psychometric approach to the measurement of narcissism that we utilize in this paper can provide a significant new tool to the upper echelons literature for the measurement of a wide array of personal charac- teristics of leaders that were previously difficult to assess consistently for strategy researchers. Third, this research also contributes to agency

theorizing. Drawing on upper echelons theory we find that narcissism may result in agentic behaviors that may have negative financial implications for organizations. Finally, we contribute to research on corporate social responsibility. We introduce narcissism as a novel determinant of CSR and suggest that motivations for CSR that are not strategic, like CEO narcissism, may help explain the mixed findings in the relationship between CSR and firm performance.

THEORY AND HYPOTHESES

Narcissism and corporate social responsibility

The concept of CSR refers to voluntary manage- rial “actions that appear to further some social good, beyond the interests of the firm and that which is required by law” (McWilliams and Siegel, 2001: 117; see also Waddock, 2004). When man- agers look beyond shareholders and decide to exert organizational effort on employees, customers, the environment, and other stakeholders, they have con- siderable discretion as such decisions are difficult to evaluate (Margolis and Walsh, 2003; Waldman and Siegel, 2008). Building on the logic of the upper echelons perspective, CEOs will have a significant influence in such discretionary decisions and there- fore a firm’s propensity to engage in CSR may be affected by chief executives’ preferences and pri- orities that derive from their values and personali- ties (Chatterjee and Hambrick, 2007; Gerstner et al., 2013; Hambrick and Mason, 1984).

Because top executives, and especially CEOs, can affect the behaviors and outcomes of firms and have substantial influence on organizational efforts and outlays (Chandler, 1962; Finkelstein and Hambrick, 1996), researchers have explored the psychological qualities and experiences of top executives to understand organizational behaviors and outcomes. For example, research has shown that top executives’ personality (Kets de Vries and Miller, 1985), charisma (Flynn and Staw, 2004), and locus of control (Miller, Kets de Vries, and Toulouse, 1982) affect organizational outcomes. More recently, Chatterjee and Hambrick (2007, 2011) have shown the impact of CEO’s narcis- sism on firm strategies. They found that narcissism predicts the dynamism and grandiosity of firms’ strategic actions and that narcissistic CEOs react differently to their success and seek social praise (Chatterjee and Hambrick, 2011).

This shows that there are tendencies for CEO to use CRS as a means to fulfil their personal needs. and it this is driven by managerial self interest it may hamper the policies of CSR.

Therefore Narcissism can result in agentic behaviors that can hinder the and create negative implication on the organization

Corporate Social Responsibility or CEO Narcissism? 265

Narcissistic individuals construe reality in part as it reflects on their self-image and are constantly seeking attention and reinforcement of their positive self-views (Campbell et al., 2004; Chatterjee and Hambrick, 2007, 2011). An organization’s social responsibility projects and overall external percep- tions often bring attention in the form of praise or criticism not just to the firm, but often more directly to the CEO. As such, these projects are likely to engage aspects of the CEO’s self-image. There are at least three reasons to view CSR initiatives as providing an especially relevant domain to gener- ate narcissistic supply for narcissistic CEOs. First, CSR are value loaded initiatives that appear to fur- ther some social good and offer an opportunity for exhibitionism (Bogart et al., 2004) by taking the moral high ground on socially acceptable behaviors. Second, CSR engages sets of value sensitive audi- ences in adulation, media attention, and praise, all of which are external sources of narcissistic sup- ply (Wallace and Baumeister, 2002). Finally, CSR offers a variety of avenues to change the status quo supplying continuity and variety to the opportuni- ties narcissistic CEOs have to exhibit themselves to attentive and responsive audiences.

Additionally, while most CEOs may want to avoid criticism for negative CSR, narcissistic CEOs will be more strongly motivated to abstain from socially irresponsible actions that can bring easy criticism from investors, media, the public or employees because narcissists respond particularly negatively to criticism (Kernis and Sun, 1994; Rhodewalt and Morf, 1998; Rhodewalt and Sorrow, 2003) and therefore avoid it. Social responsibility can be viewed as a continuum with extremes being “socially responsible” and “socially irresponsible.” Because discrimination, lawsuits, scandals, and similar CSR concerns often bring about serious criticism of the CEO, narcissistic CEOs are likely to avoid engaging in negative CSR. This logic may seem at odds with the idea that narcissists will seek attention at any cost, even when such attention implies negative attention. For example, narcis- sistic CEOs have been found to favor acquisitions because, even when acquisitions “do not always garner positive acclaim (Shleifer and Vishny, 1991; Sirower, 1994), they are highly visible, attract the audience that is needed by the narcissistic CEO” (Chatterjee and Hambrick, 2007: 359). But avoid- ing negative CSR or, more specifically, eliminating CSR concerns, just like allowing CSR concerns to emerge, is subject to media attention, but in

a positive tone and with much less institutional cost. Consistently, narcissistic CEOs are likely to seek positive attention by avoiding the emergence of CSR concerns and working to reduce existing CSR concerns. Therefore, we expect narcissistic CEOs to approach or be motivated to engage in positive CSR initiatives and to avoid negative CSR developments in their organizations, creating a positive relationship between CEO narcissism and organizational CSR.

Hypothesis 1: There will be a positive relation- ship between CEO narcissism and corporate social responsibility.

Narcissism and high profile corporate philanthropy

As we previously noted, narcissists are exception- ally susceptible to adulation, media attention, and praise, all of which are external sources of narcis- sistic supply (Wallace and Baumeister, 2002). This need for adulation and praise implies that narcissists are more likely to behave as exhibitionists (Raskin and Terry, 1988) and are preoccupied with receiv- ing attention from others (Kernberg, 1975, 1986; Kohut, 1977). Therefore, narcissistic CEOs can be expected to favor especially visible initiatives that reflect on themselves, like for example philan- thropic initiatives that attract media attention and praise because, even when spending organizational effort in such actions may draw some criticism from sectors of the investor community and media, nar- cissistic CEOs are “exceptionally emboldened by social praise” (Chatterjee and Hambrick, 2011: 202) and will engage in more bold and noticeable actions than less narcissistic CEOs. High profile philan- thropic actions are a noticeable example of actions that will attract both praise and attention for the CEO, because they are usually both construed as contributing to a social good and reflective of the CEOs values or preferences. As such, we expect a positive relationship between CEO narcissism and the high profile of their corporate philanthropy. We focus here on predictions about the media profile of the philanthropic actions by firms because, if nar- cissists are likely to engage in CSR in order to gen- erate social approval as we argue in our previous hypothesis, predicting differences in the corporate philanthropy media profile of firms run by more narcissistic CEOs provides additional evidence for the mechanisms in our theory.

266 O. V. Petrenko et al.

Additionally, two aspects of narcissistic person- ality profiles suggest that narcissistic CEOs will not only engage their organizations in higher pro- file philanthropic initiatives, but will also be driven to continue such initiatives that reward them with media or analyst praise. First, narcissists have a ten- dency to be restless (Deluga, 1997), to constantly seek attention and reinforcement of their positive self-views (Campbell et al., 2004; Chatterjee and Hambrick, 2007, 2011), and to maintain protag- onism (Bogart et al., 2004; Morf and Rhodewalt, 2001). Second, research on narcissists’ responses to environmental feedback (Kernis and Sun, 1994; Rhodewalt and Eddings, 2002) suggests that narcis- sists will be particularly responsive to visible social praise and attention (Vazire and Funder, 2006) and respond to praise with additional effort (Chatter- jee and Hambrick, 2011; Wallace and Baumeis- ter, 2002). Narcissism and the reinforcement of self-images resulting from the high profile of their corporate philanthropy will provide an avenue for narcissistic CEOs to maintain protagonism. While we expect that all CEOs are going to continue to commit to actions that have generated positive media attention for them, we expect that narcissistic CEOs will be more responsive to the media atten- tion resulting from such actions (Chatterjee and Hambrick, 2011). Implicit in this hypothesis is that we expect both narcissism and previous high pro- file corporate philanthropy to have positive direct and interactive effects on later high profile corporate philanthropy. Therefore, we hypothesize that more narcissistic CEOs will be more likely to continue to generate higher profile corporate philanthropy than their less narcissistic peers:

Hypothesis 2: CEO narcissism will positively moderate the relationship between prior corpo- rate philanthropy media profile and current cor- porate philanthropy media profile.

Narcissistic CSR and financial performance

A dominant concern of the CSR literature has been the concern with the instrumental relation- ship between CSR and firm financial performance. Researchers have argued that CSR can serve as strategic choices that may lead to firm perfor- mance because, by focusing on stakeholders other than the shareholders, CSR reduces the cost of committing resources to the organization (Hill- man and Keim, 2001; Jones, 1995; Turban and

Greening, 1997), reduces risk premiums, (Cornell and Shapiro, 1987), provides insurance protection against litigation and regulation costs (Kacper- czyk, 2009), serves as advertisement and goodwill (Knauer, 1994), and enhances corporate reputation (Schnietz and Epstein, 2005). With findings about the relationship between CSR and firm performance generally positive but mixed, the question remains: if CSR choices are made in part based on CEO char- acteristics, or more specifically to satisfy narcissis- tic CEOs needs for praise and attention, will they be less likely to relate to financial performance?

We expect CEO narcissism to moderate the pos- itive relationship between CSR and firm financial performance for three main reasons. First, CSR decisions by narcissistic CEOs are less likely to be aligned with consideration of organizational out- comes, and therefore CEO narcissism will mod- erate the relationship between CSR by their firms and organizational performance. Narcissistic CEOs consider themselves highly intelligent and superior in their ability to control the environment (Campbell et al., 2004; Judge, LePine, and Rich, 2006; Pratto et al., 1994) and are much less responsive to objec- tive performance indicators compared to their less narcissistic peers (Chatterjee and Hambrick, 2011). As such, narcissistic CEOs’ strong confidence in their ability to control the environment and their need to attract attention and praise are more likely to result in CSR decisions that disregard the effects on stakeholders beyond simple praise and enhance their own reputation above the organizational rep- utation, therefore leading to CSR efforts with less positive effects on firm performance.

Second, because CSRs can have negligible or even negative effects on performance in the absence of absorptive capacity or complementary assets (Darnall and Edwards, 2006; Zahra and George, 2002), CSR by firms with more narcissistic CEOs who are more likely to disregard the availability of complimentary resources in the organization because of their lower responsiveness to indicators and strong sense of personal ability to control their environment (Campbell et al., 2004) should have a less positive relationship with organizational performance.

Finally, the tendency of narcissists to be rest- less (Deluga, 1997), to constantly seek attention and reinforcement of their positive self-views (Camp- bell et al., 2004; Chatterjee and Hambrick, 2007, 2011), and to maintain protagonism (Bogart et al., 2004; Morf and Rhodewalt, 2001), will be reflected

Corporate Social Responsibility or CEO Narcissism? 267

in their engagement in multiple and widely varied CSR initiatives with different sets of stakeholders. This will result in a widely unconcentrated CSR effort that will aid them in avoiding feelings of bore- dom (Wink and Donahue, 1997) and satisfy sensa- tion seeking (Emmons, 1981) in their contact and interaction with multiple constituencies. This ten- dency of narcissistic CEOs to be broad and scattered in their decisions affecting CSR leads the situation when the CSR efforts by firms with more narcis- sistic CEOs may be seen by stakeholders as less regular or ad-hoc (Husted and Salazar, 2006; Vergne and Durand, 2010), or as simple responses to exter- nal pressure (Frooman, 1999) reducing the credibil- ity of a serious commitment to stakeholder interests and therefore reducing its impact on firm financial performance.

Therefore, CSR by organizations with highly nar- cissistic CEOs may not have as positive or efficient effects on the cost of committing resources to the organization, risk premiums, and corporate reputa- tion, as those by organizations with less narcissis- tic CEOs. CEO narcissism may negate or diminish the mechanisms that link CSR decisions with orga- nizational outcomes and provide one explanation for some of the mixed findings in some previous research. We argue that the level of CSR of organi- zations with more narcissistic CEOs is less likely to be positively related to financial performance than CSR of organizations with less narcissistic CEOs.

Hypothesis 3: CEO narcissism will negatively moderate the relationship between CSR and performance.

METHODS

Sample and data collection

Our annual financial and corporate data come from Standard and Poor’s COMPUSTAT industrial databases, our corporate social responsibility data come from the KLD database (Godfrey, Merrill, and Hansen, 2009; Orlitzky and Benjamin, 2001), and our CEO characteristics data was collected with a novel video survey methodology. Our starting pop- ulation included all S&P 500 firms between the years 1997 and 2012 inclusive, and excluded 24 pri- vate firms for which no financial data is available in COMPUSTAT. Following prior research, we then excluded 69 firms in highly regulated industries

such as financial, insurance, and utilities. Firms in highly regulated industries such as financials are not only subject to differences in their regulatory environments that limit discretion of these firms’ CEO over outlays such as CSR (McNamara, Aime, and Vaaler, 2005; Sanders, 2001a) but also their results are not comparable with those of other indus- tries based on differences in accounting criteria (McGahan and Porter, 1997). To verify these the- oretical rationales, we run K-S tests to verify if the distributions of CSR and performance variables are different for regulated firms and especially financial firms compared to others in the population. There were significant differences with respect to compar- ison variables; p-values for ROA and CSR measures ranged from 0.000 to 0.01 for both financials specif- ically and regulated industries as a whole when compared with the broader population of firms. We identified the CEO for every firm in 2007 and included all firm-years in this time-frame for which they were CEOs of their respective firms. We then imposed four necessary filters on our data. First, we omitted 15 CEOs who held temporary appointments (e.g., interim, acting) because the effects on firms of temporary CEOs are different to those of per- manent CEOs (Ballinger and Marcel, 2010). Sec- ond, we excluded 16 CEOs that were with their companies for only one year. Third, we omitted 139 CEOs because adequate videos (as discussed in the Measurement of narcissism section of the paper) were unavailable through public sources. Finally, we excluded 86 firms for which data was not avail- able in the KLD database. The final sample repre- sented in our models ranges between a total of 911 and 1,051 CEO-year observations based on con- trol and year lag availabilities, for which financials and corporate social responsibility were measured annually for each CEO-year available in the sample. We assessed the representativeness of the sample by comparing included and nonincluded firms/CEOs in the final sample using the Kolmogorov-Smirnov (K-S) two-sample test (e.g., Siegel and Castellan, 1988; Westphal and Bednar, 2005). K-S tests if the distribution of given variables are different for firms/CEOs in the final sample compared to others in the broader population. There were no signifi- cant differences with respect to variables included in the study that were available for the broader pop- ulation, including measures of CSRs, CEO duality, CEO age, CEO ownership, CEO tenure, short and long term pay, ROA, TQ, and Market Value Added); p-values ranged from 0.19 to 0.65.

268 O. V. Petrenko et al.

CEO narcissism was measured as invariant in line with previous research that has viewed nar- cissism as a stable disposition (Campbell, Foster, and Finkel, 2002; Chatterjee and Hambrick, 2007, 2011; Cramer, 1998). We also tested our models with alternative sample configurations with identi- cal results. We proceeded to create a sample includ- ing all firm-year observations from 2007 to 2012. By testing our findings in this alternative sample, we checked for robustness of our results to tem- poral precedence of the measurement of narcis- sism. While narcissism is considered an invariant in research, we wanted to confirm that results would not differ between an invariant measurement with strict temporal precedent and an invariant measure- ment without temporal precedent. Our results are robust to alternative measurement periods.

Independent variables

Measurement of narcissism

We follow the prevailing instrument for measur- ing narcissism, the Narcissistic Personality Inven- tory (NPI), through third-party ratings of video samples of CEOs. Third-party ratings have been meta-analytically shown to provide higher oper- ational validities of personality traits when com- pared to self-reports (Oh et al., 2011) and do not suffer from the inflation of self-reports (Van Iddekinge, Raymark, and Roth, 2005) because observers have “clearer lenses” for identifying tar- gets’ personality traits (Connelly and Hulsheger, 2012). Utilizing third-party ratings of video sam- ples provides valid access to direct but unobtrusive measurement of CEO characteristics. This novel videometric approach allows us to overcome several limitations in the measurement of CEO characteris- tics. First it provides unobtrusive but direct access to a large sample of CEOs because video samples of CEOs in the public domain are already ubiquitous online and are likely to increase in availability and quality as the presence of video documentation con- tinues to grow for all aspects of social and corporate life. Therefore, it finds a way around the reluctance of top executives to participate in survey research (Chatterjee and Hambrick, 2007, 2011) while pro- viding a direct physical trace or documentary sam- ple of them for measurement (Webb et al., 1966) that is not mediated by other participants like in the cases in which investment relations departments prepare annual reports or other company public

statements. Second, it provides the opportunity to measure the sample with previously validated psy- chometric scales (like, in this case, the NPI) without concerns about loss of responses based on the sen- sitivity of the traits being measured (Cycyota and Harrison, 2006) or about social desirability biases.

Several third-party ratings of individual char- acteristics have recently been used by researchers to assess, for example, personality, competencies, likeability, and expected election performances (Benjamin and Shapiro, 2009; Borkenau and Liebler, 1993; Judge et al., 2002; Mount, Bar- rick, and Strauss, 1994; Oh et al., 2011; Raskin, Novacek, and Hogan, 1991; Riemann, Angleitner, and Strelau, 1997; Rubenzer, Faschingbauer, and Ones, 2000; Zimmerman, Triana, and Barrick, 2010), including media supported observations (Benjamin and Shapiro, 2009; Borkenau et al., 2009). To the best of our knowledge, our study is the first to utilize this videometric approach to measure CEO characteristics. This new approach provides a novel and valid access to an otherwise potentially biased and rarely accessible population.

Videos showcasing focal chief executive offi- cers in our population of interest in 2007 were acquired from public internet sources and edited to omit identifying information that could bias evaluations by coders, including their position and the name of their company. Doctoral students in psychology with experience in personality assess- ment were recruited to serve as raters and were offered a monetary incentive to participate in the study. To assess CEO narcissism, we turned to the NPI (Raskin and Terry, 1988; Raskin et al., 1991; Rhodewalt and Morf, 1995), the prevail- ing instrument for measuring narcissism (Chatter- jee and Hambrick, 2007, 2011), by employing a short third-party rater adapted variation on Kansi (2003). Three expert raters, blind to the study hypotheses, rated each focal CEO on the narcis- sism items using a seven-point Likert scale. This scale demonstrated high coefficient alpha relia- bility (𝛼 = 0.95) (Nunnally, 1978). Moreover, the expert raters demonstrated significant agreement on their ratings of CEO narcissism, ICC (1, 3) = 0.54, p < 0.001, rwg = 0.85 (Bliese, 2000).

We developed the samples and procedures for rating and trained the expert raters on the rating scale and procedures prior to executing the rat- ings. First, we conducted a pilot study to estab- lish acceptable video lengths in order to have suit- able video length for valid measurement while

Corporate Social Responsibility or CEO Narcissism? 269

avoiding rater exhaustion. More specifically, we had different sets of raters observe and rate videos of CEOs of different lengths including: 45 seconds, 1 minute and 20 seconds, 2 minutes and 30 seconds, 5 minutes, and 10 minutes. ICC (1,3) for narcissism is negative for 45 second video samples (−0.66), turns positive for 1 minute and 20 second video samples (0.45) and grows significantly to 0.71 with 2.5 minute video samples. It does not signifi- cantly change for lengthier video samples (0.72 for both 5 and 10 minute video samples). We further compared means for narcissism with video sam- ples of different lengths and the results showed that means were significantly different between 45 seconds and 2.5 minutes (p < 0.01) but there was no significant change in the means between sam- ples of 2.5 minutes and samples of 5 and 10 minutes (p > 0.65 and p > 0.87 respectively). The video sam- pling design was developed following these results by creating video samples of 2 hours and 30 minutes in length plus/minus a few seconds for sentence completion in the sample.

Raters were notified in shared training sessions that the focus of the study was to get their percep- tions regarding the individuals in the video sam- ples, that there was no alternative interest in their responses, and they were not being evaluated in any way. Raters were then logged into a training video sample survey in Qualtrics and were able to code three video examples and the observa- tions were discussed with researchers as part of the training. After the training, raters were provided an individual login code to access the Qualtrics embedded video sample surveys. There was random assignment of CEO to raters and all ratings were performed in individual sessions across two weeks with sessions limited to not more than an hour per session to avoid exhaustion and guarantee rating independence.

To ensure validity of our measures we have con- ducted a number of robustness checks. First, in order to ensure our ratings were robust to media effects or timing of the video sample, we cre- ated a random subsample of CEOs (n = 32) for which we collected multiple available video sam- ples from public sources. We rated these videos in the same manner as the core sample videos and pro- ceeded to analyze mean differences between dif- ferent videos for the same CEO. There were no significant differences in the narcissism measure (p < 0.43) between different videos of the same

CEOs, showing that the approach is generally con- sistent across video samples. As a final robustness check, we assessed the consistency of our video sampling rating approach and both self-ratings and personal acquaintance ratings of narcissism for a set of individuals. Specifically, we recruited 10 doc- toral students and video recorded their responses to a set of open questions and edited the length of the videos per our study’s design. We then col- lected the narcissism scale through self-reports in a survey of the video sampled individuals, a rating of narcissism for the individuals by two direct per- sonal acquaintances of the subjects, and our video sampling procedure with three expert raters with no prior relationship to the focal subjects of this robustness test. We found no significant differences across the three measurement approaches showing consistency between our approach and other survey procedures. As a comparison to previous research that has utilized unobtrusive measures of narcis- sism based on indicators of narcissistic tendencies, we compared videometric approach to Chatterjee and Hambrick’s (2011) measure of CEO narcissism that combines indicators for (1) measures of relative pay, (2) prominence of CEO’s photograph in annual reports, and (3) prominence of CEO in company press releases into an index of CEO narcissism. The correlation between the measures was high and significant (0.404, p < 0.001). Our measure of narcissism is available by request.

Dependent variables

Corporate social responsibility (CSR)

We used KLD ratings to measure corporate social responsibility with data from Kinder, Lyndenberg, Domini, and Company (KLD), a financial advisory firm with a focus on Corporate Social Responsi- bility evaluations (Godfrey et al., 2009; Mattingly and Berman, 2006; Waddock and Graves, 1997) that has been broadly regarded as the most com- prehensive data available to measure CSR (e.g., Choi and Wang, 2009; Graves and Waddock, 1994; Kacperczyk, 2009). While KLD data has limitations related to potential confusion of industry effects (Rowley and Berman, 2000) and possible subjec- tivity (Entine, 2003), it has generally been shown to have good empirical reliability (Walls, Berrone, and Phan, 2012) and has widely been seen as less sub- jective and more representative of the actual con- struct than alternative measures and therefore has

270 O. V. Petrenko et al.

been more widely adopted in empirical research (Choi and Wang, 2009; Coombs and Gilley, 2005; Dahlmann and Brammer, 2011; Hillman and Keim, 2001; Hull and Rothenberg, 2008; Kacperczyk, 2009; Wong, Ormiston, and Tetlock, 2011). KLD data is based on ratings by independent analysts of a variety of categories of CSR characteristics of firms including, for example, community, diver- sity, employee relations, environment, and human rights. We operationalize our measure of CSR as an aggregate net score at t + 1 of the various dimen- sions reported in the data following the most com- monly utilized approach in the literature (Choi and Wang, 2009; Dahlmann and Brammer, 2011; David et al., 2007; Graves and Waddock, 1994; Hillman and Keim, 2001; Hull and Rothenberg, 2008; Wong et al., 2011).1

Corporate philanthropy media profile (CPMP)

Media praise for CEOs following company philan- thropic actions was assessed by a content analysis of key publications that cover general and busi- ness news through targeted searches in the Fac- tiva database of media coverage. We first searched Factiva for all articles that included the company name and the firm/year combination in our sample. Then, we isolated articles that cover philanthropic events using the predefined Philanthropy reputation driver of the Factivia Expert Search module. This search option employs a predetermined query which includes relevant keywords associated with philan- thropy (e.g. philanthropy, donation, charities) and their derivatives included in a dictionary of expres- sions that identifies media publications which cover philanthropic actions by corporations. Two inde- pendent raters then coded the number of articles that positively mention the CEO associated with philan- thropic events for each firm year observation. Inter- rater agreement was high (ICC1 = 0.78). To further ensure validity of our measure, we resolved each instance of disagreement between the coders by independently conducting a separate content anal- ysis to resolve the disagreement. To account for the possibility that certain industries attract more media attention we standardized the variable by industry (two-digit SIC code).

1 As a robustness check, we also assess CSR using a weighted indicies (Waddock and Graves, 1997) and find results consistent across both measures.

Performance

We examine performance using Return on Assets (ROA), a common measure of firm performance, calculated as net income divided by assets at t (e.g., Finkelstein and Boyd, 1998; Ridge, Aime, and White, 2014; Schmalensee, 1985). Given the nature of our hypothesized effects and arguments, a widely accepted measure of operational performance cap- tures the expected effects on the performance of firm operations caused by CSR decisions made by firms in the study.

Additionally, we included two common measures of market performance for supplementary analysis: Tobin’s Q (TQ), calculated by dividing the firm’s market value by firm’s asset replacement costs and Market Value Added (MVA), calculated by subtracting capital (i.e., the debt and shareholders’ equity invested in the firm) from the equity market valuation of the firm. Tobin’s Q provides an approx- imation of the stock market’s estimation of net present value (Tobin and Barnard, 1968) and MVA captures the ability of firms to maximize share- holder value through efficient allocation and man- agement of resources (Hillman and Keim, 2001).

Control variables

We control for CEO-, firm-, and industry-level potential confounding factors.

CEO control variables

Because views about the importance of corporate social responsibility may vary with age, we con- trolled for CEO age. We also controlled for indi- cators of CEO structural power (Finkelstein, 1992) that might influence their ability to promote CSR projects in their firms, including CEO tenure, dual- ity (coded as a 1 if CEO is also a chairman of the board), the percentage of company stock owned by the CEO, and the level of independence of the board. Additionally, we controlled for CEO incen- tives: short term pay focus was measured as the ratio of the dollar value of bonuses earned by the execu- tive during the year and the total value of all CEO compensation; long term pay focus was calculated as the ratio of the dollar value of restricted stock and stock options to the total compensation (Deckop et al., 2006). Finally, we controlled for CEO polit- ical ideology in order to capture CEO preferences about CSR (Chin et al., 2013). To measure CEO

Corporate Social Responsibility or CEO Narcissism? 271

political ideology, we examined political donations made by executives for the 10 years before they became CEOs. The data was obtained from U.S. Federal Election Commission (FEC), which records all individual contributions of more than $200 to candidates, campaign committees, parties, and to political action committees (PACs). We obtained the data from the Center for Responsive Politics (www.opensecrets.org) that reports data provided by FEC. Following Chin et al. (2013) we created four indicators of liberalism: (1) the number of donations to Democrats divided by the total number of donations (similar to Chin et al. (2013) we added 0.1 to all numerators and 0.2 to all denominators to handle zero values); (2) the dollar amount of dona- tions to Democrats divided by the dollar value of all donations; (3) the number of years the CEO made donations to Democrats divided by the number of years that the CEO donated to either party; (4) the number of distinct Democratic recipients divided by the number of total recipients. We included contri- butions to individual candidates, party committees, and PACs that were identified as either Democratic or Republican. We excluded any PACs with unclear orientation. The aggregate liberalism score was an average of all four indicators. We also included the dummy.

Firm control variables

To control for firm-specific conditions that might influence CSR intensity of a firm, we controlled for the availability of slack resources, measured as the ratio of current assets to current liabilities. We also controlled for prior year performance by including the firm’s ROA in the previous time period. Because large firms may face different pressures to be involved in CSR, we controlled for the size of the firm, which we measured as the natural logarithm of sales. To account for the possibility that CSR spending might be subject to previous trends idiosyncratic to a firm, we included previous year CSR in the model as well.

Industry control variables

We also accounted for the possibility that cer- tain industries have different levels of performance by including an industry dummy (three-digit SIC code) in the models. Table 1 provides the means, standard deviations, and bivariate correlations for all data.

Model and estimation

We estimated our models using generalized estimating equations (GEE) (Liang and Zeger, 1986) with an endogeneity control for consistency with previous research on the outcomes of invari- ant personal characteristics like CEO narcissism (Chatterjee and Hambrick, 2007, 2011). This estimation technique derives maximum likelihood estimates while controlling for nonindependence of observations. When our outcome measure (CPMP) had a Poisson zero-inflated distribution we specified a negative binomial distribution with a log link function (Chatterjee and Hambrick, 2007). All other models are specified including a Gaussian distribution with an identity link function. To ensure robustness of our estimation, we used a robust variance estimator (White, 1980). Because we expected endogeneity to bias our results we added an endogeneity control in our GEE esti- mation (Chatterjee and Hambrick, 2007, 2011).2

Additionally, we proceeded to reanalyze the per- formance model winsorizing ROA at both the 2.5 and 5 percent levels to control for the prevalence of extreme ROA values in COMPUSTAT and results are robust.

RESULTS

Table 2 reports the results for Hypotheses 1 and 2. We first show the results for our base control model (Model 1). Hypothesis 1 posits that there will be a positive relationship between CEO narcissism and

2 Endogeneity control. We controlled for the possibility that narcissistic CEOs might be drawn to firms exhibiting certain characteristics following Chatterjee and Hambrick (2007, 2011). First, we regressed CEO narcissism against a set of antecedent and contemporaneous variables. Antecedent variables are meant to capture CEO’s entry conditions and were measured in a year prior to the individual becoming the CEO of the firm. They included firm revenues, age, and ROA. We have also accounted for the possibility that early improvements in performance might stimulate narcissistic tendencies by including ROA change in the first year of CEO’s tenure in the model (Chatterjee and Hambrick, 2007). Other variables were measured in the year of CEO start and included measures of CEO power (duality, CEO ownership and board independence) and a dummy for whether CEO was an insider (hired from within a firm) or outsider. To account for narcissistic CEOs being drawn to certain industries, we also included the two digit SIC code indicator variable. We then create a predicted narcissism score based on this model and included the endogeneity control in our models. As a robustness check, we also analyze whether the exclusion of this variable affects our results and find that results did not change significantly if we omit the endogeneity control.

272 O. V. Petrenko et al.

T ab

le 1.

D es

cr ip

ti ve

st at

is ti

cs an

d co

rr el

at io

n co

ef fi

ci en

V ar

ia bl

e M

ea n

s. d.

1 2

3 4

5 6

7 8

9 10

11 12

13 14

15 16

17 18

19 20

1. C

S R

0. 34

4. 14

2. C

S R

co nc

er ns

4. 35

2. 87

− 0.

38 3.

C S

R st

re ng

th s

4. 69

4. 02

0. 75

0. 31

4. C

P M

P 0.

00 1.

00 0.

08 0.

01 0.

08 5.

R O

A 0.

05 0.

07 0.

24 −

0. 07

0. 20

0. 06

6. T

ob in

’s Q

1. 82

1. 21

0. 19

− 0.

18 0.

07 0.

09 0.

38 7.

M ar

ke t

va lu

e ad

de d

12 .2

7 35 .9

9 0.

26 0.

08 0.

33 0.

07 0.

38 0.

54 8.

N ar

ci ss

is m

4. 41

1. 31

0. 18

− 0.

19 0.

05 0.

07 0.

08 0.

20 0.

03 9.

A ge

54 .7

7 5.

89 −

0. 08

0. 27

0. 11

− 0.

08 0.

03 −

0. 16

− 0.

06 −

0. 10

10 .D

ua li

ty 0.

65 0.

47 0.

03 0.

15 0.

14 0.

03 0.

02 −

0. 07

− 0.

05 −

0. 04

0. 16

11 .C

E O

ow ne

rs hi

p 0.

46 1.

98 0.

09 −

0. 07

0. 04

0. 14

0. 06

0. 16

0. 09

0. 01

− 0.

13 0.

03 12

.T en

ur e

5. 5

5. 54

0. 08

− 0.

10 0.

01 0.

03 0.

03 −

0. 06

0. 09

0. 36

0. 23

0. 12

13 .U

na bs

or be

d sl

ac k

3. 17

17 .4

9 −

0. 01

− 0.

11 −

0. 08

0. 02

0. 03

0. 11

− 0.

01 0.

01 −

0. 02

− 0.

08 −

0. 01

0. 04

14 .S

iz e

(l og

of sa

le s)

9. 59

0. 99

0. 04

0. 50

0. 40

0. 06

0. 04

− 0.

18 0.

23 −

0. 09

0. 21

0. 14

0. 05

− 0.

06 −

0. 01

15 .E

nd og

en ei

ty co

nt ro

l 3.

96 2.

15 0.

06 −

0. 01

0. 05

0. 02

0. 03

0. 04

− 0.

33 0.

42 −

0. 19

0. 03

− 0.

05 −

0. 01

16 .C

E O

sh or

t- te

rm pa

y fo

cu s

0. 09

0. 14

− 0.

09 −

0. 04

− 0.

12 −

0. 01

0. 13

0. 07

0. 17

− 0.

02 −

0. 09

− 0.

16 −

0. 10

− 0.

06 −

0. 03

− 0.

04 −

0. 05

17 .C

E O

lo ng

-t er

m pa

y fo

cu s

0. 72

0. 23

0. 06

0. 13

0. 15

− 0.

01 −

0. 02

− 0.

16 −

0. 21

0. 09

0. 16

0. 22

0. 05

− 0.

01 0.

12 0.

04 −

0. 75

18 .C

E O

li be

ra li

sm 0.

36 0.

28 0.

02 −

0. 15

− 0.

08 0.

08 −

0. 02

0. 14

0. 01

0. 19

0. 10

− 0.

10 0.

13 0.

03 −

0. 07

0. 01

− 0.

04 19

.A dv

er ti

si ng

in te

ns it

y 0.

01 0.

02 0.

19 0.

09 0.

26 0.

04 0.

08 0.

23 0.

34 0.

13 0.

06 −

0. 12

0. 03

− 0.

01 −

0. 01

0. 16

− 0.

11 0.

03 −

0. 04

− 0.

02 20

.R &

D in

te ns

it y

0. 03

0. 07

0. 19

− 0.

05 0.

15 0.

02 0.

05 0.

50 0.

37 0.

22 −

0. 02

− 0.

07 0.

01 −

0. 06

0. 07

− 0.

15 0.

01 0.

02 0.

01 0.

09 0.

19 21

.C ap

it al

in te

ns it

y 0.

07 0.

14 −

0. 07

0. 06

− 0.

03 0.

01 −

0. 04

− 0.

07 −

0. 08

0. 02

0. 01

0. 12

0. 01

0. 26

− 0.

05 −

0. 09

0. 05

− 0.

08 −

0. 10

− 0.

C or

re la

ti on

s gr

ea te

r th

an 0.

05 ar

e si

gn ifi

ca nt

at p <

0. 05

Corporate Social Responsibility or CEO Narcissism? 273

Table 2. Effects of CEO narcissism on corporate social responsibility (CSR) and corporate philanthropy media profile (CPMP) (GEE analyses)

Model 1 Model 2 Model 3 Model 4 Model 5 Model 6

DV: CSR DV: CSR DV: Strengths DV: Concerns DV: CPMP DV: CPMP

Constant −1.76* (0.82) −2.63** (0.79) −4.99** (1.76) −1.13 (1.69) −5.38*** (1.78) −5.41*** (1.77) Lagged DV 0.98*** (0.01) 0.97*** (0.02) 0.95*** (0.02) 0.81*** (0.02) 0.03 (0.07) −0.02 (0.07) CEO duality 0.12 (0.12) 0.15 (0.11) −0.07 (0.17) −0.05 (0.03) 0.08 (0.27) 0.07 (0.27) CEO age 0.01 (0.01) 0.01 (0.01) 0.02 (0.02) 0.05* (0.03) −0.07* (0.03) −0.06* (0.02) CEO

ownership −0.03 (0.03) −0.02 (0.02) 0.05 (0.05) 0.08* (0.04) −0.02 (0.03) −0.02 (0.03)

CEO tenure 0.01 (0.01) 0.02 (0.01) 0.10*** (0.02) −0.11*** (0.02) 0.04 (0.02) 0.04 (0.02) Unabsorbed

slack 0.01 (0.01) −0.01 (0.01) −0.01 (0.01) 0.01 (0.01) −0.01 (0.01) −0.01 (0.01)

Size (log of sales)

0.12 (0.05) 0.15** (0.05) 0.16*** (0.04) 0.20*** (0.05) 0.55*** (0.14) 0.53*** (0.14)

ROA 2.34** (0.76) 2.23*** (0.58) 2.08** (0.67) −1.71* (0.70) 2.62 (1.37) 2.49 (1.38) Endogeneity

control −0.01 (0.03) −0.01 (0.03) 0.09 (0.08) 0.20* (0.09) −0.10 (0.06) −0.10 (0.05)

Short-term pay focus

−0.63 (0.60) −0.58 (0.48) 0.39 (0.45) 0.01 (0.54) −0.47 (0.86) −0.22 (0.86)

Long-term pay focus

0.53 (0.34) 0.36 (0.27) −0.05 (0.26) 0.04 (0.32) −0.79 (0.48) −0.70 (0.48)

Liberalism −0.25 (0.19) −0.38 (0.21) −0.30 (0.17) −0.11 (0.15) 0.51 (0.48) 0.62 (0.49) Narcissism 0.15*** (0.03) 0.07** (0.02) −0.08** (0.03) 0.35** (0.12) 0.33** (0.12) CPMPt-1 0.16** (0.06) 0.03 (0.11) Narcissism ×

CPMPt−1 0.05** (0.02)

Observations 1,004 1,004 1,004 1,004 1,004 1,004 Wald Chi2 6,374*** 6,961*** 7,590*** 4,559*** 46.48*** 52.99***

Lagged DV at t - 1 for Models 1 – 4 and t - 2 for Models 5 and 6. Standard errors are in parentheses. *p < 0.05; **p < 0.01; ***p < 0.001

corporate social responsibility. The results of Model 2 provide strong support for Hypothesis 1 (0.15, p < 0.001). Additionally, we performed some sup- plementary analyses to evaluate if our hypothesis was supported for both the strength and concerns aspect of CSR. According to our arguments for Hypothesis 1, we expected the composite aspects of CSR to be independently related to CEO nar- cissism. Consistent with this logic, Models 3 and 4 show that CEO narcissism is positively related to the Strengths dimension of CSR (0.07, p < 0.01) and negatively related to the Concerns dimension of CSR (−0.08, p < 0.01), implying the narcissis- tic CEOs emphasize activities associated with CSR strengths and avoid actions that raise CSR concerns.

Hypothesis 2 posits that CEO narcissism will positively moderate the relationship between the previous level of media profile for firm corpo- rate philanthropy and the level of media profile for firm corporate philanthropy. Model 5 shows that CEO narcissism is positively related to cor- porate philanthropy media profile (0.35, p < 0.01)

implying that narcissistic CEOs engage in higher profile corporate philanthropy. Additionally, Model 6 shows that CEO narcissism strengthens the rela- tionship between previous corporate philanthropy high profile and subsequent corporate philanthropy high profile (0.05, p < 0.01), indicating that firms with more narcissistic CEOs are more likely to continue to generate higher profile corporate phi- lanthropy than those of their less narcissistic peers.3

Figure 1 illustrates this interaction. In both models we added a second lag of the DV to control for pre- vious value of CPMP. Results are the same with or without this additional control for firm specific lev- els of CPMP. Models 5 and 6 in Table 2 provide support for Hypothesis 2.

3 We tested and found no support for ceiling effect (b = −0.16, z = −1.38 and b = −0.01, z = −0.41 respectively) for both the quadratic and the quadratic interaction terms. Ceiling effects could be expected because of media saturation but are not present probably due to lack of maturity in coverage of such effects because of either recency in interest in coverage of CSR or resets of CSR focus at the firm or industry levels resulting from economic or business crises.

274 O. V. Petrenko et al.

Figure 1. The relationship between corporate philan- thropy media profile (CPMP) at time t and t + 1 for low

narcissists and high narcissists CEOs

Table 3 presents results for Hypothesis 3. Hypothesis 3 posits that CEO narcissism will neg- atively moderate the relationship between CSR and performance. Model 7 in Table 3 shows that CEO narcissism weakens the relationship between CSR and firm performance measured as ROA (0.02, p < 0.05). Our result has practical significance because it indicates ROA will become a nominal 2.16 percent higher for firms with less narcissistic

Figure 2. The relationship between corporate social responsibility (CSR) and performance (ROA) for low nar-

cissists and high narcissists CEOs

CEOs than for firms with more narcissistic CEOs, when CSR increases one standard deviation above the mean in the sample. Figure 2 illustrates this interaction and shows that the relationship between firm performance and CSR is much more posi- tive for relatively low narcissism CEOs than for relatively high narcissism CEOs.

We performed two supplementary analyses to add depth to these findings. First, we investigated

Table 3. Results for the effects of CEO narcissism on performance (GEE analyses)

Model 7 Model 8 Model 9 Model 10

DV: ROA DV: ROA DV: TQ DV: MVA

Constant −0.01 (0.04) −0.01 (0.03) 1.67*** (0.29) 139.94 (9.83) Lagged DV 0.21*** (0.03) 0.26*** (0.02) 0.62*** (0.01) 0.83*** (0.04) Endogeneity control 0.01 (0.01) 0.01 (0.01) 0.01 (0.01) 0.27 (0.18) Size (log of sales) 0.02*** (0.01) 0.02*** (0.01) −0.04 (0.02) 0.91 (0.06) Advertising intensity 0.04 (0.17) 0.14 (0.16) 4.60** (1.33) 5.20 (4.48) R&D intensity −0.28*** (0.04) −0.07 (0.04) 1.31*** (0.33) 8.18 (9.59) Capital intensity −0.01 (0.03) 0.03 (0.03) −0.14 (0.17) −2.87 (2.60) Unabsorbed slack −0.01 (0.01) 0.01 (0.01) 0.01 (0.00) 0.02* (0.01) CSR 0.07** (0.02) 0.06*** (0.01) 0.93** (0.35) CEO narcissism −0.03 (0.02) −0.07* (0.03) −0.01 (0.02) −0.76** (0.29) CEO narcissism × CSR −0.02* (0.01) −0.03*** (0.01) −0.14* (0.06) CSR strengths 0.05* (0.02) CSR concerns −0.11*** (0.02) CEO narcissism × CSR strengths −0.09* (0.04) CEO narcissism × CSR concerns 0.02* (0.01) Observations 1,051 1,051 911 911 Wald Chi2 786*** 964*** 14,333*** 16,099***

Standard errors are in parentheses. *p < 0.05; **p < 0.01; ***p < 0.001

Corporate Social Responsibility or CEO Narcissism? 275

if the effects of CEO narcissism on the relationship between CSR and performance held for both the strength and concerns dimensions of CSR. Model 8 shows support for these analyses. As expected, CEO narcissism negatively moderates the relationship between CSR Strengths and ROA (−0.09, p < 0.05) and positively moderates the relationship between CSR Concerns and ROA (0.02, p < 0.05). Therefore ROA will be lower for firms with more narcissistic CEOs when they have more CSR strengths or less CSR concerns than for the firms of their less narcissistic peers.

Second, while our predictions are about opera- tional performance making ROA our variable of choice, we replicated this analysis for two com- monly utilized measures of market performance — Tobin’s Q and Market Value Added — as additional validation for our findings and found consistent results. Models 9 and 10 show results for TQ (0.03, p < 0.001) and MVA (0.14, p < 0.05).

DISCUSSION

We began by suggesting that CSR initiatives may result from leaders’ personal needs for attention and image reinforcement and that, in such cases, CSR initiatives may be less strategic in terms of financial performance for their organizations. With insights from both upper echelons and agency per- spectives, we have theorized and shown that the corporate social responsibility of firms can be sig- nificantly affected by CEO narcissism. Consistent with the logic that narcissistic CEOs crave attention, we have also theorized and shown that firms with more narcissistic CEOs have higher profile corpo- rate philanthropy than the firms of their less narcis- sistic peers, and that they also continue to engage in high profile corporate philanthropy once they suc- ceed at it. Finally, we have shown that the positive relationship between CSR and firm performance is weaker for firms with more narcissistic CEOs, pos- sibly indicating that CEO narcissism may preempt other stakeholder or strategic considerations in CSR decisions. These insights contribute to the litera- ture on executive characteristics and organizational decision making, to the agency literature, and to the growing literature on CSR.

Theoretical contribution

Our study opens two main pathways for consider- ation by CSR researchers. First, we extended the

integration of upper echelons approaches with the CSR literature by showing that CSR can be an outcome that helps satisfy leaders’ personal needs for attention and image reinforcement as in the case of narcissistic CEOs. With this, we provide new insight about CSR antecedents, which will be relevant for researchers interested in the strategic nature of CSR decisions. Second, we reconceptu- alized CSR as an agency outcome. As such, CSR can be constructed as an opportunity to extract per- sonal value from the organization by organizational incumbents. If CSR provides organizational incum- bents opportunities to satisfy their personal needs, it may also be an opportunity for actual recogni- tion, prestige, personal network development and other personally relevant outcomes for organiza- tional decision makers that may help better under- stand the motivations for CSR. This potentially dark side of CSR opens new opportunities for research in CSR.

Our study also contributes to the agency theory literature. Agency researchers have viewed agency behaviors mostly from a risk-centered perspective by studying agency effects on risk decision making by executives. Recasting CSR as a visible opportu- nity to acquire prestige and praise through organi- zational resource allocations, we open an avenue to explore alternative agency behaviors that may have personal value for executives based on their charac- teristics but that may also provide executives with marketable value for their services beyond their actual employment contract.

Our findings also raise intriguing questions for upper echelons researchers. If two strategic deci- sions are driven by different antecedents, should we assume similar organizational meanings for the decisions, or should we expect the decisions to vary in their strategic value or their effects on firm perfor- mance? While we tend to assume similar meaning for similar strategic decisions or investments, our study shows that antecedents may not only result in varying strategic decisions but also in varying effects on performance. As an example of this, our study shows that CSR levels do not result in the same performance outcomes when performed by organizations in which the decision may be moti- vated by CEO narcissism. In our study, CSR by organizations with CEOs that were relatively lower in narcissism is related to performance but not in the case of organizations in which the CEOs had relatively higher narcissistic characteristics. Simi- larly, we could question, given an upper echelons

276 O. V. Petrenko et al.

prediction for a relationship between executive characteristics and a particular strategic decision like, for example, R&D expenditures, if the decision will have the same meaning for the organization and take a similar implementation form or have a similar effect on firm performance when driven by different antecedents.

Methodological contribution

Beyond our theoretical contribution to the upper echelons, the agency and the CSR literatures, our paper contributes to the upper echelons method- ology arsenal by developing a new approach to measurement of CEO characteristics. Utilizing third-party ratings of video samples, we provide a method for executive measurement that allows valid access to direct but unobtrusive measurement of CEO characteristics. This novel videometric approach overcomes several limitations in the measurement of CEO characteristics. It allows researchers to circumvent executive reticence to participate in surveys because it provides unobtru- sive but direct access to a large sample of CEOs since video samples of CEOs in the public domain are already ubiquitous online and are likely to increase in availability and quality as the presence of video documentation continues to grow for all aspects of social and corporate life. Also, it makes it possible to use previously validated psychometric scales, without concerns for desirability biases, selective participation, or responder identity. As such, the new videometric approach to executive characteristics measurement that we espouse, demonstrate, and check for robustness in this paper provides a basis for the development of research in upper echelons because of its ability to provide valid measurement of a wide range of personal characteristics that may not have been easily addressed through alternative measurement.

In sum, we show that CEO narcissism has a positive effect on organizational CSR because CSR may be a response to leaders’ personal needs for attention and image reinforcement. We also show that CSR in firms with more narcissistic CEOs may be less strategic in terms of financial performance in part because of a resulting lack of focus in CSR configurations. We hope that our study stimulates future research not only because of its implications for the upper echelons, the agency theory and the CSR literatures, but also because of the opportunities it provides for extending research

on executive characteristics in strategy through our novel methodological approach to the measurement of executive characteristics.

ACKNOWLEDGMENTS

We gratefully acknowledge Jim Westphal and two anonymous reviewers for their constructive sugges- tions that helped shape this manuscript.

REFERENCES

Agle BR, Mitchell RK, Sonnenfeld JA. 1999. Who matters to CEOs? An investigation of stakeholder attributes and salience, corporate performance, and CEO values. Academy of Management Journal 42: 507 – 525.

Aguilera RV, Rupp DE, Williams CA, Ganapathi J. 2007. Putting the S back in corporate social responsibility: a multilevel theory of social change in organizations. Academy of Management Review 32: 836 – 863.

Arthur W, Day EA, McNelly TL, Edens PS. 2003. A meta-analysis of the criterion-related validity of assess- ment center dimensions. Personnel Psychology 56: 125 – 154.

Ballinger GA, Marcel JJ. 2010. The use of an interim CEO during succession episodes and firm performance. Strategic Management Journal 31: 262 – 283.

Benjamin DJ, Shapiro JM. 2009. Thin-slice forecasts of gubernatorial elections. Review of Economics and Statistics 91: 523 – 536.

Bliese PD. 2000. Within-group agreement, non- independence, and reliability: implications for data aggregation and analysis. In KJ Klein and SWJ Kozlowski. Multilevel Theory, Research, and Methods in Organizations (pp. 349 – 381). San Francisco, CA: Jossey-Bass.

Bogart LM, Benotsch EG, Pavlovic JD. 2004. Feeling superior but threatened: the relation of narcissism to social comparison. Basic and Applied Social Psychol- ogy 26: 35 – 44.

Borkenau P, Brecke S, Mottig C, Paelecke M. 2009. Extraversion is accurately perceived after a 50-ms exposure to a face. Journal of Research in Personality 43: 703 – 706.

Borkenau P, Liebler A. 1993. Convergence of stranger ratings of personality and intelligence with self-ratings, partner ratings, and measured intelligence. Journal of Personality and Social Psychology 65: 546 – 553.

Campbell WK, Foster CA, Finkel EJ. 2002. Does self-love lead to love for others?: a story of narcissistic game playing. Journal of Personality and Social Psychology 83: 340 – 354.

Campbell WK, Goodie AS, Foster JD. 2004. Narcissism, confidence, and risk attitude. Journal of Behavioral Decision Making 17: 297 – 311.

Carpenter MA, Geletkanycz MA, Sanders WG. 2004. Upper echelons research revisited: antecedents, ele- ments, and consequences of top management

Corporate Social Responsibility or CEO Narcissism? 277

team composition. Journal of Management 30: 749 – 778.

Chandler A. 1962. Strategy and Structure. MIT Press: Cambridge, MA.

Chatterjee A, Hambrick DC. 2007. It’s all about me: narcissistic CEOs and their effects on company strategy and performance. Administrative Science Quarterly 52: 351 – 386.

Chatterjee A, Hambrick DC. 2011. Executive personality, capability cues, and risk taking: how Narcissistic CEOs react to their successes and stumbles. Administrative Science Quarterly 56: 202 – 237.

Chin MK, Hambrick DC, Treviño LK. 2013. Political ide- ologies of CEOs: the influence of executives’ values on corporate social responsibility. Administrative Science Quarterly 58: 197 – 232.

Choi J, Wang H. 2009. Stakeholder relations and the per- sistence of corporate financial performance. Strategic Management Journal 30: 895 – 907.

Clark GL, Hebb T. 2004. Corporate engagement: the fifth stage of capitalism. Relations Industrielles/Industrial Relations 59: 142 – 169.

Connelly BS, Hulsheger UR. 2012. A narrower scope or a clearer lens for personality? Examining sources of observers’ advantages over self-reports for predicting performance. Journal of Personality 80: 603 – 631.

Connelly BS, Ones DS. 2010. Another perspective on per- sonality: meta-analytic integration of observers’ accu- racy and predictive validity. Psychological Bulletin 136: 1092 – 1122.

Coombs JE, Gilley KM. 2005. Stakeholder management as a predictor of CEO compensation: main effects and interactions with financial performance. Strategic Management Journal 26: 827 – 840.

Cornell B, Shapiro AC. 1987. Corporate stakeholders and corporate finance. Financial Management 16: 5 – 14.

Cramer P. 1998. Freshman to senior year: a follow-up study of identity, narcissism, and defense mechanisms. Journal of Research in Personality 32: 156 – 172.

Cycyota CS, Harrison DA. 2006. What (not) to expect when surveying executives: a meta-analysis of top man- ager response rates and techniques over time. Organi- zational Research Methods 9: 133 – 160.

Dahlmann F, Brammer S. 2011. Exploring and explaining patterns of adaptation and selection in corporate envi- ronmental strategy in the USA. Organizational Studies 32: 527 – 553.

Darnall N, Edwards D. 2006. Predicting the cost of envi- ronmental management system adoption: the role of capabilities, resources and ownership structure. Strate- gic Management Journal 27: 301 – 320.

David P, Bloom M, Hillman AJ. 2007. Investor activism, managerial responsiveness, and corporate social performance. Strategic Management Journal 28: 91 – 100.

Deckop JR, Merriman KK, Gupta S. 2006. The effects of CEO pay structure on corporate social performance. Journal of Management 32: 329 – 342.

Deluga RJ. 1997. Relationship among American presiden- tial charismatic leadership, narcissism, and rated per- formance. Leadership Quarterly 8: 49 – 65.

Donaldson T, Preston L. 1995. The stakeholder theory of corporation: concepts, evidence, and implications. Academy of Management Review 20: 65 – 91.

Emmons RA. 1981. Relationship between narcissism and sensation seeking. Psychological Reports 48: 247 – 250.

Entine J. 2003. The myth of social investing: a critique of its practice and consequences for corporate social performance research. Organization and Environment 16: 352 – 368.

Finkelstein S. 1992. Power in top management teams: dimensions, measurement, and validation. Academy of Management Journal 35: 505 – 538.

Finkelstein S, Boyd BK. 1998. How much does the CEO matter? The role of managerial discretion in the setting of CEO compensation. Academy of Management Jour- nal 41: 179 – 199.

Finkelstein S, Hambrick DC. 1996. Strategic Leadership: Top Executives and their Effects on Organizations. West: St. Paul, MN.

Flynn FJ, Staw BM. 2004. Lend me your wallets: the effect of charismatic leadership on external support for an organization. Strategic Management Journal 25: 309 – 330.

Freeman RE. 1984. Strategic Management: A Stakeholder Approach. Pitman: Boston, MA.

Friedman M. 1970. The social responsibility of business is to increase its profits. New York Times Magazine 13 September: 32 – 33.

Frooman J. 1999. Stakeholder influence strategies. Academy of Management Review 24: 191 – 205.

Gerstner W, König A, Enders A, Hambrick DC. 2013. CEO Narcissism, audience engagement, and organiza- tional adoption of technological discontinuities. Admin- istrative Science Quarterly 58(2): 257 – 291.

Godfrey PC, Merrill CB, Hansen JM. 2009. The rela- tionship between corporate social responsibility and shareholder value: an empirical test of the risk manage- ment hypothesis. Strategic Management Journal 30(4): 425 – 445.

Graves SB, Waddock SA. 1994. Institutional owners and corporate social performance. Academy of Management Journal 37: 1035 – 1046.

Hambrick DC. 2007. Upper echelons theory: an update. Academy of Management Review 32: 334 – 343.

Hambrick DC, Mason PA. 1984. Upper echelons: the orga- nization as a reflection of its top managers. Academy of Management Review 9: 193 – 206.

Hillman AJ, Keim GD. 2001. Shareholder value, stakeholder management, and social issues: what’s the bottom line? Strategic Management Journal 22: 125 – 139.

Hull CE, Rothenberg S. 2008. Firm performance: the inter- actions of corporate social performance with innova- tion and industry differentiation. Strategic Management Journal 29: 781 – 789.

Husted BW, Salazar JDJ. 2006. Taking Friedman seri- ously: maximizing profits and social performance. Journal of Management Studies 43: 75 – 91.

Jones MT. 1995. The institutional determinants of social responsibility. Journal of Business Ethics 20: 163 – 179.

278 O. V. Petrenko et al.

Judge TA, Bono JE, Illies R, Rehardt M. 2002. Personality and leadership: a qualitative review. Journal of Applied Psychology 87: 765 – 780.

Judge TA, LePine J, Rich BL. 2006. Loving yourself abun- dantly: relationship of narcissistic personality to self and other perceptions of workplace deviance, leader- ship, and task and contextual performance. Journal of Applied Psychology 91: 762 – 776.

Kacperczyk A. 2009. With greater power comes greater responsibility? Takeover protection and corporate attention to stakeholders. Strategic Management Journal 30: 261 – 285.

Kansi J. 2003. The narcissistic personality inventory: applicability in a Swedish population sample. Scandi- navian Journal of Psychology 44: 441 – 448.

Kernberg OF. 1975. Borderline Conditions and Patholog- ical Narcissism. Jason Aronson: New York.

Kernberg OF. 1986. Narcissistic personality disorder. In The Personality Disorder and Neuroses (Volume 1), Cooper AM, Frances AJ, Sachs MH (eds). Basic Books: New York; 219 – 231.

Kernis MH, Sun CR. 1994. Narcissism and reactions to interpersonal feedback. Journal of Research in Person- ality 28: 4 – 13.

Kets de Vries MFR, Miller D. 1985. Narcissism and leader- ship: an object relations perspective. Human Relations 38: 583 – 601.

Knauer NJ. 1994. The paradox of corporate giving: tax expenditures, the nature of the corporation, and the social construction of charity. DePaul Law Review 44: 1 – 97.

Kohut H. 1977. The Restoration of the Self . International Universities Press: New York.

Lewis M. 2008. The History of doing good. Forbes 16 October. Available at http://www.forbes.com/2008/ 10/16/csr-history-reform-lead-corprespons08-cx_ml_ 1016lewis.html.

Liang KY, Zeger SL. 1986. Longitudinal data analy- sis using generalized linear models. Biometrika 73: 13 – 22.

Mattingly JE, Berman SL. 2006. Measurement of corpo- rate social action: discovering taxonomy in the Kinder Lydenburg Domini ratings data. Business Society 45: 20 – 46.

Margolis JD, Walsh JP. 2003. Misery loves companies: rethinking social initiatives by business. Administrative Science Quarterly 48: 268 – 305.

Marquis C, Glynn MA, Davis GF. 2007. Community isomorphism and corporate social action. Academy of Management Review 32: 925 – 945.

McGahan AM, Porter ME. 1997. How much does industry matter really? Review of Economics and Statistics 81: 143 – 153.

McGuire J, Dow S, Argheyd K. 2003. CEO incentives and corporate social performance. Journal of Business Ethics 45(4): 341 – 359.

McNamara G, Aime F, Vaaler PM. 2005. Is performance driven by industry- or firm-specific factors? A response to Hawawini, Subramanian, and Verdin. Strategic Man- agement Journal 26: 1075 – 1081.

McWilliams A, Siegel D. 2000. Corporate social respon- sibility and financial performance: correlation or

misspecification? Strategic Management Journal 21: 603 – 609.

McWilliams A, Siegel D. 2001. Corporate social respon- sibility: a theory of the firm perspective. Academy of Management Review 26: 117 – 127.

Miller D, Kets de Vries M, Toulouse JM. 1982. Top executive locus of control and its relationship to strategy-making, structure, and environment. Academy of Management Journal 25: 237 – 253.

Morf CC, Rhodewalt F. 2001. Unraveling the paradoxes of narcissism: a dynamic self-regulatory processing model. Psychological Inquiry 12: 177 – 196.

Mount MK, Barrick MR, Strauss JP. 1994. Validity of observer ratings of the big five personality factors. Journal of Applied Psychology 79: 272 – 280.

Muller A, Kolk A. 2010. Extrinsic and intrinsic drivers of corporate social performance: evidence from foreign and domestic firms in Mexico. Journal of Management Studies 47: 1 – 26.

Neubaum DO, Zahra SA. 2006. Institutional ownership and corporate social performance: the moderat- ing effects of investment horizon, activism, and coordination. Journal of Management 32: 108 – 131.

Nunnally JC. 1978. Psychometric Theory. McGraw-Hill: New York.

Oh I, Wang G, Mount MK. 2011. Validity of observer ratings of the five-factor model of personality traits: a meta-analysis. Journal of Applied Psychology 96: 762 – 773.

Orlitzky M, Benjamin JD. 2001. Corporate social perfor- mance and firm risk: a meta-analytic review. Business and Society 40: 369 – 396.

Pratto F, Sidanius J, Stallworth LM, Malle BF. 1994. Social dominance orientation: a personality variable predicting social and political attitudes. Journal of Personality and Social Psychology 67: 741 – 763.

Ramchander S, Schwebach RG, Staking K. 2012. The informational relevance of corporate social responsibil- ity: evidence from DS400 index reconstitutions. Strate- gic Management Journal 33: 303 – 314.

Raskin R, Novacek J, Hogan R. 1991. Narcissistic self-esteem management. Journal of Personality and Social Psychology 60: 911 – 918.

Raskin R, Terry H. 1988. A principal-components analy- sis of narcissistic personality inventory and further evi- dence of its construct validity. Journal of Personality and Social Psychology 54: 890 – 902.

Rhodewalt F, Eddings SK. 2002. Narcissus reflects: mem- ory distortion in response to ego-relevant feedback among high- and low-narcissistic men. Journal of Research in Personality 36: 97 – 116.

Rhodewalt F, Morf CC. 1995. Self and interpersonal correlates of the Narcissistic Personality Inventory: a review and new findings. Journal of Research in Personality 29: 1 – 23.

Rhodewalt F, Morf CC. 1998. On self-aggrandizement and anger: a temporal analysis of narcissism and affective reactions to success and failure. Journal of Personality and Social Psychology 74: 672 – 685.

Rhodewalt F, Sorrow DL. 2003. Interpersonal self-regulation: lessons from the study of narcis- sism. In Handbook of Self and Identity, Leary

Corporate Social Responsibility or CEO Narcissism? 279

MR, Tangney JP (eds). Guilford Press: New York; 519 – 535.

Ridge JW, Aime F,, White MA. 2014. When much more of a difference makes a difference: social comparison and tournaments in the CEO’s top team. Strategic Management Journal , DOI: 10.1002/smj.2227.

Riemann R, Angleitner A, Strelau J. 1997. Genetic and environmental influences on personality: a study of twins reared together using the self- and peer report NEO-FFI scales. Journal of Personality 65: 449 – 475.

Rowley T, Berman S. 2000. A brand new brand of corporate social performance. Business and Society 39: 397 – 418.

Rubenzer SJ, Faschingbauer TR, Ones DS. 2000. Assess- ing the U.S. presidents using the Revised NEO Person- ality Inventory. Assessment 7: 403 – 420.

Sanders G. 2001a. Behavioral responses of CEOs to stock ownership and stock option pay. Academy of Management Journal 44(3): 477 – 492.

Sanders WG. 2001b. Incentive alignment, CEO pay level, and firm performance: a case of “heads I win, tails you lose”? Human Resource Management 40(2): 159 – 170.

Schmalensee R. 1985. Do markets differ much? American Economic Review 75: 341 – 351.

Schnietz KE, Epstein MJ. 2005. Exploring the financial value of a reputation for corporate social responsibil- ity during a crisis. Corporate Reputation Review 7: 327 – 345.

Sen S, Gurhan-Canli Z, Morwitz V. 2001. Withholding consumption: a social dilemma perspective on con- sumer boycotts. Journal of Consumer Research 28: 399 – 417.

Shleifer A, Vishny RW. 1991. Takeovers in the ‘60s and the ‘80s: evidence and implications. Strategic Management Journal 12: 51 – 61.

Siegel S, Castellan NJ. 1988. Nonparametric Statistics for the Behavioral Sciences. McGraw-Hill: New York.

Sirower ML. 1994. Acquisition behavior, strategic resource commitments and the acquisition game: a new perspective on performance and risk in acquiring firms. Unpublished doctoral diss., Colombia University, New York.

Svindseth MF, Nottestad JA, Wallin J, Roaldset JO, Dahl AA. 2008. Narcissism in patients admitted to psy- chiatric acute wards: its relation to violence, suici- dality and other psychopathology. BMC Psychiatry 8: 13 – 24.

Tobin J, Barnard W. 1968. Pitfalls in financial model building. American Economic Review 58: 99 – 122.

Turban DB, Greening DW. 1997. Corporate social per- formance and organizational attractiveness to prospec- tive employees. Academy of Management Journal 40: 658 – 672.

Van Iddekinge CH, Raymark PH, Roth PL. 2005. Assess- ing personality with a structured employment inter- view: construct-related validity and susceptibility to

response inflation. Journal of Applied Psychology 90: 536 – 552.

Vazire S, Funder DC. 2006. Impulsivity and the self-defeating behavior of narcissists. Personality and Social Psychology Review 10: 154 – 165.

Vergne J-P, Durand R. 2010. The missing link between the theory and empirics of path dependence: concep- tual clarification, testability issue, and methodologi- cal implications. Journal of Management Studies 47: 736 – 759.

Waddock S. 2004. Parallel universes: companies, aca- demics and the progress of corporate citizenship. Busi- ness and Society Review 109: 5 – 42.

Waddock S, Graves S. 1997. The corporate social performance — financial performance link. Strategic Management Journal 18: 303 – 319.

Waldman DA, Siegel D. 2008. Defining the socially responsible leader. Leadership Quarterly 19: 117 – 131.

Wallace HM, Baumeister RF. 2002. The performance of narcissists rises and falls with perceived opportunity for glory. Journal of Personality and Social Psychology 82: 819 – 834.

Walls JL, Berrone P, Phan PH. 2012. Corporate gover- nance and environmental performance: is there really a link? Strategic Management Journal 33: 885 – 913.

Webb E, Campbell D, Schwartz R, Sechrest L. 1966. Unobtrusive Methods: Non Reactive Measures in the Social Sciences. Rand McNally: Chicago, IL.

Weidenbaum ML, Jensen M. 2009. Introduction to the transaction edition. In The Modern Corporation and Private Property, Berle AA, Means GC (eds). Trans- action Publishers: New Brunswick, NJ; ix – xviii

Westphal JD, Bednar MK. 2005. Pluralistic ignorance in corporate boards and firms’ strategic persistence in response to low firm performance. Administrative Science Quarterly 50: 262 – 298.

White HC. 1980. A heteroskedasticity-consistent covari- ance matrix estimator and a direct test for heteroskedas- ticity. Econometrica 48: 817 – 838.

Wink P, Donahue K. 1997. The relation between two types of narcissism and boredom. Journal of Research in Personality 31: 136 – 140.

Wright P, Ferris SP. 1997. Agency conflict and corporate strategy: the effect of divestment on corporate value. Strategic Management Journal 18: 77 – 83.

Wong EM, Ormiston ME, Tetlock PE. 2011. The effects of top management team integrative complexity and decentralized decision making on corporate social performance. Academy of Management Journal 54: 1207 – 1228.

Zahra SA, George G. 2002. Absorptive capacity: a review, reconceptualization, and extension. Academy of Man- agement Review 27: 185 – 203.

Zimmerman R, Triana MC, Barrick M. 2010. The criterion-related validity of a structured letter using multiple raters and multiple performance criteria. Human Performance 23: 361 – 378.

View publication statsView publication stats

https://www.researchgate.net/publication/268528664