Content uploaded by Aaron D. Hill
Author content
All content in this area was uploaded by Aaron D. Hill on Apr 30, 2018
Content may be subject to copyright.
Strategic Management Journal
Strat. Mgmt. J.,37: 262– 279 (2016)
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,1FEDERICO AIME,1*JASON RIDGE,2
and AARON HILL1
1Department of Management, Oklahoma State University, Stillwater, Oklahoma,
U.S.A.
2Department 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 prole of organizational CSR; additionally, CEO narcissism will reduce the effect
of CSR on performance. We nd 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 nance 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 nan-
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
Copyright © 2015 John Wiley & Sons, Ltd.
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 reect a form of alignment
of rm policy to its stakeholder value preferences;
fewer have examined the effects of internal values
at the organizational or individual levels (e.g., the
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 dened as “actions that appear to further some
social good, beyond the interests of the rm 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 rm 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 nancial 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.
Specically 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
attery (Bogart, Benotsch, and Pavlovic, 2004; Wal-
lace and Baumeister, 2002). Since CSR activities
offer a difcult 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—dened as CEOs who have
inated 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 signicant benets 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
rms’ 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 rst
hypothesize and nd that organizations controlled
by narcissistic CEOs will engage in more CSR.
In line with this same logic, we also look at spe-
cic 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 inuence of CEOs’ personal interests on CSR.
We specically show that CEO narcissism will
negatively moderate the relationship between CSR
and rm performance. This last nding is important
for the discussion of the relationship between CSR
and rm performance because it highlights a CSR
cost that may not even be values driven, like in
Friedman’s depiction of CSRs as “spending
Copyright © 2015 John Wiley & Sons, Ltd. Strat. Mgmt. J.,37: 262– 279 (2016)
DOI: 10.1002/smj
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 fulll their
personal needs may put at issue the idea that while
CSR can certainly be a valuable strategic choice for
rms, they can also be nonstrategic when driven by
managerial self-interest. Given the complex nature
of the relationship between CSR and nancial
performance, with about half of the studies on the
relationship between CSR and nancial perfor-
mance reporting positive relationships and the rest
of the studies suggesting mixed or nonsignicant
ndings (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 nancial 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. Speci-
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 nd 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 signicant
new tool to the upper echelons literature for the
measurement of a wide array of personal charac-
teristics of leaders that were previously difcult
to assess consistently for strategy researchers.
Third, this research also contributes to agency
theorizing. Drawing on upper echelons theory we
nd that narcissism may result in agentic behaviors
that may have negative nancial 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 ndings in the relationship between CSR
and rm 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 rm 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 difcult
to evaluate (Margolis and Walsh, 2003; Waldman
and Siegel, 2008). Building on the logic of the upper
echelons perspective, CEOs will have a signicant
inuence in such discretionary decisions and there-
fore a rm’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 rms
and have substantial inuence 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 rm strategies. They found that narcissism
predicts the dynamism and grandiosity of rms’
strategic actions and that narcissistic CEOs react
differently to their success and seek social praise
(Chatterjee and Hambrick, 2011).
Copyright © 2015 John Wiley & Sons, Ltd. Strat. Mgmt. J.,37: 262– 279 (2016)
DOI: 10.1002/smj
Corporate Social Responsibility or CEO Narcissism? 265
Narcissistic individuals construe reality in part
as it reects 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 rm, 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 specically, 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 prole 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 reect 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 prole 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 reective of the
CEOs values or preferences. As such, we expect a
positive relationship between CEO narcissism and
the high prole of their corporate philanthropy. We
focus here on predictions about the media prole of
the philanthropic actions by rms 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 prole of rms run by more
narcissistic CEOs provides additional evidence for
the mechanisms in our theory.
Copyright © 2015 John Wiley & Sons, Ltd. Strat. Mgmt. J.,37: 262– 279 (2016)
DOI: 10.1002/smj
266 O. V. Petrenko et al.
Additionally, two aspects of narcissistic person-
ality proles suggest that narcissistic CEOs will
not only engage their organizations in higher pro-
le 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 prole 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-
le corporate philanthropy to have positive direct
and interactive effects on later high prole corporate
philanthropy. Therefore, we hypothesize that more
narcissistic CEOs will be more likely to continue to
generate higher prole corporate philanthropy than
their less narcissistic peers:
Hypothesis 2: CEO narcissism will positively
moderate the relationship between prior corpo-
rate philanthropy media prole and current cor-
porate philanthropy media prole.
Narcissistic CSR and nancial performance
A dominant concern of the CSR literature has
been the concern with the instrumental relation-
ship between CSR and rm nancial performance.
Researchers have argued that CSR can serve as
strategic choices that may lead to rm 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 ndings about
the relationship between CSR and rm performance
generally positive but mixed, the question remains:
if CSR choices are made in part based on CEO char-
acteristics, or more specically to satisfy narcissis-
tic CEOs needs for praise and attention, will they be
less likely to relate to nancial performance?
We expect CEO narcissism to moderate the pos-
itive relationship between CSR and rm nancial
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 rms
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 condence 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 rm 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 rms 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 reected
Copyright © 2015 John Wiley & Sons, Ltd. Strat. Mgmt. J.,37: 262– 279 (2016)
DOI: 10.1002/smj
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 rms 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 rm nancial
performance.
Therefore, CSR by organizations with highly nar-
cissistic CEOs may not have as positive or efcient
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 ndings 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 nancial 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 nancial 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 rms between the
years 1997 and 2012 inclusive, and excluded 24 pri-
vate rms for which no nancial data is available
in COMPUSTAT. Following prior research, we then
excluded 69 rms in highly regulated industries
such as nancial, insurance, and utilities. Firms in
highly regulated industries such as nancials are
not only subject to differences in their regulatory
environments that limit discretion of these rms’
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 rms and especially nancial
rms compared to others in the population. There
were signicant differences with respect to compar-
ison variables; p-values for ROA and CSR measures
ranged from 0.000 to 0.01 for both nancials specif-
ically and regulated industries as a whole when
compared with the broader population of rms.
We identied the CEO for every rm in 2007 and
included all rm-years in this time-frame for which
they were CEOs of their respective rms. We then
imposed four necessary lters on our data. First, we
omitted 15 CEOs who held temporary appointments
(e.g., interim, acting) because the effects on rms
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 rms for which data was not avail-
able in the KLD database. The nal 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 nancials
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 rms/CEOs
in the nal 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
rms/CEOs in the nal sample compared to others
in the broader population. There were no signi-
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.
Copyright © 2015 John Wiley & Sons, Ltd. Strat. Mgmt. J.,37: 262– 279 (2016)
DOI: 10.1002/smj
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 congurations with identi-
cal results. We proceeded to create a sample includ-
ing all rm-year observations from 2007 to 2012.
By testing our ndings 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 conrm 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 ination 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 nds 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 rst 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 of-
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 coefcient alpha relia-
bility (𝛼=0.95) (Nunnally, 1978). Moreover, the
expert raters demonstrated signicant 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
Copyright © 2015 John Wiley & Sons, Ltd. Strat. Mgmt. J.,37: 262– 279 (2016)
DOI: 10.1002/smj
Corporate Social Responsibility or CEO Narcissism? 269
avoiding rater exhaustion. More specically, 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 45second video samples (−0.66),
turns positive for 1 minute and 20 second video
samples (0.45) and grows signicantly to 0.71
with 2.5 minute video samples. It does not signi-
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 signicantly different between
45 seconds and 2.5minutes (p<0.01) but there was
no signicant 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 notied 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
signicant 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 nal 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. Specically, 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 signicant 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 signicant (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 nancial advisory
rm 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
Copyright © 2015 John Wiley & Sons, Ltd. Strat. Mgmt. J.,37: 262– 279 (2016)
DOI: 10.1002/smj
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 rms 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 prole (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 rst searched
Factiva for all articles that included the company
name and the rm/year combination in our sample.
Then, we isolated articles that cover philanthropic
events using the predened 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 identies 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 rm 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).
1As a robustness check, we also assess CSR using a weighted
indicies (Waddock and Graves, 1997) and nd results consistent
across both measures.
Performance
We examine performance using Return on Assets
(ROA), a common measure of rm 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
rm operations caused by CSR decisions made by
rms in the study.
Additionally, we included two common measures
of market performance for supplementary analysis:
Tobin’s Q(TQ), calculated by dividing the rm’s
market value by rm’s asset replacement costs
and Market Value Added (MVA), calculated by
subtracting capital (i.e., the debt and shareholders’
equity invested in the rm) from the equity market
valuation of the rm. Tobin’s Qprovides an approx-
imation of the stock market’s estimation of net
present value (Tobin and Barnard, 1968) and MVA
captures the ability of rms to maximize share-
holder value through efcient allocation and man-
agement of resources (Hillman and Keim, 2001).
Control variables
We control for CEO-, rm-, 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 inuence their ability to promote CSR
projects in their rms, 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
Copyright © 2015 John Wiley & Sons, Ltd. Strat. Mgmt. J.,37: 262– 279 (2016)
DOI: 10.1002/smj
Corporate Social Responsibility or CEO Narcissism? 271
political ideology, we examined political donations
made by executives for the 10years 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 identied 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 rm-specic conditions that might
inuence CSR intensity of a rm, 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 rm’s ROA in the previous time
period. Because large rms may face different
pressures to be involved in CSR, we controlled
for the size of the rm, 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 rm, 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-inated distribution
we specied a negative binomial distribution with
a log link function (Chatterjee and Hambrick,
2007). All other models are specied 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 rst show the results for our base control model
(Model 1). Hypothesis 1 posits that there will be a
positive relationship between CEO narcissism and
2Endogeneity control. We controlled for the possibility that
narcissistic CEOs might be drawn to rms 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 rm. They
included rm 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
rst 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 rm) 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 nd that results did not change signicantly if we omit the
endogeneity control.
Copyright © 2015 John Wiley & Sons, Ltd. Strat. Mgmt. J.,37: 262– 279 (2016)
DOI: 10.1002/smj
272 O. V. Petrenko et al.
Table 1. Descriptive statistics and correlation coefcients
Variable Means.d.1234567891011121314151617181920
1. CSR 0.34 4.14
2. CSR concerns 4.35 2.87 −0.38
3. CSR strengths 4.69 4.02 0.75 0.31
4. CPMP 0.00 1.00 0.08 0.01 0.08
5. ROA 0.05 0.07 0.24 −0.07 0.20 0.06
6. Tobin’s Q1.82 1.21 0.19 −0.18 0.07 0.09 0.38
7. Market value added 12.27 35.99 0.26 0.08 0.33 0.07 0.38 0.54
8. Narcissism 4.41 1.31 0.18 −0.19 0.05 0.07 0.08 0.20 0.03
9. Age 54.77 5.89 −0.08 0.27 0.11 −0.08 0.03 −0.16 −0.06 −0.10
10. Duality 0.65 0.47 0.03 0.15 0.14 0.03 0.02 −0.07 −0.05 −0.04 0.16
11. CEO ownership 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. Tenure 5.55.54 0.08 −0.10 0.01 0.01 0.03 0.03 −0.06 0.09 0.36 0.23 0.12
13. Unabsorbed slack 3.17 17.49 −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. Size (log of sales) 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. Endogeneity control 3.96 2.15 0.06 −0.01 0.05 0.05 0.05 0.02 0.03 0.04 −0.33 0.42 −0.19 0.03 −0.05 −0.01
16. CEO short-term pay focus 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. CEO long-term pay focus 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 0.01 0.12 0.04 −0.75
18. CEO liberalism 0.36 0.28 0.02 −0.15 −0.08 0.08 −0.02 0.14 0.01 0.19 0.10 −0.10 0.10 0.13 0.03 −0.07 0.01 0.01 −0.04
19. Advertising intensity 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 intensity 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. Capital intensity 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 0.05 0.05 −0.08 −0.10 −0.05
Correlations greater than 0.05 are signicant at p<0.05.
Copyright © 2015 John Wiley & Sons, Ltd. Strat. Mgmt. J.,37: 262– 279 (2016)
DOI: 10.1002/smj
Corporate Social Responsibility or CEO Narcissism? 273
Table 2. Effects of CEO narcissism on corporate social responsibility (CSR) and corporate philanthropy media prole
(CPMP) (GEE analyses)
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6
DV: CSR DV: CSR DV: Strengths DV: Conc erns 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
Wal d Chi26,374*** 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 prole for rm corpo-
rate philanthropy and the level of media prole
for rm corporate philanthropy. Model 5 shows
that CEO narcissism is positively related to cor-
porate philanthropy media prole (0.35, p<0.01)
implying that narcissistic CEOs engage in higher
prole corporate philanthropy. Additionally, Model
6 shows that CEO narcissism strengthens the rela-
tionship between previous corporate philanthropy
high prole and subsequent corporate philanthropy
high prole (0.05, p<0.01), indicating that rms
with more narcissistic CEOs are more likely to
continue to generate higher prole 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 rm specic lev-
els of CPMP. Models 5 and 6 in Table 2 provide
support for Hypothesis 2.
3We 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 rm or industry levels resulting from
economic or business crises.
Copyright © 2015 John Wiley & Sons, Ltd. Strat. Mgmt. J.,37: 262– 279 (2016)
DOI: 10.1002/smj
274 O. V. Petrenko et al.
Figure 1. The relationship between corporate philan-
thropy media prole (CPMP) at time tand 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 rm performance measured as ROA (0.02,
p<0.05). Our result has practical signicance
because it indicates ROA will become a nominal
2.16 percent higher for rms 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 rms 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
rm 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 ndings. 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 Chi2786*** 964*** 14,333*** 16,099***
Standard errors are in parentheses.
*p<0.05; **p<0.01; ***p<0.001
Copyright © 2015 John Wiley & Sons, Ltd. Strat. Mgmt. J.,37: 262– 279 (2016)
DOI: 10.1002/smj
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 rms with more narcissistic
CEOs when they have more CSR strengths or
less CSR concerns than for the rms 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 Qand Market Value Added—as additional
validation for our ndings 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
nancial 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 rms can be sig-
nicantly affected by CEO narcissism. Consistent
with the logic that narcissistic CEOs crave attention,
we have also theorized and shown that rms with
more narcissistic CEOs have higher prole corpo-
rate philanthropy than the rms of their less narcis-
sistic peers, and that they also continue to engage in
high prole corporate philanthropy once they suc-
ceed at it. Finally, we have shown that the positive
relationship between CSR and rm performance is
weaker for rms 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 ndings 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 rm 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
Copyright © 2015 John Wiley & Sons, Ltd. Strat. Mgmt. J.,37: 262– 279 (2016)
DOI: 10.1002/smj
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 rm 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 rms with more narcissistic
CEOs may be less strategic in terms of nancial
performance in part because of a resulting lack
of focus in CSR congurations. 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 rm 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,
condence, 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
Copyright © 2015 John Wiley & Sons, Ltd. Strat. Mgmt. J.,37: 262– 279 (2016)
DOI: 10.1002/smj
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 inuence 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 nancial performance. Strategic
Management Journal 30: 895– 907.
Clark GL, Hebb T. 2004. Corporate engagement: the fth
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 nancial performance. Strategic
Management Journal 26: 827– 840.
Cornell B, Shapiro AC. 1987. Corporate stakeholders and
corporate nance. 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 prots. New York Times Magazine 13
September: 32–33.
Frooman J. 1999. Stakeholder inuence 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 reection 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 prots 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.
Copyright © 2015 John Wiley & Sons, Ltd. Strat. Mgmt. J.,37: 262– 279 (2016)
DOI: 10.1002/smj
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. For b e s 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 rm-specic 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 nancial performance: correlation or
misspecication? Strategic Management Journal 21:
603–609.
McWilliams A, Siegel D. 2001. Corporate social respon-
sibility: a theory of the rm 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 ve 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 rms 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 ve-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 rm 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 reects: 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 ndings. 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
Copyright © 2015 John Wiley & Sons, Ltd. Strat. Mgmt. J.,37: 262– 279 (2016)
DOI: 10.1002/smj
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 inuences 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 rm 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 nancial
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 rms.
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 nancial 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 ination. 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 clarication, 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—nancial performance link. Strategic
Management Journal 18: 303– 319.
Waldman DA, Siegel D. 2008. Dening 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 rms’ strategic persistence
in response to low rm 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 conict 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.
Copyright © 2015 John Wiley & Sons, Ltd. Strat. Mgmt. J.,37: 262– 279 (2016)
DOI: 10.1002/smj
