Introduction
Organizational scholarship on narcissism typically conceptualizes narcissism as a set of subclinical personality traits that reflect an inflated sense of self, excessive need for admiration, grandiosity, overconfidence, entitlement, and self-absorption (Blair et al., Reference Blair, Hoffman and Helland2008; Campbell & Campbell, Reference Campbell and Campbell2009; Campbell & Foster, Reference Campbell, Foster, Sedikides and Spencer2007; Paulhus, Reference Paulhus2014; Raskin & Terry, Reference Raskin and Terry1988; Rosenthal & Pittinsky, Reference Rosenthal and Pittinsky2006). The prevalence of narcissistic behavior has increased over time in individualistic societies (e.g., Foster et al., Reference Foster, Campbell and Twenge2003; Twenge et al., Reference Twenge, Konrath, Foster, Campbell and Bushman2008), and its prevalence in organizational leadership positions has been lamented for decades. Indeed, some scholars have gone so far as to conclude that, “It is probably not an exaggeration to state that if individuals with significant narcissistic characteristics were stripped from the ranks of public figures, the ranks would be perilously thinned” (Post, Reference Post1993, p. 99). Further, in the face of numerous high-profile corporate leaders (e.g., Elon Musk, Jeff Bezos, Shai Agassi) and scandals (e.g., at Enron, Lehman Brothers, and the Royal Bank of Scotland; Sözen & Basım, Reference Sözen and Basım2022), the nature and consequences of narcissistic leadership have been extensively documented in popular media (Ben-Hur & Bolton, Reference Ben-Hur and Bolton2018; Campbell & Crist, Reference Campbell and Crist2020; Choi & Phan, Reference Choi and Phan2022; Gruda & Hanges, Reference Gruda and Hanges2023; O’Connell, Reference O’Connell2021; Smiech, Reference Smiech2020). This in turn has fueled a general distrust of leaders in organizations and brought attention to the need to understand how organizational systems and practices can contribute to such dynamics.
Alongside this, there has also been growing interest in the topic of leader narcissism among organizational scholars and practitioners. Academic interest in leader narcissism also stems from the fact that it has been found to be a reliable predictor of detrimental outcomes at different levels of abstraction. Specifically, at high levels, narcissism is problematic (a) for individuals (e.g., leading to counterproductive work behaviors, Penney & Spector, Reference Penney and Spector2002; and leader ineffectiveness, Grijalva et al., Reference Grijalva, Harms, Newman, Gaddis and Fraley2015), (b) for teams (e.g., contributing to reduced coordination and performance, Grijalva et al., Reference Grijalva, Maynes, Badura and Whiting2020), and (c) for organizations (e.g., in the form of fluctuating organizational performance, Chatterjee & Hambrick, Reference Chatterjee and Hambrick2007; and failure to heed others’ advice, Zhu & Chen, Reference Zhu and Chen2015). For all these reasons then, the study of narcissism is very much in vogue (Lasch, Reference Lasch and Wilkinson2019).
Yet, although extensive research has examined the effects of leader narcissism on organizational outcomes (see Braun, Reference Braun2017, for a systematic review), far less attention has been directed toward factors that shape leader narcissism. In part this is because dominant thinking assumes that narcissism is a personality trait shaped by early developmental influences such as parental income or parenting style (e.g., Bergman et al., Reference Bergman, Westerman and Daly2010; Martin et al., Reference Martin, Côté and Woodruff2016). With these factors being largely beyond the remit or control of organizations, this perspective implies that organizations have little ability or responsibility to manage or influence leader narcissism in the workplace.
At the same time, though, this perspective is somewhat surprising considering evidence that narcissism can change during adulthood and can be influenced both by entry into the workforce and by early experiences within the workplace (e.g., Grapsas et al., Reference Grapsas, Brummelman, Back and Denissen2020; Heyde et al., Reference Heyde, Wille, Vergauwe, Hofmans and De Fruyt2024). In line with this observation, psychologists across various subdisciplines have observed that expressions of personality variables are not simply a fixed input into organizational life but also can be shaped by interactions and experiences in social and organizational contexts (e.g., Haslam et al., Reference Haslam, Jetten, Reynolds and Reicher2013; Heyde et al., Reference Heyde, Wille, Vergauwe, Hofmans and De Fruyt2024; Li et al., Reference Li, Li, Feng, Wang, Zhang, Frese and Wu2020; Reynolds et al., Reference Reynolds, Turner, Branscombe, Mavor, Bizumic and Subašic2010; Roberts et al., Reference Roberts, Caspi and Moffitt2003; Steffens et al., Reference Steffens, Chong and Haslam2022). Nevertheless, apart from a few studies that have investigated the relationships between uncertainty and corporate culture and leader narcissism in organizations (e.g., Nevicka et al., Reference Nevicka, De Hoogh, Van Vianen and Ten Velden2013; O’Reilly et al., Reference O’Reilly, Chatman and Doerr2021), most organizational scholarship has construed leader narcissism as an input rather than as a context-dependent view of the self (Onorato & Turner, Reference Onorato and Turner2004) that might also, at least in part, be an output of organizational life. Consequently, questions about the ways in which the character and behavior of narcissistic leaders might be influenced by organizational, human resource (HR) practices are underexplored and hence left largely unanswered.
We therefore call on I-O psychologists and HR scholars and practitioners to expand the lens of leader narcissism by examining how the very systems and practices they design and implement may contribute to its expression. For instance, scholars have theorized that self-enhancing experiences—such as being selected for or promoted to a leadership role—have the potential to increase an individual’s view of themself as special and superior to others (e.g., Campbell & Campbell, Reference Campbell and Campbell2009). Given the consensus among scholars that HR practices (e.g., selection, training, and compensation/rewards) can effectively influence and shape the effort, behaviors, performance, and interactions of employees (Bartel, Reference Bartel2004; Huselid, Reference Huselid1995; Jiang et al., Reference Jiang, Lepak, Hu and Baer2012; Sun et al., Reference Sun, Aryee and Law2007; Wright et al., Reference Wright, Dunford and Snell2001, Reference Wright, Gardner, Moynihan and Allen2005), we argue that these same practices may unintentionally amplify narcissistic and self-interested characteristics and behaviors among leaders, and thus, deserve a critical analysis. It is this possibility that we focus on in this paper.
In what follows we therefore seek to redirect the focus on leader narcissism inward to organizations—with a view to exploring how different HR practices might either stimulate or suppress destructive forms of leader narcissism at all levels of organizations. So, although certain HR practices (e.g., the provision of unrestricted opportunities for personal development) are less likely to contribute to people’s narcissism or superiority complex, we note that other practices (e.g., the provision of exclusive high-potential development programs, Finkelstein et al., Reference Finkelstein, Costanza and Goodwin2018) may encourage narcissism by cultivating a sense that a particular employee is “special” and has extraordinary qualities that make them uniquely fit to be a leader. Relatedly, we suggest that particular selection and promotion practices can cultivate narcissism and hubris among leaders by signaling to those who come out on top that they are in rare and elite company (Berger et al., Reference Berger, Osterloh, Rost and Ehrmann2020). Similarly, compensation and remuneration structures that create a significant vertical pay gap between leaders and followers can signal to those leaders that they are of a different breed—set apart and different from everyday workers (Peters et al., Reference Peters, Fonseca, Steffens and Hauser2024; Steffens et al., Reference Steffens, Haslam, Peters and Quiggin2020b). In these and other ways, then, we suggest that some HR practices can serve to stimulate leader narcissism.Footnote 1 But equally, desisting from these practices, as well as pursuing alternative ones, may suppress leader narcissism and its destructive effects. We offer this specific perspective to help start a broader conversation on the role that organizations play in influencing leader narcissism—one that we hope I-O psychology and HR scholars and practitioners will engage in, build on, and extend.
To provide a theoretical framework for our analysis, we draw on self-categorization theory (SCT; Hogg & Terry, Reference Hogg and Terry2000; Turner et al., Reference Turner, Hogg, Oakes, Reicher and Wetherell1987). In particular, we argue that some HR practices (e.g., competitive leader selection, elevated executive pay) are highly personalizing in so far as they serve to promote a person’s sense of their special personal identity (or “me-ness”) over their collective social identity (or “we-ness”), while at the same time other HR practices (e.g., rotational leader selection, team rewards) are depersonalizing in so far as they are groupcentric and encourage leaders to see themselves in terms of a collective identity that they share with other organizational members. Furthermore, we argue that hyperpersonalizing HR practices are more likely than depersonalizing HR practices to encourage narcissism on the part of organizational leaders. Conversely, we argue that compared to hyperpersonalizing HR practices, depersonalizing HR practices are more likely to encourage leaders to think and act in group-oriented ways that serve the needs of other organizational members. Through this novel lens, we invite organizational scholars and practitioners to explore the capacity for HR practices to be either “me-sources” or “we-sources” and, through this, to shape leader characteristics and behaviors that are either destructive or protective of healthy organizational cultures.
Self-categorization theory (SCT)
Identity relates to the ways in which an individual views or defines the self in ways that bear upon their interaction with the world (Chryssochoou, Reference Chryssochoou2003; Cornelissen et al., Reference Cornelissen, Haslam and Balmer2007). Identity can be defined in a range of ways but, critically, can be understood to vary along a continuum where, at one extreme, the self is defined in terms of factors that are unique to a person (e.g., their personality and unique tastes), and, at the other extreme, the self is defined by factors that are shared with others (e.g., group memberships, roles, and norms; Alvesson & Wilmott, Reference Alvesson and Willmott2002; Ashforth, Reference Ashforth2000; Turner, Reference Turner and Tajfel1982, Reference Turner and Lawler1985). Critically, identity is not fixed but varies with social context (Turner et al., Reference Turner, Oakes, Haslam and McGarty1994). In particular, individualistic contexts and cultures tend to emphasize personal achievement and accountability in ways that lead people to define themselves in terms of a personal identity as a unique individual (as “I” and “me”), whereas collectivistic contexts and cultures tend to emphasize group-level factors that make salient a social identity in which the self is defined as part of a group or collective (as “we” and “us”; Tajfel & Turner, Reference Tajfel, Turner, Worchel and Austin1986). SCT speaks to the social–cognitive process involved in this operation of self, focusing especially on the ways in which people come to define the self (i.e., self-categorize) in different ways across diverse contexts (Turner et al., Reference Turner, Hogg, Oakes, Reicher and Wetherell1987, Reference Turner, Reynolds, Haslam, Veenstra, Postmes and Jetten2006). More specifically, it sets out hypotheses related to three keys aspects of self-functioning (Haslam, Reference Haslam2004): (a) the process through which individuals come to define themselves in terms of a given personal or social identity, (b) the contextual process that amplify or diminish the salience of a given personal or social identity, and (c) the consequences of defining the self in terms of a given personal or social identity. When, for example, will an employee define themselves as a member of a particular team rather than as a unique individual? What will make them more or less likely to do so? And how will this affect what they think, feel, and do?
Regarding the first of these elements, SCT argues that individuals internalize a meaningful group membership as part of the self (e.g., of “us, software engineers”) through a process of depersonalization, whereby the self is defined in terms of the shared attributes that define category membership (e.g., being highly analytical, problem-solvers, dressing casually). To be clear, depersonalization does not involve a loss of self but rather a redefinition of self that aligns it with a salient social identity and salient aspects of the group membership (e.g., values, norms, goals) to which that identity relates (Turner, Reference Turner and Tajfel1982). Indeed, it is this capacity to internalize a group membership and engage in “we-thinking” that provides a psychological platform for group behavior in which people act in the interests of others rather than simply themselves as individuals (Richerson et al., Reference Richerson, Baldini, Bell, Demps, Frost, Hillis, Mathew, Newton, E., Naar, Newson, Ross, Smaldino, Waring and Zefferman2016; Turner, Reference Turner and Tajfel1982). As Haslam and colleagues put it: “We can only act as group members because, and to the extent that, we are able to think about ourselves as ‘we’ and not just ‘I’” (Haslam et al., Reference Haslam, Reicher and Platow2011, p. 52).
Concerning the second element, SCT outlines two major determinants of self-categorization in terms of a given social identity: the perceived fit of a particular self-categorization and an individual’s readiness to use it (Oakes et al., Reference Oakes, Haslam and Turner1994; Turner et al., Reference Turner, Oakes, Haslam and McGarty1994). Fit has two components: normative and comparative. Normative fit captures the extent to which the characteristics and behaviors associated with a particular group membership match a perceiver’s expectations about what it means to belong to that category (e.g., using jargon and engaging in practices that define “us” as software engineers). Comparative fit captures the extent to which categorizing self and others in terms of shared group membership maximizes within-group similarities and between-group differences (e.g., so that the things that define “us” are distinct from the things that define “them”). Perceiver readiness refers to a person’s readiness to use a particular social category as the basis for self-definition. This is driven by both previous experiences and present expectations, goals, needs, and values. For example, a person may be more likely to self-categorize as a software engineer if this is the only job they have ever had and if they are attending a conference on software engineering.
Furthermore, whether the self is defined in terms of a specific social identity has critical implications for individual and group behavior (Hogg & Terry, Reference Hogg and Terry2000). This is because when people define themselves in terms of a shared social identity, group interests become self-interests, and people are motivated to act in ways that align with the norms and goals of the group. As a result, shared social identity has been argued to be a basis for people to mutually influence one another (Turner, Reference Turner1991), to cooperate with one another and coordinate their behavior to achieve collective (group or organizational) goals (Tyler & Blader, Reference Tyler and Blader2003), and to go the “extra-mile” by engaging in organizational citizenship behavior (Lee et al., Reference Lee, Park and Koo2015).
Importantly, this dynamic also shapes the experience and expression of individual differences (Turner et al., Reference Turner, Reynolds, Haslam, Veenstra, Postmes and Jetten2006). In particular, just as the salience of social identities changes, so too does the salience of personal identities (e.g., personality). This means that whether or not people see themselves and act in terms of their unique personality (e.g., as a socially dominant, entitled, narcissist) will vary as a function of the context in which they are operating and the cues within that context. Among other things, this means that particular personality characteristics will only come to the fore in contexts where individuals come to see themselves in unique terms—and are encouraged to do so (e.g., as a result of incentive structures, organizational culture, and ideology). In this sense, the analysis provided by SCT aligns with the core predictions of trait activation theory, which proposes that traits need to be activated by context in order to find expression (Tett et al., Reference Tett, Simonet, Walser, Brown, Christiansen and Tett2013, Reference Tett, Toich and Ozkum2021). Nevertheless, in SCT this analysis is couched within a broader analysis of the self-system that has far broader ramifications (e.g., as the basis for an analysis of social influence and leadership; Haslam et al., Reference Haslam, Reicher and Platow2020; Turner, Reference Turner1991).
In line with this analysis, we posit that some HR practices can be hyperpersonalizing in leading individuals to see themselves as individuals and serving to create a competitive climate that privileges a leader’s sense of their special personal identity. This can also be reinforced by organizational cultures and ideologies that implicitly or explicitly promote the “romance of leadership,” in ways that glorify leaders as heroes and attribute collective success to their singular efforts (Haslam et al., Reference Haslam, Alvesson and Reicher2024; Meindl et al., Reference Meindl, Ehrlich and Dukerich1985). Indeed, these do not so much “activate” narcissism as actively cultivate and reinforce it (e.g., through organizational cues that make this expression of personal identity appear to be both desired and desirable). In contrast, other HR practices treat individuals as group members and serve to create cooperative climates that valorize the social identity that leaders share with other organizational members. In these contexts, narcissism may be less likely to rear its head, both because it is incongruent with group norms and because the environment fails to incentivize this particular expression of superior personal identity. We further argue that hyperpersonalizing HR practices are more likely than depersonalizing HR practices to enhance (vs. suppress) leader narcissism, whereas depersonalizing HR practices are more likely than the former to facilitate leaders’ engagement in group-oriented activities. Below, we discuss three key HR practices that might fuel leader narcissism in this way: (a) leader promotion and selection procedures, (b) training and development programs, and (c) compensation and rewards. Further, we argue that depersonalizing HR can encourage leaders to think and behave in collective-oriented ways and advance group interests.
Human resource practices and leader narcissism
Leader selection and promotion practices
Hyperpersonalizing practices
Selection methods that are competitive in nature typically involve making interpersonal comparisons highly salient with the aim of selecting the most competent leaders within a given organization or organizational unit. This is a common practice across both private and public sectors. Nevertheless, we argue that hypercompetitive leadership selection practices have the capacity to catalyze and amplify narcissistic tendencies for those who are selected into leadership positions. Illustrative leadership selection methods include: (a) those that make salient each applicant’s personal identity vis-à-vis the personal identities of other candidates and/or other members of the organization (e.g., by requiring applicants to highlight their superiority relative to others), (b) those that ask applicants to compete with others in a set of bespoke competitions beyond the scope of their normal work in order to secure advancement (i.e., requiring them to “beat” those who might otherwise be collaborators), and (c) those that require participants to invest considerable time and effort in the process of curating their personal identities (e.g., requiring employees to create portfolios that showcase their unique abilities and accomplishments).
Hypercompetitive selection methods of this form are often used to select the person who is most skilled or competent for a given leadership position or role. More often than not, these identify a “winner,” and this is the person with the strongest record of performance or fit with the role in question (in line with implicit leadership theories; see Lord et al., Reference Lord, Epitropaki, Foti and Hansbrough2020, for review). This is apparent in a range of common practices—including executive searches, elections, competency-based assessments, assessment centers, succession planning, and talent development programs. These approaches can create two problems: (a) attracting biased samples of individuals with high levels of narcissism and (b) validating and encouraging self-serving and narcissistic behavior.Footnote 2 Moreover, because narcissists seek social validation, these problems are likely to become more pronounced the more publicity the selection process and outcomes attract. Indeed, because overt narcissism is positively related to competitiveness (Luchner et al., Reference Luchner, Houston, Walker and Houston2011; Raskin & Terry, Reference Raskin and Terry1988), the competitive nature of many modern selection processes can make these especially alluring for narcissists. At the same time, success in a hypercompetitive selection process will reward narcissistic tendencies and validate overconfident individuals’ belief that they are superior to others and in some ways “special” or extraordinary (Berger et al., Reference Berger, Osterloh, Rost and Ehrmann2020; Haslam et al., Reference Haslam, McGarty, Brown, Eggins, Morrison and Reynolds1998).
Similarly, the power that flows from success in a competitive selection process can lead candidates to become overconfident in ways that have detrimental effects on organizational performance (Kipnis, Reference Kipnis1972; Vitanova, Reference Vitanova2021). Hypercompetitive selection practices can also facilitate self-serving biases (Miller & Ross, Reference Miller and Ross1975) that lead people to attribute past achievements to their own personal ability and effort at the same time that they downplay the contributions of others (Ross, Reference Ross and Bekowitz1977; Sandel, Reference Sandel2020). For those who succeed, leadership selection processes can thus be a powerful affirmation of their inherent superiority to others and of their entitlement to leadership positions.
SCT provides a useful lens through which to understand the influence of hypercompetitive selection practices on leader narcissism. In particular, the principle of comparative fit suggests that the salience of social identity will be reduced to the degree that the interaction context serves to highlight differences between individuals rather than similarities (Haslam, Reference Haslam2004; Oakes et al., Reference Oakes, Haslam and Turner1994). Accordingly, when key aspects of a selection process highlight individual attributes and interpersonal differences, successful candidates are more likely to define themselves in terms of their personal identity. At the same time, they are less likely to define themselves in terms of a shared social identity (Haslam et al., Reference Haslam, McGarty, Brown, Eggins, Morrison and Reynolds1998). Following intraindividual hypercompetitive processes, this means that when individuals are selected to assume leadership roles or positions of power, they are more likely to act selfishly in line with their personal values, beliefs, and interests, and less in line with the values, beliefs, and interests of the collective. As Kipnis (Reference Kipnis1972) found in his classic research on the corrupting influence of power, this means that when a selection process confers sovereignty on a leader, concern for others is often the first casualty.
Depersonalizing practices
Evidence suggests that individuals who recognize the role that luck or randomness has played in their success are more likely to express humility and act prosocially toward others than those who believe that their success is an entirely justified reflection of their own ability and effort (Bartlett & DeSteno, Reference Bartlett and DeSteno2006; Berger et al., Reference Berger, Osterloh, Rost and Ehrmann2020; Frank, Reference Frank2016). This also means that recognizing randomness in, or injecting a degree of randomness into, leadership selection can help to attenuate perceptions that leadership roles are “made for” and necessarily occupied by extraordinary individuals. Yet although random leadership selection methods have been used in many spheres throughout history (e.g., to appoint magistrates, judges, and spiritual leaders; Berger et al., Reference Berger, Osterloh, Rost and Ehrmann2020; Duxbury, Reference Duxbury2002; Hansen, Reference Hansen1991), this practice is generally rather uncommon in organizations.
Unsurprisingly, perhaps, scholars and practitioners have questioned the appropriateness of random selection for leadership roles and tend to eschew practices of this form. Indeed, in the spirit of Taylor and Münsterberg, they often see their role as one that necessarily entails removing as much randomness as possible from the selection process (Moskowitz, Reference Moskowitz1977). This is based on the assumption that “you cannot randomly allocate leadership responsibility and expect the interests of justice or society to be well-served” (Emler & Hogan, Reference Emler, Hogan, Kurtines and Gerwitz1991, p. 86).
There is evidence, however, that, under some conditions, building randomness into a selection process—and making this explicit—can help to temper the self-serving behavior of narcissistic leaders and therefore be beneficial for the groups they lead. For example, in a series of experimental studies, Haslam and colleagues found that team performance increased when leaders were randomly selected rather than chosen on the basis of systematic competition (Haslam et al., Reference Haslam, McGarty, Brown, Eggins, Morrison and Reynolds1998). Along related lines, Berger and colleagues (2020b) conducted an experiment that explored the benefits of using a partly random selection procedure—that combined competitive and random selection methods—to select leaders from a pool whose members had been identified as high-performing or highly competent. The researchers found that overconfident leaders who were selected using partly random (vs. entirely competitively) selection procedures acted less selfishly by claiming less money for themselves and allocating more money to their subordinates (Berger et al., Reference Berger, Osterloh, Rost and Ehrmann2020b).
In this sense, partly random selection practices can function as a depersonalizing HR practice that encourages group members to define themselves in terms of a collective identity (as “we” rather than “me”) and thereby avoids intragroup division (Haslam et al., Reference Haslam, McGarty, Brown, Eggins, Morrison and Reynolds1998). This is because depersonalizing leadership selection encourages candidates to define themselves in terms of a social identity that they share with other group or organizational members—with the result that once they are selected for a leadership role, they are more likely act in ways that promote the interests of that group or organization. There is evidence too that this strategy can support the interests of diversity and inclusion more broadly (e.g., by reducing gender bias; Berger et al., Reference Berger, Osterloh and Rost2020a).
That said, it is important to note that random or partly random selection practices are not always appropriate and will not solve all the problems associated with leader narcissism. Instead, they are likely to be most appropriate when groups or teams have clearly defined shared goals, when they are self-managed and have decision-making autonomy, and when they would otherwise have a strong sense of shared social identity in the absence of an appointed leader (Haslam et al., Reference Haslam, McGarty, Brown, Eggins, Morrison and Reynolds1998). Also, we note that effective implementation of random leader selection practices in organizations begins by drawing leaders randomly from a pool of motivated and capable candidates. It also needs to be combined with other methods—which we discuss below—that build support for the policy and help to create a pool of potential leaders (e.g., shared leadership). In short, we would theorize that there is equifinality to the goal of reducing leader narcissism, and that random leader selection is but one path to this goal (Doty et al., Reference Doty, Glick and Huber1993; Hunter et al., Reference Hunter, Doctor, Allen and Ligon2023).
If random leader selection is seen to be too drastic or too radical, there are therefore other leadership selection methods that might serve the same end of seeking to suppress narcissism. For example, one viable alternative pathway is provided by shared leadership. Shared leadership can be defined as an emergent team property associated with shared responsibility and distributed influence across multiple team members (Carson et al., Reference Carson, Tesluk and Marrone2007). In contrast to leader-centric perspectives that assume top-down influence from a single designated leader to subordinates (vertical leadership), shared leadership recognizes and promotes lateral or peer-to-peer influence that takes place in the presence or absence of a designated leader (Pearce & Conger, Reference Pearce and Conger2003). Importantly, models of shared leadership reduce leader–follower distinctions because they recognize that team members can assume leadership roles at any point in time (Mehra et al., Reference Mehra, Smith, Dixon and Robertson2006; Nicolaides et al., Reference Nicolaides, LaPort, Chen, Tomassetti, Weis, Zaccaro and Cortina2014). In this sense, shared leadership encourages members to see themselves and others as part of a collective because everyone has a responsibility to serve and represent the team (Wang et al., Reference Wang, Waldman and Zhang2014).
By emphasizing collective over individual needs, shared leadership should generally tend to reduce narcissistic tendencies on the part of group members and formal leaders who are seeking personal admiration and validation. Moreover, by helping to create teams that are leaderful it can be a powerful stimulus for teamwork and team effectiveness (Steffens et al., Reference Steffens, Fransen, Haslam, Haslam, Fransen and Boen2020a). This is something that has been well documented in a range of domains—from sport and health (e.g., Fransen et al., Reference Fransen, Van Puyenbroeck, Loughead, Vanbeselaere, De Cuyper, Broek and Boen2015, Reference Fransen, Haslam, Steffens, Peters, Mallett, Mertens and Boen2020, Reference Fransen, Cruwys, Haslam, Iserbyt, Seghers, Vanderlinden, van Uffelen, Verbaanderd and Boen2022; Mertens et al., Reference Mertens, Boen, Steffens, Cotterill, Haslam and Fransen2020) to business and government (D’Innocenzo et al., Reference D’Innocenzo, Mathieu and Kukenberger2016; Nicolaides et al., Reference Nicolaides, LaPort, Chen, Tomassetti, Weis, Zaccaro and Cortina2014; Wang et al., Reference Wang, Waldman and Zhang2014; Zhu et al., Reference Zhu, Liao, Yam and Johnson2018). In this vein, agencies such as the National Aeronautics and Space Administration (NASA) recognize that the success of complex and high-risk missions (e.g., exploring Mars) typically requires a shared leadership approach that draws on different members’ unique expertise at different points in time (as opposed to hierarchical and “tight-fisted control”; Mulhearn et al., Reference Mulhearn, McIntosh, Gibson, Mumford, Yammarino, Connelly, Day and Vessey2016). Hence, as part of the selection process, team members complete a ropes course that involves changing positions after each challenge or obstacle faced. As one NASA representative noted when reflecting on the people they are looking for in this process, “[We] want to see if they are good leaders as well as good followers” (Mulhearn et al., Reference Mulhearn, McIntosh, Gibson, Mumford, Yammarino, Connelly, Day and Vessey2016, p. 9).
The process of embedding shared leadership is often supported by social network analysis that draws on the experiences and perceptions of all members of a given network (e.g., a specific team or organization; Fransen et al., Reference Fransen, Haslam, Steffens, Peters, Mallett, Mertens and Boen2020; Mayo et al., Reference Mayo, Meindl, Pastor, Pearce and Conger2003). By providing insight into the degree to which all members influence one another, these can also help to create a consensual understanding that ultimately allows leaders to be seen as serving their group in ways that also minimize friction within it. That said, a shared leadership approach is not a silver bullet and is most appropriate when work is highly complex and requires high levels of interdependence (D’Innocenzo et al., Reference D’Innocenzo, Mathieu and Kukenberger2016; Nicolaides et al., Reference Nicolaides, LaPort, Chen, Tomassetti, Weis, Zaccaro and Cortina2014; Wang et al., Reference Wang, Waldman and Zhang2014). However, in general, leader selection that embeds elements of shared leadership can help to nip narcissism in the bud by reinforcing leaders’ sense that they have been appointed by the group and for the group.
Another approach that can achieve similar ends is rotating leadership. The goal of this is to give all qualified members of a group or team the opportunity to lead the group at some point. Precisely when each group member is given this opportunity can be decided randomly, systematically (e.g., according to a rota), or tactically (e.g., as a function of suitability, need, or fit to the specific task at hand). In sport, this approach was pioneered by the coach of the Australian Women’s Hockey team (the Hockeyroos), Ric Charlesworth—whose teams went on to win nine major international tournaments including gold medals at the 1996 and 2000 Olympic Games. For Charlesworth, this strategy was partly about ensuring that no one leader got too big for their boots; but, more generally, it was a strategy designed to encourage everyone to see themselves as a leader and to contribute to the team on this basis (Charlesworth, Reference Charlesworth2001).
Yet, although there is evidence that this practice is used widely in organizations when selecting leaders to take part in leadership development programs, it is less commonly used as a method for appointing them to leader roles. Nevertheless, research by Erez et al. (Reference Erez, Lepine and Elms2002) observed that teams with rotating leaders outperformed those in which leaders were either chosen on merit or left to emerge naturally. Members of teams with rotating leaders were also more cooperative and reported having a greater sense of voice and job satisfaction. Their research did not explore the role of narcissism in these effects, but it seems plausible that the positive outcomes were attributable, at least in part, to the fact that leader rotation held egregious self-promotion in check.
Leadership training and development practices
Hyperpersonalizing practices
The training of leaders is of critical importance for organizations (Lacerenza et al., Reference Lacerenza, Reyes, Marlow, Joseph and Salas2017). However, the theories of leadership that routinely inform training programs are leader-centric, focusing on the attributes and qualities of leaders as key drivers of group or organizational success (Bass & Bass, Reference Bass and Bass2008; Northouse, Reference Northouse2018). In this way, programs that are grounded in leader-centric theories (e.g., charismatic, directive, trait-based approaches) generally reinforce a “heroic” narrative that positions leaders as organizational messiahs and saviors (Haslam et al., Reference Haslam, Alvesson and Reicher2024; Steffens et al., Reference Steffens, Chong and Haslam2022). At the same time, the romance of leadership (Meindl et al., Reference Meindl, Ehrlich and Dukerich1985) encourages leaders and followers alike to fall prey to the fundamental attribution error by explaining collective success in terms of the personal characteristics and contributions of leaders rather than the abilities and efforts of the group as a whole. This tendency to fetishize leaders is also apparent in leadership theories that center on individuals’ extraordinary attributes, and these have been found to have particular appeal for leaders with narcissistic tendencies (Steffens et al., Reference Steffens, Chong and Haslam2022).
These problems have led researchers and commentators alike to observe that leadership training can easily promote narcissism among those who are singled out to receive it (Tourish, Reference Tourish2013; Westerman et al., Reference Westerman, Bergman, Bergman and Daly2012). Scholars have also highlighted the perversity of some high-profile organizations being lauded for their leadership programs at the very same time that they been found to be engaging in wholesale fraud and corruption (Kellerman, Reference Kellerman2012).
Nevertheless, business executives consistently identify the development of leaders within their organizations as a key area of focus, and estimates of the global annual spend on leadership development suggest this is around $80 billion (Statifacts, 2025). In this context, it is common for organizations to identify and select a small group of employees to receive a large chunk of training expenditure (typically around 5% of the workforce) with a view to maximizing their potential as leaders within those organizations (Finkelstein et al., Reference Finkelstein, Costanza and Goodwin2018; Kwok et al., Reference Kwok, Shen and Brown2021). Indeed, the label “high potential” that often accompanies this selection process can itself become an implicit extrinsic reward—signaling a person’s elevated status relative to their peers (Kohn, Reference Kohn1993).
Yet although they are seemingly innocuous, there is evidence that such attempts to cultivate high performance by singling individuals out for excessive praise and preferential treatment can backfire. One reason for this is that they can cultivate narcissism among recipients (Brummelman & Grapsas, Reference Brummelman, Grapsas and Brummelman2020). Another is that they overlook the importance of the group (Beer et al., Reference Beer, Finnström and Schrader2016), and in the process of fast-tracking the few, they may leave the many behind. Indeed, such programs may serve to demotivate those who are excluded from them as much as (if not more than) they motivate those who are selected (Steffens et al., Reference Steffens, Fonseca, Ryan, Rink, Stoker and Pieterse2018).
It is nevertheless the case that once they are on the “high-potential” track, employees are more likely to be promoted into more senior leadership roles (Kwok et al., Reference Kwok, Shen and Brown2021). Consistent with the Matthew Effect (Merton, Reference Merton1968)—the tendency for (dis)advantage to beget (dis)advantage in ways that compound accumulating disparity—those who experience early success are likely to go on to have more—not least because they are identified as appropriate beneficiaries of yet more rewards and resources. In light of this, it is hardly surprising that those beneficiaries come to see themselves as superior and “special.” Even if they are not narcissistic to start with, the pull of intoxicating self-love may be hard to resist.
Depersonalizing practices
One potential way to minimize the narcissism that can result from high-potential programs is to offer equivalent training support to all employees who meet relevant performance criteria and to ensure that these criteria are attainable by all (i.e., so that they do not exclude members of specific groups). A similar approach is taken by the global software company Adobe, whose employees have access to professional development benefits provided they are in good standing performance wise. This decreases the likelihood of programs creating division and of attendees coming to see themselves as superior to the rest of the organization. Another way to minimize developmental disparity in organizations is for programs to focus on recruiting any employees who have specific developmental needs and who are developmentally ready to take part in them. In this way, leader development can be focused on those who stand to gain most from it rather than those for whom it is a vanity project and who may need it least (Kwok et al., Reference Kwok, Shen and Brown2021). In the process, this should suppress the sense of entitlement and self-importance that may otherwise accompany elevation (de Cremer & van Dijk, Reference De Cremer and Van Dijk2005).
More generally, there may be value in rebranding high-potential programs so that these are marketed in more inclusive ways that have a more collective focus around people’s contribution to group achievement. For example, these can be styled as “leadership readiness programs” (Kwok et al., Reference Kwok, Shen and Brown2021) that flag an organization’s commitment to developing employees for the future rather than to elevating a particular individual above their peers.
To further suppress narcissistic tendencies, reframing should not be limited to simply the labeling of such programs but needs to extend to program content. In contrast to leader-centric programs that zero in on the heroic and special attributes of leaders, the concept of leadership needs to be framed as a group process in which “teamful” leaders work with and for the groups they lead to co-create organizational success (Haslam et al., Reference Haslam, Reutas, Bentley, McMillan, Lindfield, Luong, Peters and Steffens2023). Although such programs have been positively received by attendees and positively impact their team functioning—as evidenced in studies with employees in public-sector organizations as well as manufacturing and construction companies (Haslam et al., Reference Haslam, Reutas, Bentley, McMillan, Lindfield, Luong, Peters and Steffens2023; McMillan et al., Reference McMillan, Steffens, Haslam and Peters2025)—such programs may have little appeal to narcissists. But that is very much the point.
Leader compensation and rewards
Hyperpersonalizing practices
The compensation that senior leaders receive has surged in recent years, with the median total compensation for CEOs of the 500 largest US-based companies reaching $12.3 million in 2019. This is a 22% increase since 2015, and it equates to 191 times more than the median salary of employees (Equilar, 2020). Challenging the conclusion that this is good value for money, the link between CEO pay and firm performance is tenuous. Indeed, at the highest levels of CEO pay it is completely absent (Aguinis et al., Reference Aguinis, Martin, Gomez-Mejia, O’Boyle and Joo2018; Hambrick, Reference Hambrick2018). Moreover, within this select group, narcissistic CEOs receive especially high compensation (O’Reilly et al., Reference O’Reilly, Doerr, Caldwell and Chatman2014).
Although incentive and shareholder value models advance the notion that increasing CEOs’ pay ensures that they will be more effective in motivating employees to accomplish key organizational goals, evidence suggests that it reduces the ability of CEOs to be seen as acting in the interests of the group and to motivate their employees (Steffens et al., Reference Steffens, Haslam, Peters and Quiggin2020b). Why so? From an SCT perspective, this can be understood to be a consequence of the fact that huge pay disparities undermine perceptions of shared identity between leaders and followers. More particularly, SCT predicts that leaders will be most influential when they are perceived as embodying the norms, values, and interests of the group—that is, when they are seen as prototypical group members who represent “who we are” and “what we stand for” (Haslam et al., Reference Haslam, Reicher and Platow2011; Turner, Reference Turner1991). Huge compensation gaps do violence to such perceptions, in part because they increase the comparative fit and hence the salience of distinctions between “us” (employees) and “them” (executives). This weakens a leader’s ability to be seen as “one of us” and thereby undermines their influence and legitimacy (Haslam et al., Reference Haslam, Alvesson and Reicher2024; Steffens et al., Reference Steffens, Haslam, Peters and Quiggin2020b). By the same token, for CEOs themselves, the same disparities have been found to promote narcissistic qualities such as superiority and entitlement (Chatterjee & Hambrick, Reference Chatterjee and Hambrick2007).
Beyond extreme pay, the trappings of executive leadership do little to encourage modesty and groundedness. On the contrary, private planes, company limousines, chauffer services, exclusive club memberships, executive lounges, and luxury office suites all serve to limit employee–executive interactions and exacerbate a sense of physical and psychological distance between leaders and the led (Rajan & Wulf, Reference Rajan and Wulf2006). As Jordan Belfort (Reference Belfort2007) observes in The Wolf of Wall Street: “If I earn a million dollars a week and the average American earns a thousand dollars a week, then when I spend 20 thousand dollars on something it’s the equivalent of the average American spending 20 dollars on something, right?”
In SCT terms, then, the striking disconnect between the world of leaders and that of employees makes it hard for either to see each other as part of a larger “we.” Psychologically, plethoric perks signal to CEOs that they have “earned” special treatment, fostering entitlement and reducing empathy for those with less (Keltner et al., Reference Keltner, Gruenfeld and Anderson2003). Simultaneously, from the perspective of employees, those same perks signal that executives are more invested in their personal indulgence and advancement than they are in such things as shared mission and the greater good.
It is also the case that pay disparities in organizations more generally can create salient and meaningful ingroup–outgroup distinctions (e.g., Tanjitpiyanond et al., Reference Tanjitpiyanond, Jetten and Peters2023a, Reference Tanjitpiyanond, Peters and Jetten2023b), not least between leaders and followers at levels lower than the C-suite (Peters et al., Reference Peters, Fonseca, Steffens and Hauser2024). As a case in point, a recent experimental study by Peters and colleagues showed that when individuals served in a supervisory role on a task and received a significantly larger bonus than workers (one that was 10 times higher), workers identified less with those leaders and their organizations than they did when the pay gap was smaller (Peters et al., Reference Peters, Fonseca, Steffens and Hauser2024). Such findings further illustrate how salient pay disparities between lower level leaders and their followers can prime self-categorical distinctions between us and them, undermine shared identity (i.e., a sense that we are in this together), and lead followers to perceive leaders as going into bat for their own self-interest not that of the larger collective.
Additionally, excessive valorization through awards and recognition often reinforces a sense that successful outcomes are attributable solely to those who occupy high-profile formal leadership positions. Such recognition—and the accolades that accompany it—are indeed linked to motivation but not always in a beneficial way. For example, this may reduce the intrinsic motivation of those who receive the awards, rewards, and recognition (i.e., leaders), particularly when the award is monetary and given regularly (Gallus & Frey, Reference Gallus and Frey2016). Moreover, by promoting “I-thinking,” awards can engender overconfidence in recipients, leading to negative behaviors such as disrespect toward colleagues or demands for higher compensation (Gallus & Frey, Reference Gallus and Frey2016). And because they receive no recognition for their contribution, awards that are given only to leaders can be demotivating for followers in ways that ultimately reduce both performance and morale (Gallus & Frey, Reference Gallus and Frey2016).
The narrow focus of awards and recognition also perpetuates a hierarchical view of leadership in which the contributions of those in lowly positions are undervalued and underappreciated (Tourish, Reference Tourish2014). More generally, the emphasis on glorifying high-profile leaders can promulgate a culture that prioritizes individual achievements over collective success and feeds the furious entitlement of leaders and their narcissistic tendencies (Haslam et al., Reference Haslam, Reicher and Platow2011; Owen, Reference Owen2006; Owen & Davidson, Reference Owen and Davidson2009). This in turn can embolden risky leader behavior and presage highly variable firm performance (Chatterjee & Hambrick, Reference Chatterjee and Hambrick2007).
Depersonalizing practices
To assist in depersonalizing leader compensation, we emphasize two critical points. First, it is to make leader compensation conditional—at least in part—on team/organizational performance. This encourages team and organizational members to become interdependent and to look to each other as a basis for extrinsic reward (Aguinis et al., Reference Aguinis, Gottfredson and Joo2013; Grijalva et al., Reference Grijalva, Maynes, Badura and Whiting2020). For example, in highly interdependent work settings, hybrid reward structures that provide both individual and shared rewards for performance have been found to reduce self-serving behavior at the same time that they increase behavior that benefits the collective (Pearsall et al., Reference Pearsall, Christian and Ellis2010). Considering that many organizations currently use hybrid reward structures and yet still observe narcissistic leader behavior, our second point becomes even more important to the process of depersonalizing leader compensation. That is, when providing individual rewards, organizations should strive to minimize large disparities in rewards between leaders and followers. Failing to do so will ultimately make distinctions between leaders and followers more salient (Tanjitpiyanond et al., Reference Tanjitpiyanond, Jetten and Peters2023a, Reference Tanjitpiyanond, Peters and Jetten2023b), reduce followers’ identification with leaders (Peters et al., Reference Peters, Fonseca, Steffens and Hauser2024), and fuel leader narcissism.
To counteract the negative effects of excessive praise of leaders and foster more inclusive and collaborative organizational culture, it is imperative to shift the focus from individual leaders to the collective efforts of teams. By emphasizing the importance of group-based achievements and transparent award selection processes, organizations can also redirect attention toward shared success rather than individual glory. Additionally, recognizing and appreciating the contributions of all employees, irrespective of their position in any given hierarchy, can help to mitigate the overvalorization of individual leaders and promote a sense of community and mutual respect and support (Kellerman, Reference Kellerman2012). However, along the lines of the negative outcomes (e.g., turnover) that organizations experience when their practices are conflicting (e.g., Trzebiatowski et al., Reference Trzebiatowski, Jiang, Zhang, Eckardt and Kim2025), we caution leaders against only adopting “quick fix” approaches to reward modification (e.g., emphasizing collective achievements) while neglecting other important features of pay structures (e.g., large pay disparities). All things considered, organizations that take thoughtful, comprehensive, and coherent steps to neutralize narcissism are generally better places to work, and they are less likely to be destroyed from within.
Limitations and future directions
We offer this perspective not as a definitive account, but as a provocation—an invitation to scholars and practitioners to engage with these ideas, extending, or refining them to deepen our understanding not only of who becomes a leader but also of how those leaders see themselves and serve others. Although we have pointed to ways in which selection, personnel development, and compensation systems can personalize or depersonalize leadership, our discussion is not exhaustive. Indeed, our sense is that the processes we have discussed are relevant to most, if not all, HR policies and practices (e.g., those relating to recruiting, conduct, antiharassment, safety, accommodations, termination, and resignation; Holliday, Reference Holliday2021). For example, job design practices such as flexible job assignments (Sun et al., Reference Sun, Aryee and Law2007) can either personalize or depersonalize leadership depending on how they are implemented. If such assignments emphasize interpersonal differences in competence or are reserved only for some (e.g., managers deemed “special” or deserving), they may inadvertently promote narcissism by reinforcing a lack of depersonalization.
There is therefore a pressing need for research to test the theoretical claims we have made across diverse organizational contexts and time. Future work should adopt a systematic approach to cataloging how HR practices vary in their degree of (de)personalization and to the effects of this on organizational life. In particular, it needs to clarify the extent to which hyperpersonalizing practices fuel leader narcissism and organizational dysfunction and the extent to which these can be mitigated by more depersonalizing practices.
It is also critical to identify the boundary conditions to these effects (e.g., the degree to which they are affected by organizational culture as well as broader ideologies) and their role in advancing (or else hindering) a range of organizational and societal goals (e.g., of diversity and inclusion; Boucher et al., Reference Boucher, Fuesting, Diekman and Murphy2017; Lagowska et al., Reference Lagowska, Sobral, Jacob, Hafenbrack and Goldszmidt2023). In this we are mindful too of the capacity for both personalizing and depersonalizing practices to have paradoxical effects, in which an emphasis on the one leads people to hanker for, and respond more positively to, the other.
Conclusion
In this paper we have argued that, for all their worthy efforts to help organizations develop good leaders and citizens, prevailing HR practices can all too easily end up doing the opposite. From a self-categorization perspective (after Turner et al., Reference Turner, Hogg, Oakes, Reicher and Wetherell1987), we have argued that this is because hyperpersonalizing practices may not only attract narcissistic individuals to those organizations but also then go on to validate, reward, and reinforce narcissistic cognitions, emotions, and behaviors among the leaders they anoint. In this way, HR can all too easily end up being “me-sources.” In contrast, we have suggested that developing and implementing HR practices that are more depersonalizing can help suppress narcissistic tendencies among leaders and orient them to think and act in the best interests of organizations—so that human resources are ultimately “we-sources” from which everyone stands to benefit.
We recognize that for readers who have been schooled in, and are deeply committed to, the wisdom of singling out leaders and developing pedestals on which to elevate them this will likely be a confronting message. We recognize too that this approach is one that is widely honored in contemporary Western societies and hence that our analysis goes against the cultural grain. Nevertheless, we urge organizational scholars and practitioners to reexamine the underlying logic of organizational and HR practices through a more critical lens. Not least, this is because we are interested in having leaders who see—and who want to increase—the value of we and us.
Appendix. A Hyperpersonalizing Checklist
Description: This checklist is designed to help organizational members consider to what extent their human resource (HR) practices in the areas of selection, development, and rewards are hyperpersonalizing—that is, whether they set (potential) leaders apart from other members of their team and/or the organization. Individuals and groups in organizations can use the checklist and examples of hyperpersonalizing HR practices below and supplement with additional examples of their own organizational practices to reflect on ways in which various practices can contribute to narcissistic behavior by leaders.
