top of page
Search

Achilles at the Top Table: What the Research Says About Narcissistic and Hubristic Leaders, and Working Beside One.

Writer: Meg Steinschauer, MBA
Meg Steinschauer, MBA
Mar 24, 2025
45 min read

Updated: Sep 28

The meeting was booked for an hour and the decision was made by ten past nine. The executive had met the founder of the company they were thinking of buying, liked the energy, had "a good feel for it", and said so with the kind of certainty that makes a room nod. You had built the pre-read. On page fourteen there was a chart showing that the last two deals of this kind had taken three years longer to pay back than anyone expected, and you were fairly sure nobody at the table had turned to page fourteen. That evening you drafted the follow-up email three times, trying to find the version of "have you seen page fourteen?" that would be read rather than filed.


I wrote about narcissistic and hubristic leadership in 2025, with Achilles in the title and a list of red flags underneath, and the distinction at its heart holds up well: one kind of leader needs to look invincible, the other has come to believe it. I have spent years as the right hand to some truly inspiring leaders, and what I did not have then was the research on why confident people get chosen to lead, what the evidence says they actually do at the top of an organisation, how power quietly changes almost anyone who holds it, and what the person in the seat beside them can do to keep good information flowing, keep the team safe and keep their own footing. That research is large, careful and a good deal kinder to everyone involved than the word "narcissist" suggests.


A single black chess king standing alone on a dark wooden chessboard with warm walnut grain in its squares, against a black background, lit low from one side so the piece casts a long shadow across the board, the Connected Assistants wordmark at the foot.
The shadow is always taller than the piece. It helps to know which one you are working with.



What you will get out of this


  • Why narcissism and hubris are two different things, and why the difference changes what you do

  • The meta-analysis that explains why confident people get chosen to lead, and why the same confidence tells you almost nothing about whether they will lead well

  • What studies of chief executives found about narcissism, overconfidence and the decisions that follow, including one large review that says overconfidence usually pays

  • How power changes the way anyone listens, takes advice and weighs risk, and the two conditions under which it does not

  • How to get a hard fact heard by someone who is very sure of themselves, without flattery and without a fight

  • What protected the people working for self-focused leaders in the studies, and the kind of coping that made things worse

  • The steadiness check: a tool that reads your situation and suggests where to put your effort first

  • The Steady Hand Pages: three pages for the brief, the decision and your own footing




Two different things, and why the difference helps you


One thing before the research, because it matters more than anything else in this post. None of what follows is a diagnostic kit, and the most useful thing the research does for you is to take the diagnosis off the table.


Psychologists themselves disagree about whether professionals should ever offer opinions on the mental health of people they have not examined. A 2018 review argued that the long-standing rule against it in psychiatry rests on shaky assumptions, and even its authors kept that argument to select cases and to people with the training to make them. If trained clinicians argue about it, the rest of us can safely leave it alone. And there is a better reason to leave it alone than modesty. A study of matched leaders and followers found that what harmed followers was the leader's narcissistic behaviour as the followers saw it; the leaders' own scores on the same trait predicted nothing. The behaviour is the thing that matters, and the behaviour is also the thing you can work with.


So in this post, narcissism and hubris are patterns of behaviour and ways of relating to power. They are not labels for a person, and certainly not for the person whose diary you run.


With that said, the two patterns are different, and knowing which one you are looking at changes what helps.


Narcissism is a trait, and everyone has some of it. The research treats it as a dimension rather than a category, and separates two forms: a grandiose form that is openly self-assured, entitled and dominant, and a vulnerable form that is self-focused too, and also anxious, distrustful and withdrawn. The grandiose form is the one that turns up in leadership research. One of the most useful models of it splits it into two routes to the same goal of a special, admired self. Admiration is the charming route: self-promotion, big visions, a strong first impression. Rivalry is the defensive route: running others down, reacting badly to challenge, guarding status. The two tend to travel together, but they have quite different consequences for the people nearby, and most of the damage in the workplace studies comes from the second. A process model published in 2020 describes the underlying pattern as a moment-by-moment pursuit of status: seeking situations that offer it, watching closely for cues about who has it, and then either promoting oneself or diminishing someone else depending on which looks more likely to work.


The older clinical view underneath all of this is quietly sympathetic. It describes a self that is grandiose on the surface and fragile underneath, in constant need of outside confirmation, whose efforts to get that confirmation often push away the people who could give it.


Hubris is something that happens to people. The researchers who named hubris syndrome, two doctors, one of them a psychiatrist, writing in a medical journal in 2009, described it as an acquired condition. The trigger is power, particularly power that has come with overwhelming success, been held for years and met little constraint, and it usually fades when the power does. That is the crucial difference. Narcissism is largely something a person brings to the role. Hubris is something the role, the track record and the lack of friction can do to a person who did not arrive with it.


A 2020 paper put the distinction in one memorable image. Hubristic leaders become intoxicated with their positional power and their past success. For narcissistic leaders, power is the means by which they stay intoxicated with themselves. Another line of work links hubris to a very high level of what psychologists call core self-evaluation, meaning a person's settled sense of their own worth, competence and control, pushed past the point where it helps. And one group of researchers has argued that the ingredient most often overlooked is intuition: a leader who has been right on instinct many times comes to trust instinct over evidence, and nobody around them is in a position to say otherwise.


I mentioned in 2025 a paper that set out four families of factors that bring narcissistic leadership to the surface: the individual, the culture, the environment and the structure around the leader. That framing holds up, and it is worth keeping, because three of the four are about the situation.


A graphic titled Two different things, with the eyebrow From the research on narcissism in leaders and on hubris syndrome. Standfirst: Narcissism is mostly something a person brings to the role. Hubris is mostly what power, success and a lack of friction do to a person over time. They often appear together, and respond to different levers. Five rows. One, Where it comes from, Trait or acquired: Narcissism is a trait everyone has in some measure, in a grandiose or a vulnerable form. Hubris is acquired: it grows with power held for years, after big success, with little constraint. Two, What it feeds on, The fuel: Narcissism needs to be seen and admired, and uses power to keep that going. Hubris feeds on a track record: past success, and the power it brought. Three, How it meets a challenge, When someone disagrees: Narcissism guards status: running others down, reacting badly to criticism. Hubris waves the challenge away, trusting instinct over evidence. Four, What makes it worse, Conditions: For both: flattery, uncertainty, a board that rarely asks questions, and decisions made fast with nobody checking. Five, What helps, Where to push: Narcissism responds to audience, credit and saving face. Hubris responds to friction, a slower pace and regular contact with how things turn out. Foot: Miller and colleagues (2017); Back and colleagues (2013); Owen and Davidson (2009); Asad and Sadler-Smith (2020).
One arrives with the person; the other arrives with the corner office. Often they share it.

Here is why this helps you. If the pattern in front of you is mostly about being seen, the levers are about audience, credit and face. If it is mostly about being sure after a long run of success, the levers are about friction, contact with reality and pace. Often you will see some of both, and that is fine. The point is that each one has a direction you can push in, and neither of them requires you to decide what kind of person your executive is.


The behaviour is the thing that matters, and the behaviour is also the thing you can work with.

Try this. Next time something at work makes you reach for a label, write down the behaviour instead, in one sentence, with the situation it happened in. "Announced the decision before anyone had spoken, in a meeting where the board chair was present." After three or four of those sentences you will usually see whether the pattern is about audience or about certainty, and you will have something you could describe calmly to anyone, which a label never gives you.


Lilienfeld, Miller and Lynam (2018) reviewed the Goldwater Rule, which prohibits psychiatrists from offering diagnostic opinions on people they have not examined, and argued that it rests on dubious scientific assumptions, while confining their case to select instances involving psychological scientists with relevant expertise. Fehn and Schütz (2021) studied matched leaders and followers and found that followers of leaders high in narcissistic rivalry reported less supervisor support, lower quality leader-member relationships, lower performance-based self-esteem and lower engagement, but only when leaders' rivalry was rated by followers and not when it was self-rated, implying that the effects are driven by expressed behaviour. Miller and colleagues (2017) reviewed controversies in narcissism research and argued that failing to distinguish grandiose (overtly immodest, self-centred, entitled, domineering) from vulnerable (self-centred, distrustful, neurotic, introverted) presentations has made the literature less coherent. Back and colleagues (2013) validated the two-dimensional model of narcissistic admiration and rivalry across seven studies, showing the dimensions to be positively correlated with distinct consequences. Grapsas and colleagues (2020) proposed a process model in which narcissists select status-affording situations, attend to status cues and pursue status through self-promotion or other-derogation. Morf and Rhodewalt (2001) described narcissism as a dynamic self-regulatory system built around a grandiose yet vulnerable self-concept and a chronic need for external affirmation, whose strategies often undermine the self they aim to maintain. Owen and Davidson (2009) proposed hubris syndrome as an acquired condition triggered by power, particularly power associated with overwhelming success held for years with minimal constraint, and usually remitting when power fades. Sadler-Smith and colleagues (2017) reviewed hubristic leadership, describing leaders who overestimate their abilities, make over-ambitious decisions and are resistant to criticism and contemptuous of advice, and distinguished hubris from overconfidence, narcissism, core self-evaluation and pride. Asad and Sadler-Smith (2020) argued that hubristic leaders are intoxicated with positional power and prior success, while for narcissists power facilitates self-intoxication and maintains a grandiose self-view. Hiller and Hambrick (2005) argued that a very high level of core self-evaluation aligns with what is colloquially called hubris. Claxton, Owen and Sadler-Smith (2015) proposed unbridled intuition as an overlooked source of leaders' hubris. Ouimet (2010) synthesised the research on narcissistic leadership and identified idiosyncratic, cultural, environmental and structural factors that can trigger it.




Why the confident one gets chosen, and what happens next


The single most useful finding in this whole field is a distinction between two questions that feel like one. Does a person come across as a leader? And do they lead well?


In three studies published in 2008, people took part in leaderless group discussions and the researchers watched who took charge. Across all three, the more narcissistic participants were more likely to emerge as the leader, and in the third study the same link turned up in expert ratings of a group of executives.


A 2015 meta-analysis pulled the research together and gave the answer that has shaped the field since. Narcissism was positively related to emerging as a leader, and most of that was explained by extraversion: narcissistic people tend to be outgoing, talkative and visible, and groups read that as leadership. But narcissism had no straight-line relationship with leadership effectiveness at all. When the leaders rated themselves, the more narcissistic ones rated their own effectiveness higher; when supervisors, peers and the people they led did the rating, there was no relationship. And underneath that flat line the authors found a curve: leaders with a moderate level of narcissism did better than those with very little or a great deal. A little self-belief helps. A lot of it stops helping.


A 2011 experiment showed what that gap can cost. Groups worked on a problem where each member held different pieces of information and the right answer only appeared if people pooled what they knew. Narcissistic leaders were seen as more effective, because they projected authority. Their groups performed worse, because the leader's style inhibited the exchange of information between members. The picture and the result pointed in opposite directions.


Time and distance are what close the gap. In two studies that followed groups over twelve weeks, narcissism predicted being rated as a leader when the group members were strangers, and stopped predicting it as they got to know each other; in groups who already knew each other, it predicted lower leadership ratings later on. A study of 311 people meeting weekly for three weeks, all filmed and coded, found the same decline in popularity, and pinned down why: the charming admiration route started strong and faded, while the defensive rivalry route grew steadily more costly as people saw more of it.


The most striking version comes from a study of 175 retail stores. Narcissistic store leaders had less organisational experience when they were hired. Where the staff had few chances to see their leader, the leader's narcissism went with higher ratings of effectiveness and better job attitudes. Where the staff saw their leader often, that positive link disappeared.


Which brings us to you. Nobody in an organisation sees a leader more often, or from closer, than the person who runs their working life. If your view of your executive is more mixed than the view from the board or the town hall, that is what the research would predict. Neither view is wrong. You are simply standing where the gap closes first.


A graphic titled How the picture changes with time and distance, with the eyebrow From studies of leaderless groups, twelve-week teams and 175 retail stores. Standfirst: Narcissism helps people get chosen to lead and tells you almost nothing about how well they will lead. The gap between the two closes with time, and it closes first for whoever stands closest. Five numbered steps, each with a label. One, The first meeting, labelled Emergence: In leaderless groups, the more narcissistic people were more likely to take charge. Groups read confidence and talk as leadership. Two, When things feel uncertain, labelled The pull: People chose more narcissistic leaders in uncertain conditions, while fully aware of the arrogance that came with them. Three, A few weeks in, labelled Honeymoon: Charm faded week by week and rivalry cost more, so popularity fell as groups got to know the person. Four, Seen from a distance, labelled The board's view: Where staff rarely saw their store leader, the leader's narcissism went with better ratings of effectiveness. Five, Seen up close, labelled Your view: Where staff saw their leader often, that advantage disappeared. Nobody sees an executive more often than the person beside them. Foot: Brunell and colleagues (2008); Nevicka and colleagues (2013, 2018a); Leckelt and colleagues (2015).
Magnificent from the back of the town hall. Rather different from the desk outside the door.

There is also the question of why organisations keep choosing this style. In three studies, people rated narcissistic characteristics as more desirable in a leader when the situation was uncertain, and chose more narcissistic leaders more often in uncertain conditions, while being fully aware of the arrogance and exploitativeness that came with them. A larger set of studies found that economic uncertainty lowered people's sense of personal control and raised their preference for dominant leaders over respected ones. When things feel shaky, certainty is reassuring, even when we can see its price.


And the honest counterweights, of which there are several. Narcissistic people were no more creative than others in a 2010 set of studies, though they were better at persuading people that their ideas were creative, largely through enthusiasm, and groups with a moderate number of narcissists produced more creative outcomes than groups with none. In a large health insurer, narcissistic leaders who also showed humility had more engaged and better-performing staff; the combination worked where narcissism alone did not. Two studies of more than 1,500 managers found inverted-U relationships between dark-side personality composites and rated leadership performance, and a study of 320 managers rated by 4,906 colleagues found that scores in the middle went with the most effective behaviour, and that very low scores, as well as very high ones, went with ineffective extremes. A leader with no self-belief at all is not the answer either.


One review made a useful point: much of what we admire as transformational leadership overlaps heavily with grandiose narcissism, which is why the two are so hard to tell apart from a distance.


Try this. When you find your view of your executive at odds with how others see them, resist the urge to correct either picture. Instead, ask what the others are not positioned to see, and what you might not be positioned to see either. The board sees results and presence; you see process and the people who carry it. Both are information, and your value lies in bringing the second kind to where it is needed, calmly and without a verdict attached.


Brunell and colleagues (2008) found across three studies of four-person leaderless group discussions that narcissism was linked to leader emergence, with the power dimension of narcissism predicting reported emergence after controlling for sex, self-esteem and the Big Five, and in a third study found an association with expert ratings of emergence among executives. Grijalva and colleagues (2015) meta-analysed the narcissism and leadership literature and found a positive relationship with leadership emergence explained by extraversion, no linear relationship with leadership effectiveness, a positive relationship with self-reported but not observer-reported effectiveness, and a curvilinear trend suggesting an optimal midrange level of narcissism. Judge, LePine and Rich (2006) found in two studies that narcissism was related to enhanced self-ratings of leadership, and in the second study was positively correlated with self-ratings and negatively related to other ratings of leadership. Nevicka and colleagues (2011) used a hidden-profile task and found that narcissistic leaders were perceived as effective because of their displays of authority while their narcissism inhibited information exchange between members and lowered group performance. Ong and colleagues (2016) followed groups for twelve weeks and found narcissism positively associated with peer-rated leadership at initial formation among unacquainted people but not later, and among acquainted people unrelated at first and negatively associated later. Leckelt and colleagues (2015) followed 311 participants interacting weekly for three weeks and found that narcissists became less popular over time, with an initially positive but decreasing effect of admiration and an increasingly negative effect of rivalry. Nevicka and colleagues (2018a) collected multisource longitudinal data from 175 retail stores and found that narcissistic leaders had less organisational experience when hired, and that leader narcissism was positively related to perceived effectiveness and job attitudes when followers had few opportunities to observe the leader, a relationship that disappeared when they had more. Nevicka and colleagues (2013) found that narcissistic characteristics were rated as more desirable, and high narcissists chosen more often, in uncertain contexts, despite awareness of their negative features. Kakkar and Sivanathan (2017) found across three large-scale studies that economic uncertainty lowered people's sense of personal control and increased their preference for dominant over prestigious leaders. Goncalo, Flynn and Kim (2010) found that narcissists were not more creative but were rated as more creative when pitching ideas, partly through enthusiasm, and that groups showed a curvilinear effect of the number of narcissists on creativity. Owens, Wallace and Waldman (2015) found in a health insurance organisation that the interaction of leader narcissism and humility was associated with perceived leader effectiveness and with follower engagement and performance. Benson and Campbell (2007) found inverted-U relationships between derailing personality composites and leadership performance in samples of 1,306 and 290 managers. Kaiser, LeBreton and Hogan (2015) studied 320 managers rated by 4,906 co-workers and found that scores near the mean on dark-side traits went with optimal leader behaviour, while both high and low scores went with extreme, ineffective behaviour. O'Reilly and Chatman (2020) reviewed the overlap between transformational leadership and grandiose narcissism and described "pseudo-transformational" leaders. Paulhus (1998) examined trait self-enhancement as a mixed blessing for interpersonal and personal adjustment. Hogan and Hogan (2001) reviewed managerial derailment and proposed that the base rate of managerial incompetence is high. Campbell and colleagues (2010) reviewed narcissism in organisational contexts, and Rosenthal and Pittinsky (2006) reviewed narcissistic leadership. Fatfouta (2019) reviewed the facets of narcissism in leaders and how to assess, select and develop them. Judge, Piccolo and Kosalka (2009) set out the bright and dark sides of leader traits, Harms, Spain and Hannah (2011) examined leader development and the dark side of personality, and Smith and colleagues (2018) reviewed the upsides of dark traits and the downsides of bright ones. Owens, Johnson and Mitchell (2013) developed an observer-report measure of expressed humility across eight samples and linked leader humility to employee retention and engagement; Owens and Hekman (2012) interviewed 55 leaders about humble leader behaviours; and Ou and colleagues (2014) found in 63 private companies that humble chief executives were associated with empowering leadership, top management team integration and middle managers' engagement, commitment and performance. Braun (2017) systematically reviewed 45 studies of leader narcissism across dyadic, team and organisational levels and found persistent inconsistencies about whether its consequences are positive or negative.




What the boardroom studies found


The research on chief executives is where this subject gets its reputation, and it deserves a careful reading, because the headline and the detail do not always match.


The landmark study measured narcissism in 111 chief executives in the computer hardware and software industries without asking them anything. The researchers used four quiet clues: how prominent the chief executive's photograph was in the annual report, how often they featured in the company's press releases, how often they said "I" in interviews, and how their pay compared with the next highest-paid executive. More narcissistic leaders ran companies with more dynamic and grandiose strategies, and made more and larger acquisitions. Their companies' results swung more widely, with bigger wins and bigger losses. And here is the detail people tend to leave out: on average, those companies performed no better and no worse than the others.


A follow-up found something even more useful for anyone who writes briefings. Highly narcissistic chief executives were much less responsive than their peers to objective evidence of how the business had recently performed, and especially emboldened by social praise: glowing press coverage and awards. The numbers moved them less, and the applause moved them more.


The pattern repeats across settings. In pharmaceutical companies facing the arrival of biotechnology, narcissistic chief executives invested more aggressively in the new technology, especially when the outside world was paying attention to it. In a survey of 173 chief executives, narcissism went with a more entrepreneurial posture, which in turn helped explain why those companies' results varied more. A study of 32 companies found those led by narcissistic chief executives were more often involved in lawsuits, and that the lawsuits dragged on longer; in two follow-up experiments, narcissistic participants were less sensitive to objective assessments of legal risk and less willing to take advice from experts. Other studies have linked chief executive narcissism to pay, to fraud and, in one study that measured it from the size of executives' signatures, to investment and performance.


A 2020 review and meta-analysis added the caution this literature needs. It found five different ways researchers have measured narcissism in chief executives, each with strengths and weaknesses, and findings that share themes but remain mixed and may depend on which method was used.


The research on hubris and overconfidence runs alongside. The idea that bidders pay too much for companies because of hubris was proposed in 1986. In 106 large acquisitions, four indicators of chief executive hubris, including recent media praise and a measure of self-importance, went with larger premiums paid, and the effect was stronger when the board was less vigilant: when more directors were insiders and when the chief executive also chaired the board. The greater the hubris and the premium, the greater the losses to the acquiring company's shareholders. Studies using a clever measure of overconfidence, chief executives who kept far more of their own company's risk than was sensible, found their investment tracked the cash available more closely, that they made more acquisitions, especially outside their core business, and that markets reacted more negatively when they announced them. Among repeat acquirers, first deals were greeted neutrally by the market and later deals negatively, consistent with success being credited to skill and overconfidence building deal by deal. Press celebrity can do the same thing: journalists credit the leader for what the company did, and the leader comes to believe it. And companies led by overconfident chief executives were slower to correct their earnings forecasts after getting them wrong.


There is a quieter finding in this literature that I think every executive assistant and chief of staff should know. A study of chief executives and the people around them found that high levels of flattery and agreement from other managers and board members made chief executives more overconfident in their own judgement, which made them less likely to change strategy after poor results, and that the persistence could keep performance low. The chorus has consequences.


Now the real disagreement, and you deserve both halves. A 2023 meta-analysis of 199 studies concluded, against the conventional wisdom, that chief executive overconfidence is on average good for company performance, partly because it drives the strategic risk taking that creates value, and more so where chief executives have more discretion. The acquisition studies say overconfidence costs shareholders money. My reading is that both can be right: across the ordinary run of strategic decisions, a leader willing to take risks tends to create more value than a timid one, and the costs concentrate in a particular kind of decision, large, irreversible and made quickly with nobody checking. Several reviews of hubris make the same point, that it has a good side and a bad side, and a model of narcissistic leaders in organisations describes a predictable arc: a burst of energy and reform, a clash as the costs accumulate, and a resolution in which the leader either leaves or accepts controls negotiated with the organisation.


A graphic titled What the boardroom studies found, with the eyebrow From studies of chief executives, their decisions and their companies. Standfirst: The headlines about narcissistic and overconfident chief executives are harsher and simpler than the evidence. Here are five common beliefs, and what the studies actually show. Five common beliefs, each marked with a cross, and beneath each what the studies found. One, Narcissistic chief executives ruin their companies: In 111 technology companies they made bolder bets and got bigger swings, up and down. On average, their results were no better and no worse. Two, Overconfidence always costs money: A meta-analysis of 199 studies found it was, on average, good for performance, through the risks it made people take. The costs show up most in the studies of acquisitions. Three, Show them the numbers and they will change course: Highly narcissistic chief executives were moved less by recent results than their peers, and more by praise and awards. Four, A little flattery keeps things smooth: Flattery and agreement made chief executives more overconfident and slower to change strategy after poor results. Five, You can only change the leader: Overconfident chief executives overpaid less for acquisitions when the process before the deal ran more slowly. Foot: Chatterjee and Hambrick (2007, 2011); Burkhard and colleagues (2023); Park, Westphal and Stern (2011); Pavićević and Keil (2021).
Less a cautionary tale than a volatility warning, with one surprisingly practical fix.

The most practical finding for your role is also the most encouraging. In a study of corporate acquisitions, overconfident chief executives paid higher premiums, as earlier research had found, and paid less when the process leading up to the deal ran at a slower pace. Nobody had to change the leader. Somebody changed the tempo.


Nobody had to change the leader. Somebody changed the tempo.

Try this. Look at the next big, hard-to-reverse decision on your executive's horizon and ask one question about its process rather than its substance: is there a point, before commitment, where someone with no stake in the answer looks at the numbers? If not, you are often the person best placed to suggest one, as a diary entry rather than an argument. "I have held forty minutes on Thursday for finance to walk through the model before the board paper goes."


Chatterjee and Hambrick (2007) measured the narcissism of 111 chief executives in the computer hardware and software industries between 1992 and 2004 through the prominence of their photograph in annual reports, their prominence in press releases, their use of first-person singular pronouns in interviews and their pay relative to the second-highest-paid executive, and found narcissism positively related to strategic dynamism and grandiosity, to the number and size of acquisitions, and to extreme and fluctuating performance, with performance generally no better or worse than firms with less narcissistic chief executives. Chatterjee and Hambrick (2011) found in two studies that highly narcissistic chief executives were much less responsive to recent objective performance than their peers and especially emboldened by media praise and awards. Gerstner and colleagues (2013) studied pharmaceutical firms' responses to biotechnology from 1980 to 2008 and found narcissistic chief executives more aggressive in adopting the discontinuous technology, especially when audience engagement was high. Wales, Patel and Lumpkin (2013) surveyed 173 chief executives and found that narcissism was associated with higher entrepreneurial orientation, which partially explained greater variability in firm performance. Petrenko and colleagues (2016) found in a sample of Fortune 500 chief executives that narcissism had positive effects on the level and profile of corporate social responsibility and reduced its effect on performance. O'Reilly, Doerr and Chatman (2018) found in a sample of 32 firms that those led by narcissistic chief executives were more often involved in litigation and that suits were more protracted, and in two experiments that narcissists were less sensitive to objective risk assessments and less willing to take expert advice. O'Reilly and colleagues (2014) examined narcissistic chief executives and executive compensation. Rijsenbilt and Commandeur (2013) examined chief executive narcissism and fraud. Ham, Seybert and Wang (2018) examined chief executive signature size, investment and performance. Zhang and colleagues (2017) examined chief executive humility, narcissism and firm innovation as a paradox. Chatterjee and Pollock (2017) set out how narcissistic chief executives construct their professional worlds. Cragun, Olsen and Wright (2020) combined meta-analysis and narrative review, identified five methods of measuring chief executive narcissism and found the findings mixed and potentially dependent on method. Roll (1986) proposed the hubris hypothesis of corporate takeovers. Hayward and Hambrick (1997) found in 106 large acquisitions that four indicators of chief executive hubris were highly associated with premiums paid, more strongly when board vigilance was lacking, and that greater hubris and premiums went with greater shareholder losses. Malmendier and Tate (2005) found that the investment of overconfident chief executives, classified by their failure to reduce personal exposure to company-specific risk, was significantly more responsive to cash flow, particularly in equity-dependent firms. Malmendier and Tate (2008) found overconfident chief executives more acquisitive, particularly through diversifying deals, and market reactions to their bids significantly more negative. Billett and Qian (2008) found that chief executives' first deals had zero announcement effects and later deals negative ones, consistent with self-attribution bias leading to overconfidence. Hayward, Rindova and Pollock (2004) theorised that chief executive celebrity can produce overconfidence and hubris. Li and Tang (2010) surveyed 2,790 chief executives of Chinese manufacturing firms and found the link between hubris and firm risk taking stronger when managerial discretion was greater, including when the chief executive also chaired the board. Tang and colleagues (2015) found in S&P 1500 firms from 2001 to 2010 that chief executive hubris was negatively related to socially responsible and positively related to socially irresponsible activities. Chen, Crossland and Luo (2015) found firms led by overconfident chief executives less responsive to corrective feedback in improving forecast accuracy. Brown and Sarma (2007) found with Australian data that chief executive dominance was at least as significant as overconfidence in the decision to acquire. Park, Westphal and Stern (2011) found that high levels of flattery and opinion conformity increased chief executives' overconfidence, reduced the likelihood of strategic change in response to poor performance, and could result in persistently low performance. Burkhard and colleagues (2023) meta-analysed 199 studies and found chief executive overconfidence on average beneficial for firm performance, partially mediated by strategic risk taking and stronger under high managerial discretion. Bollaert and Petit (2010) reviewed research on the dark side of executive psychology. Picone, Dagnino and Minà (2014) reviewed the hubris hypothesis and argued that managerial hubris has good and bad sides. Sedikides and Campbell (2017) proposed the Energy Clash Model, in which a narcissistic leader perturbs an organisation, clashes with it as costs accrue, and is then expelled or accommodated through negotiated controls. Resick and colleagues (2009) found in 75 chief executives of baseball organisations that narcissism was negatively related to contingent reward leadership. Pavićević and Keil (2021) replicated the finding that overconfident chief executives pay high acquisition premiums and found this tendency reduced when pre-deal processes were slower in pace.




What power does to anyone


This section changes the tone of everything else. Much of what looks like a difficult personality at the top of an organisation is what power does to ordinary people, and the research shows it doing so in minutes.


A theory set out in 2003 summarised it: having power switches people towards approach, meaning rewards, positive mood, quick automatic judgements and fewer inhibitions, while lacking power switches people towards inhibition, meaning attention to threats, to other people's interests and to how one is coming across. Nobody chooses this. It is the way the situation tunes attention.


People asked to recall a time they had power were more likely, when asked to draw an E on their own forehead, to draw it the right way round for themselves and backwards for anyone looking at them. They were less likely to allow for the fact that other people did not share their private knowledge, and less accurate at reading other people's facial expressions.


The same thing happens to advice. In four studies, including a field survey of working professionals rated by their colleagues, more power meant less advice taken, because it made people more confident in their own first judgement, and in the fourth study the more powerful participants ended up less accurate. Another four experiments found that people who felt powerful discounted advice from experts and novices alike, where everyone else weighted the expert more, through feelings of competitiveness and confidence. A further set of studies found that a formal leader's sense of power made them dominate the conversation, which cut the team's communication and its performance, and only when the leader held a formal leadership role, because the effect depended on everyone else going along with it.


Risk moves too. Five experiments found that feeling powerful produced overconfident decisions that lost the powerful people money. Five more found that it made people more optimistic about risk and more willing to take it.


Here is the hopeful part, and it is where your role comes in. In each of these research programmes the researchers also found what switched the effect off. The link between power and overconfidence disappeared when power was not front of mind, and when the powerful person was made to feel less than expert in the area where they held power. Power made people more optimistic about risk except when they felt a sense of responsibility for others. Powerful people who were led to feel cooperative with their adviser weighted expert advice properly again. And in a set of studies with managers and professionals, people who saw their power as a responsibility took more advice than those who saw it as an opportunity, whoever the advice came from, because they valued it more.


It helps to know which kind of overconfidence you are dealing with, too. A 2008 review separated three: overestimating how well you will do, overplacing yourself relative to others, and being too precise, too sure the answer lies in a narrow range. The third is the most persistent. It is the one that says "it will take six months" when the honest answer is "somewhere between six and eighteen". A classic paper on decision making named the mechanism behind many bold forecasts: taking the inside view, reasoning from the plan in front of you, instead of the outside view, asking how things like this usually turn out.


Researchers have even looked for hubris in the language of political speeches and letters to shareholders, and the medical account describes it waning once the power is gone. What power does, the loss of power often undoes.


A graphic titled What power does to anyone, and what undoes it, with the eyebrow From laboratory and field studies of people given a sense of power. Standfirst: A few minutes of feeling powerful changes how people listen, take advice and weigh risk. In each research programme, something also switched the effect off. The effect is on the left; what eased it is on the right. Five pairs, what power does on the left and what eased it on the right. Power, Takes less advice: Feeling cooperative with the adviser brought back the proper weight on advice from experts. Power, Overconfident calls: Feeling less than expert in the area severed the link between power and overconfident decisions. Power, Sees risk as smaller: A sense of responsibility for others took the edge off the optimism about risk. Power, Listens less to anyone: Seeing power as a responsibility rather than an opportunity went with taking more advice, from anyone. Power, Dominates the room: It relied on everyone else going along, and appeared only where the leader held a formal role, so a voice that does not simply acquiesce is part of the remedy. Foot: Tost, Gino and Larrick (2012, 2013); Fast and colleagues (2012); Anderson and Galinsky (2006); De Wit and colleagues (2017).
Power works fast. Happily, so does a sense of responsibility.

Two things follow for you. The first is that none of this requires a villain. A leader who stopped listening after a string of successes may be showing you something closer to a situation than a character, which means the situation is what to work on. The second is gentler, and worth sitting with: you hold borrowed power too. You decide who gets time, what reaches the desk and in what order. The same research applies to anyone who controls access, and the same antidotes, responsibility, contact with people who see things differently and a regular reminder of what you do not know, apply as well.


Much of what looks like a difficult personality at the top is what power does to ordinary people, and what power does, the loss of power often undoes.

Try this. Frame the next piece of advice you pass up as a shared responsibility rather than a competing view. "The team is counting on this call being right, so here is what finance and I are seeing" draws on two of the findings at once: power held as responsibility listens better, and advice from someone who feels like a collaborator is weighted more than advice from someone who feels like a rival.


Keltner, Gruenfeld and Anderson (2003) reviewed evidence that elevated power is associated with positive affect, attention to rewards, automatic information processing and disinhibited behaviour, and reduced power with negative affect, attention to threat and others' interests, controlled processing and inhibition. Galinsky and colleagues (2006) found across four experiments and a correlational study that power reduced the tendency to adopt another person's visual perspective, to allow for others' lack of privileged knowledge, and to identify others' emotional expressions accurately. Hogeveen, Inzlicht and Obhi (2014) found using transcranial magnetic stimulation that high-power participants showed lower motor resonance when observing another person's actions than low-power participants. See and colleagues (2011) found across a field survey and three studies, two of them experiments, a negative relationship between power and advice taking mediated by confidence, and in the fourth study lower accuracy among higher-power participants. Tost, Gino and Larrick (2012) found across four experiments that high-power individuals discounted both expert and novice advice, through competitiveness and confidence, and that inducing them to feel cooperative with advisers restored greater weight on expert advice. Tost, Gino and Larrick (2013) found across three studies that a formal leader's subjective power produced verbal dominance that reduced team communication and performance, with the effect emerging only when the leader held a formal position. Fast and colleagues (2012) found across five experiments that power produced overconfident decisions that generated monetary losses, and that the link was severed when access to power was not salient or when the powerful were made to feel personally incompetent in their domain. Anderson and Galinsky (2006) found across five studies that power increased optimism in perceiving risk and risk-taking, effects attenuated when the high-power individual felt a sense of responsibility. De Wit and colleagues (2017) found in a field study and two experiments that power holders who construed power as responsibility rather than opportunity took more advice, whether from subordinates, experts or teammates, through a heightened perceived value of advice. Moore and Healy (2008) reconciled three forms of overconfidence, overestimation, overplacement and overprecision, and found overprecision the most persistent. Kahneman and Lovallo (1993) argued that overly optimistic forecasts result from adopting an inside view anchored on plans and scenarios. Garrard and colleagues (2014) examined linguistic markers of hubris syndrome in the speeches of political leaders. Akstinaite, Robinson and Sadler-Smith (2020) examined linguistic markers of chief executive hubris. Craig and Amernic (2018) examined whether language markers of hubris appear in chief executives' letters to shareholders.




Getting information heard


So far this has been about understanding. Now for the part of the job that happens on a Tuesday evening, with the follow-up email open and page fourteen still unread.


Start with why the usual approach struggles. A set of studies on narcissism and advice found that the more narcissistic people were, the less advice they took, and that the reason was not greater confidence. It was disregard for other people: they thought the advisers were less competent. The same studies found that putting people under pressure to justify how they had reached a decision made narcissistic people take even less advice. In two further studies of 452 people, grandiose narcissism went with overconfidence, impulsiveness and a willingness to ignore expert advice, which led to more bad decisions, and after a wrong answer, with blaming others and staying confident. So a challenge delivered as "your process was flawed" tends to meet a wall, and the wall gets higher.


The classic study of threatened egotism explains the temperature. In two experiments, the combination of narcissism and an insult produced unusually high aggression towards the person who delivered the insult, while simple self-esteem made no difference. The heat follows the threat to the self-image, and a public correction is the most threatening form there is.


There is a counterweight here which I think is underused. In four studies, people higher in narcissism responded to negative feedback, or to a threat to their sense of being unique, by becoming more willing to take on challenging tasks that let them demonstrate their ability, and in two of the studies by producing more creative work. A challenge can be turned into a chance to show what someone can do, and for this pattern, that is often the version that moves things.


Put together with the boardroom findings, the research points to a way of passing information up that is honest, respectful and far more likely to land.


Lead with what is true and working. Research on self-affirmation found that when people's sense of their own adequacy is secure, they are less defensive about threatening information. The flattery study says flattery does real harm, so this has to be something accurate that you respect, offered first so that the hard fact arrives as one part of a picture rather than as an attack.


Bring the outside view. Instead of arguing with the plan, add the base rate. "The last two deals like this took three years longer than planned" does not say the executive is wrong; it says what usually happens. Narcissistic chief executives were less moved by objective performance than their peers, which is a reason to make the evidence concrete, local and hard to wave away.


Say who else will see it. The same leaders were strongly moved by praise and by audiences. That cuts both ways, and it gives you a legitimate lever: reputation is a real stake. "When the board sees the payback figure, this is the question they will ask" puts the fact where it matters to the person receiving it.


Make it private and early. Given what threat does, the first time a hard fact is raised should rarely be in front of others.


Build the check into the process. This is where the most robust findings sit, and where your role has the most reach. In a study of 262 leaders and their staff, the harm done by leader narcissism to people's sense of their own worth, and through it to their willingness to speak up and help, was significant only where the leader failed to consult them; with high consultation, the harm disappeared. The hidden-profile experiment showed that narcissistic leadership stopped information flowing between people. Both point to the same move: make consultation and information-sharing part of how decisions get made, rather than something that depends on the leader's mood. Pre-reads that are due before the meeting, a round where everyone gives their view before the most senior person speaks, and a standing slot for finance before any major commitment all do this. A technique called the premortem, imagining the decision has already failed and asking why, draws on a 1989 study that found imagining an outcome as certain changed the kinds of explanation people produced, and structured disagreement of the devil's advocate kind has been studied since at least a 1990 meta-analysis. And the finding from the acquisitions research bears repeating: a slower process reduced overpaying by overconfident chief executives.


The research on issue selling, meaning how people raise issues with those above them, has long found that packaging, involvement and timing shape whether an issue gets a hearing. What the narcissism research adds is which parts of the package matter most for this listener: face, audience, evidence that cannot be waved away, and a process that does not rely on the leader choosing to listen.


I wrote about the mechanics of upward feedback in the post on feedback for a difficult boss, and about how rooms of reasonable people reach one answer in the post on conformity bias, so I will not repeat them here. The difference in this case is that you are designing around a pattern rather than a single conversation.


Try this. Rewrite the email about page fourteen as a four-line brief. One line on what is working, which you believe. One line with the fact and its source. One line on how it will look to the audience that matters. One line proposing a small process step, framed as shared responsibility: "Shall I ask finance to walk us through the payback on Thursday, before the board sees it?" Then send it to the executive alone, before anyone else has it.


Kausel and colleagues (2015) found across three studies that narcissism and advice taking were negatively related when narcissism was measured at the state level or extraversion was controlled, that disregard for others rather than confidence mediated the relationship, and in a fourth study that the relationship was strongly negative under process accountability. O'Reilly and Hall (2021) reported two studies with 452 participants showing that grandiose narcissists' overconfidence, impulsivity and willingness to ignore expert advice produced a higher likelihood of bad decisions, and that after a wrong answer they were more likely to blame others and remain confident. Bushman and Baumeister (1998) found in two studies that self-esteem was irrelevant to aggression, while the combination of narcissism and insult produced exceptionally high aggression towards the source of the insult. Kernis and Sun (1994) examined narcissism and reactions to interpersonal feedback. Nevicka, Baas and Ten Velden (2016) found across four studies that following ego threat, higher nonclinical narcissism was associated with greater willingness to perform tasks enabling demonstration of ability and with enhanced creative performance, though not with anagram performance. Cohen and Sherman (2014) reviewed research showing that self-affirmation interventions reduce defensive responses to threats to self-integrity, with benefits that sometimes persist for months and years. Carnevale, Huang and Harms (2018) studied 262 leader-employee dyads and found that leader narcissism threatened employees' organisation-based self-esteem, with negative indirect effects on promotive voice and helping and positive indirect effects on dysfunctional resistance and badmouthing, significant only when leaders failed to consult their employees. Mitchell, Russo and Pennington (1989) found that outcome certainty changed the nature of the explanations people generated for events, the basis of prospective hindsight. Schwenk (1990) meta-analysed the effects of devil's advocacy and dialectical inquiry on decision making. Dutton and colleagues (2001) examined the moves by which managers sell issues to top management and shape organisational change. Dutton and colleagues (2002) found that the exclusiveness of an organisation's culture was the most potent cue affecting female managers' willingness to raise a gender-equity issue. Morrison (2014) reviewed the research on employee voice and silence.




Protecting the team, and yourself


The costs of self-focused leadership fall mostly on the people nearby, and the research is clear about that. A meta-analysis of 57 studies of destructive leadership found it related to worse attitudes towards the leader, lower well-being and lower performance, and to higher turnover intentions, resistance and counterproductive behaviour. In a study of 211 team leaders and 1,205 of their staff, narcissistic leaders behaved more self-interestedly when they felt their organisation had treated them unfairly, and that self-interest lowered their teams' helpfulness and willingness to speak up. Across three samples, manager narcissism lowered how trustworthy staff found their manager, which in turn made them more silent, except where the manager came across as sincere. And in a study of 1,017 people working for 424 leaders, leader narcissism went with poorer working relationships, lower job satisfaction and more emotional exhaustion among staff.


That last study contains a finding I did not expect, and it belongs in any fair account. The narcissistic leaders themselves had lower well-being: they identified less with their work, were less satisfied and felt less worthwhile, and that ran through the same poor relationships. Nobody in that picture was doing well.


Now the part that matters most for you: what protected people in these studies.


Colleagues, more than anything. In a study of students and their academic advisers, support from the team buffered the effects of an abusive adviser on satisfaction and anxiety. Where the support came from the abusive adviser, by contrast, the abuse went with more anxiety and lower well-being. Warmth from the source of the strain did not soothe it; warmth from peers did.


Staying engaged rather than going quiet. Across three samples, people with narcissistic supervisors who used active, engaged strategies to get their work done saw no significant effect of the supervisor on their frustration, tension, resources or performance, while those without such strategies were affected on every measure. Three more samples found the same for people skilled at managing their own time, energy and support. In a reanalysis of a two-wave study, the link between an abusive supervisor and distress was weaker for people who dealt with it through direct communication, and stronger for those who managed the relationship by avoiding and placating. And in two studies in India, avoidance coping went with worse performance under an abusive supervisor.


Someone with influence on your side. Across three studies, including one of practising lawyers, a supervisor's outbursts went with worse job satisfaction, mood and effort only when that supervisor offered little political support; where they did back their people in the wider organisation, the outbursts had little effect. It also suggests that sponsorship from someone else can do a similar protective job.


A secure sense of yourself. Two studies found that people lower in self-esteem perceived narcissistic leaders as more abusive, and that this went with lower performance and more burnout. Nobody should be blamed for being affected; the finding is a reason to put deliberate effort into the things that keep your own footing steady, especially if you are going through a wobbly patch.


Switching off. The research on recovery from work stress puts psychological detachment, meaning mentally leaving work behind in your time off, at the centre of how people recover from demanding jobs. It matters most when the job is hardest, which is also when it is hardest to do. And a meta-analysis of three decades of emotional labour research found that the gap between the feelings a role requires you to show and the ones you have takes its toll on well-being, partly through the effort of faking. You will not always be able to close that gap at work. You can make sure it closes when you get home.


Keeping your own judgement belongs here too. The flattery study showed what a chorus of agreement does to a leader; the power studies showed that dominance works only when everyone else goes along. You do not have to be the one who argues. You do have to avoid becoming one more voice in the chorus, and the kindest version of that is being known, warmly and reliably, as the person who says what the numbers say. I wrote about how people with good values drift in the post on acting with integrity, and the loyalty trap described there is sharpest in exactly this seat.


Finally, knowing when a situation is beyond managing. A style that is self-focused, overconfident or hungry for credit is one thing. Shouting, public humiliation, threats, or being made to doubt your own memory of events is another, and the research on abusive supervision is consistent that its effects are harmful everywhere it has been studied. That is not yours to absorb or to fix alone. Keep dated notes, talk to someone you trust outside the reporting line, and use the routes your organisation provides, whether that is human resources, an employee assistance programme or a senior sponsor. If it is affecting your sleep, your health or your sense of yourself, a doctor or a counsellor is the right next call, and making it is a sign of good judgement. My post on gaslighting at work covers the specific pattern of being made to doubt yourself in more depth.


Try this. Draw a small support map on the back of an envelope. One peer who understands the work, one person with influence who would speak up for you, and one person entirely outside your organisation. If any of the three is empty, filling it is the most protective thing you can do this month, and it will serve you whatever happens with your executive.


Schyns and Schilling (2013) meta-analysed 57 studies of destructive leadership and found negative correlations with attitudes towards the leader, well-being and individual performance, and positive correlations with turnover intention, resistance towards the leader and counterproductive work behaviour. Liu and colleagues (2017) found in a multisource, time-lagged survey of 211 team leaders and 1,205 subordinates that narcissistic leaders engaged in more self-interested behaviour when they perceived organisational unfairness, which decreased followers' pro-social behaviour and voice. Hamstra and colleagues (2021) found in three samples that manager narcissism lowered perceived manager trustworthiness, which was related to increased employee silence, with the effect absent for managers high in apparent sincerity. Bernerth (2022) studied 1,017 subordinates under 424 leaders and found leader narcissism negatively related to leader-member exchange, which related to subordinates' job satisfaction and emotional exhaustion, and at the group level negatively related to leaders' own identification, job satisfaction and perceived self-worth through average leader-member exchange. Kim and colleagues (2021) studied 200 middle managers and 318 working adults and found that perceived chief executive narcissism raised managers' uncertainty, which led to laissez-faire leadership and impression management as coping responses. Hobman and colleagues (2009) found that team member support buffered the relationship between abusive supervision and satisfaction and anxiety, while support from the abusive adviser exacerbated anxiety and lowered well-being. Hochwarter and Thompson (2012) found across three samples that perceived supervisor narcissism increased frustration and tension and reduced resources and performance for employees low in engaged enactment behaviour, with no significant effect for those high in it. Ellen and colleagues (2019) found across three samples of US workers (187, 199 and 136) that subordinates' resource management ability attenuated the effects of supervisor narcissism on emotional exhaustion, job tension, depressed mood, task performance and citizenship behaviour. Tepper and colleagues (2007) found that the relationship between abusive supervision and psychological distress was exacerbated by regulative maintenance communication and reduced by direct maintenance communication. Nandkeolyar and colleagues (2014) found across two studies in India that avoidance coping facilitated a negative relationship between abusive supervision and performance, and that conscientiousness weakened the relationship. Mawritz, Dust and Resick (2014) examined how employees cope with hostile climates and abusive supervision, and the role of conscientiousness. Jordan and colleagues (2020) found across three studies, including 156 practising lawyers and 161 municipal employees, that supervisor narcissistic rage was associated with unfavourable outcomes only when supervisor political support was low. Nevicka and colleagues (2018b) found in studies of 85 leaders with 128 followers and of 177 leader-follower dyads that followers low in self-esteem or core self-evaluations perceived narcissistic leaders as more abusive, which related to lower performance and more burnout. Sonnentag and Fritz (2015) set out the stressor-detachment model of recovery from job stress. Hülsheger and Schewe (2011) meta-analysed three decades of emotional labour research and found that surface acting partially mediated the relationship between emotion-rule dissonance and well-being. Park, Westphal and Stern (2011) linked flattery and opinion conformity to chief executive overconfidence and strategic persistence. Tost, Gino and Larrick (2013) found that leaders' verbal dominance hurt team performance through the acquiescence of other members. O'Boyle and colleagues (2012) meta-analysed 245 samples (43,907 people) and found counterproductive work behaviour associated with all three Dark Triad traits, moderated by authority and culture. Grijalva and Newman (2015) updated that meta-analysis and found narcissism the largest unique Dark Triad predictor of counterproductive work behaviour after controlling for the Big Five. Schmid, Pircher Verdorfer and Peus (2019) developed a measure of exploitative leadership across six studies with 1,722 people. Mackey and colleagues (2017) meta-analysed abusive supervision research and found its associations with outcomes universally negative, with magnitudes varying by study design. Tepper, Simon and Park (2017) reviewed the causes and consequences of abusive supervision.




The steadiness check


Everything in this post comes down to a working question: given what I am seeing, where should my effort go first? So here is a tool for that question.


Describe the pattern in behaviour rather than labels, answer a few questions about the setting, the way information reaches your executive and how you are doing yourself, and it will suggest where to focus, what to try, what to check in a month, and give you a short note to keep. Nothing you type leaves the page, and it never asks you to name anyone.





Where to start


If you take one thing from all of this, make it the move from person to pattern. The research is remarkably consistent that the behaviour does the harm and that power, success and a lack of friction draw it out, which means your most effective work is on the situation around your executive: the information that reaches them, the process decisions go through, and the support around you and your team.


After that, there are three moves worth making.


When you pass up a hard fact, lead with what is true and working, bring the base rate, and say who will see it, in private and early. When a big decision is coming, add a step before commitment where someone with no stake looks at the numbers, and slow the tempo if you can. And for yourself, keep your support map full and your evenings your own, and know where your line is before you reach it.


I have put the three pages I would use into a download: a one-page brief for getting a hard fact heard, a decision check for the big irreversible calls, and a steadiness plan for keeping your own footing.



A graphic titled Working beside a very confident leader, on a page, with the eyebrow Five moves, the sentence each one needs, and the study behind it. Standfirst: Each of these fits inside work you are already doing. The right-hand column is what the evidence says the move buys you. A table of 5 moves, each with the sentence to say and the evidence behind it. The pattern, Behaviour, not labels. Say: Announced the decision before anyone had spoken, with the chair in the room. Evidence: What harmed followers was the behaviour they saw, not the leader's own trait score. The brief, Private and early. Say: The pipeline is strong. The last two deals like this took three years longer to pay back. Evidence: Secure people are less defensive; the base rate is harder to wave away than an opinion. The audience, Who will see it. Say: When the board sees the payback figure, this is the question they will ask. Evidence: Narcissistic chief executives were moved less by the numbers and more by praise and attention. The tempo, A check before commitment. Say: I have held forty minutes on Thursday for finance to walk us through the model. Evidence: Overconfident chief executives overpaid less when the process ran more slowly. Your footing, Peers, sponsor, outside. Say: One colleague, one person with influence, one person outside work. Evidence: Support from colleagues buffered the strain, and staying engaged did better than going quiet. Foot: Fehn and Schütz (2021); Cohen and Sherman (2014); Chatterjee and Hambrick (2011); Pavićević and Keil (2021).
Five moves, none of which require winning an argument.

Achilles, for what it is worth, was not undone by his confidence; he was undone by the one place nobody had thought to protect. Most of the leaders in this research are not tragic heroes either. They are capable people in positions that quietly change them, surrounded by people who find it easier to agree. The seat beside them is one of the few places from which that can be seen early and handled kindly, and I have always thought that is one of the quiet privileges of this work.


The part of this that shows up in how you are seen at work sits inside what the Connected Leader Assessment calls Influence and Stakeholder Trust: whether the people you work with take your information seriously, whether you can shape a decision without owning it, and whether you are trusted by the executive and the team at the same time. It is the dimension that decides whether you are heard when it matters most.


The assessment is free, takes about twelve minutes and gives you a read across all five dimensions. If you want to go further on this one, the Skill Sprints are short, focused programmes on a single dimension, and the Irreplaceable Audit is the place to look at the judgement and the trusted voice you bring to the decisions around your executive, which is exactly what this kind of leadership needs most and values least.


Stay savvy.


Meg ✌️




Free tools


  • The steadiness check, the interactive tool in this post, free

  • The Steady Hand Pages, the download in this post, free

  • The Connected Leader Assessment, twelve minutes, free, with a read on Influence and Stakeholder Trust, the dimension this post sits in




P.S. Want to Go Deeper on Narcissistic and Hubristic Leaders?


If this resonated and you want to go further, here is the evidence the piece is built on. These are the papers that shaped my thinking on what power and confidence do to leaders, and how to stay steady and useful beside one, and they are well worth your time.


Every DOI below has been checked. Where I could not confirm a figure from the paper itself, I have left the figure out rather than guess.


📃 Grijalva and colleagues (2015) Meta-analysis. Narcissism predicted emerging as a leader, not leading well; a middling level did best. https://doi.org/10.1111/peps.12072


📃 Brunell and colleagues (2008) Three studies of leaderless groups. The more narcissistic people took charge, executives included. https://doi.org/10.1177/0146167208324101


📃 Nevicka and colleagues (2011) Narcissistic leaders looked effective and their groups did worse, because information stopped flowing. https://doi.org/10.1177/0956797611417259


📃 Chatterjee and Hambrick (2007) 111 chief executives. Narcissism meant bolder strategy and bigger swings, and no better or worse on average. https://doi.org/10.2189/asqu.52.3.351


📃 Chatterjee and Hambrick (2011) Narcissistic chief executives were moved less by the numbers and more by praise and awards. https://doi.org/10.1177/0001839211427534


📃 Ham, Seybert and Wang (2018) Narcissism measured from signature size, linked to investment and performance. https://doi.org/10.1007/s11142-017-9427-x


📃 Rijsenbilt and Commandeur (2013) Chief executive narcissism and corporate fraud. https://doi.org/10.1007/s10551-012-1528-7


📃 Owens, Wallace and Waldman (2015) Narcissistic leaders who also showed humility had more engaged, better-performing staff. https://doi.org/10.1037/a0038698


📃 Campbell and colleagues (2010) Review of narcissism in organisations, its costs and its benefits. https://doi.org/10.1016/j.hrmr.2010.10.007


📃 Rosenthal and Pittinsky (2006) An early review of narcissistic leadership. https://doi.org/10.1016/j.leaqua.2006.10.005


📃 Braun (2017) Review of 45 studies of leader narcissism at three levels. Findings still pull both ways. https://doi.org/10.3389/fpsyg.2017.00773


📃 Back and colleagues (2013) Seven studies separating narcissistic admiration from narcissistic rivalry. https://doi.org/10.1037/a0034431


📃 Fatfouta (2019) Review of the facets of narcissism in leaders and how to select and develop them. https://doi.org/10.1016/j.hrmr.2018.10.002


📃 O'Boyle and colleagues (2012) Meta-analysis of 245 samples. All three dark traits went with counterproductive behaviour at work. https://doi.org/10.1037/a0025679


📃 Owen and Davidson (2009) Hubris syndrome proposed as an acquired condition triggered by power, and fading with it. https://doi.org/10.1093/brain/awp008


📃 Hayward and Hambrick (1997) 106 large acquisitions. Hubris meant bigger premiums and bigger losses, worst with a weak board. https://doi.org/10.2307/2393810


📃 Roll (1986) The original hubris hypothesis: bidders pay too much because they trust their own valuation. https://doi.org/10.1086/296325


📃 Malmendier and Tate (2005) Overconfident chief executives invested more when cash was plentiful. https://doi.org/10.1111/j.1540-6261.2005.00813.x


📃 Malmendier and Tate (2008) Overconfident chief executives made more acquisitions, and markets liked them less. https://doi.org/10.1016/j.jfineco.2007.07.002


📃 Hiller and Hambrick (2005) Hubris as a very high core self-evaluation, and what it does to strategic decisions. https://doi.org/10.1002/smj.455


📃 Sadler-Smith and colleagues (2017) Review of hubristic leadership and how it differs from overconfidence, narcissism and pride. https://doi.org/10.1177/1742715016680666


📃 Asad and Sadler-Smith (2020) Hubrists are intoxicated with power and success; narcissists use power to stay intoxicated with themselves. https://doi.org/10.1177/1742715019885763


📃 Ouimet (2010) Four families of triggers for narcissistic leadership: the person, culture, environment and structure. https://doi.org/10.1108/02683941011075265


📃 Keltner, Gruenfeld and Anderson (2003) Power switches people towards reward and action; its lack, towards threat and other people. https://doi.org/10.1037/0033-295x.110.2.265


📃 Galinsky and colleagues (2006) Power made people worse at taking other people's perspectives, down to the letter E on the forehead. https://doi.org/10.1111/j.1467-9280.2006.01824.x


📃 See and colleagues (2011) Four studies. More power, more confidence, less advice taken, and less accuracy. https://doi.org/10.1016/j.obhdp.2011.07.006


📃 Tost, Gino and Larrick (2012) The powerful discounted expert and novice advice alike; feeling cooperative reversed it. https://doi.org/10.1016/j.obhdp.2011.10.001


📃 Tost, Gino and Larrick (2013) A leader's felt power made them dominate talk, which cut team communication and performance. https://doi.org/10.5465/amj.2011.0180


📃 Fast and colleagues (2012) Five experiments. Power produced costly overconfidence, unless the powerful felt out of their depth. https://doi.org/10.1016/j.obhdp.2011.11.009


📃 Anderson and Galinsky (2006) Power made risk look smaller, except when people felt responsible for others. https://doi.org/10.1002/ejsp.324


📃 Moore and Healy (2008) Three kinds of overconfidence, and why being too precise is the most persistent. https://doi.org/10.1037/0033-295x.115.2.502


📃 Owens, Johnson and Mitchell (2013) Eight samples. Leaders' expressed humility went with engagement and retention. https://doi.org/10.1287/orsc.1120.0795


📃 Ou and colleagues (2014) 63 companies. Humble chief executives empowered their top teams, and it reached middle managers. https://doi.org/10.1177/0001839213520131


📃 Zhang and colleagues (2017) Humility and narcissism in the same chief executive, and what the mix did for innovation. https://doi.org/10.1016/j.leaqua.2017.01.003


📃 Petrenko and colleagues (2016) Narcissistic chief executives did more social responsibility, and it did less for performance. https://doi.org/10.1002/smj.2348


📃 Gerstner and colleagues (2013) Narcissistic chief executives bet harder on new technology when the world was watching. https://doi.org/10.1177/0001839213488773


📃 Wales, Patel and Lumpkin (2013) 173 chief executives. Narcissism went with entrepreneurial posture and more variable results. https://doi.org/10.1111/joms.12034


📃 Nevicka and colleagues (2018a) 175 stores. Narcissistic leaders looked good from a distance; up close, the advantage vanished. https://doi.org/10.1037/apl0000298


📃 Ong and colleagues (2016) Twelve weeks of groups. The narcissistic leader's honeymoon faded as people got to know them. https://doi.org/10.1111/jopy.12155


📃 Paulhus (1998) Self-enhancement as a mixed blessing: good first impressions, and then what follows. https://doi.org/10.1037/0022-3514.74.5.1197


📃 Leckelt and colleagues (2015) 311 people over three weeks. Charm faded, rivalry cost more, and popularity fell. https://doi.org/10.1037/pspp0000057


📃 Judge, LePine and Rich (2006) Narcissists rated their own leadership higher; other people rated it lower. https://doi.org/10.1037/0021-9010.91.4.762


📃 Schyns and Schilling (2013) Meta-analysis of 57 studies. Destructive leadership harmed well-being, performance and retention. https://doi.org/10.1016/j.leaqua.2012.09.001


📃 Hogan and Hogan (2001) Managerial derailment and the dark side of leaders' personality. https://doi.org/10.1111/1468-2389.00162


📃 Morrison (2014) Review of why people speak up to those above them, or stay silent. https://doi.org/10.1146/annurev-orgpsych-031413-091328


📃 Cohen and Sherman (2014) Self-affirmation lowers defensiveness when information threatens how people see themselves. https://doi.org/10.1146/annurev-psych-010213-115137


📃 Nevicka and colleagues (2013) In uncertain times people chose narcissistic leaders, knowing the downsides. https://doi.org/10.1002/ejsp.1943


📃 Kahneman and Lovallo (1993) The inside view breeds bold forecasts; the outside view corrects them. https://doi.org/10.1287/mnsc.39.1.17


📃 Mitchell, Russo and Pennington (1989) Imagining an outcome as certain changed how people explained it: the root of the premortem. https://doi.org/10.1002/bdm.3960020103


📃 Schwenk (1990) Meta-analysis of devil's advocacy and dialectical inquiry in decision making. https://doi.org/10.1016/0749-5978(90)90051-a


📃 Garrard and colleagues (2014) Changes in political leaders' language as hubris develops. https://doi.org/10.1016/j.cortex.2013.08.014


📃 Picone, Dagnino and Minà (2014) Review of the hubris hypothesis, with a good side and a bad side. https://doi.org/10.5465/amp.2012.0177


📃 Bollaert and Petit (2010) Review of research on the dark side of executive psychology. https://doi.org/10.1016/j.emj.2010.01.001


📃 Tang and colleagues (2015) S&P 1500 firms. Hubris went with less social responsibility and more irresponsibility. https://doi.org/10.1002/smj.2286


📃 Li and Tang (2010) 2,790 chief executives in China. Hubris meant more risk, most where they had most freedom. https://doi.org/10.5465/amj.2010.48036912


📃 Tepper and colleagues (2007) Talking directly eased distress under an abusive supervisor; placating made it worse. https://doi.org/10.5465/amj.2007.20159918


📃 Mawritz, Dust and Resick (2014) Hostile climates, abusive supervision, and how conscientious people cope. https://doi.org/10.1037/a0035863


📃 Hobman and colleagues (2009) Support from the team buffered an abusive adviser; support from the adviser did not. https://doi.org/10.1111/j.1464-0597.2008.00330.x


📃 Mackey and colleagues (2017) Meta-analysis of abusive supervision. Its links with outcomes were negative everywhere. https://doi.org/10.1177/0149206315573997


📃 Schmid, Pircher Verdorfer and Peus (2019) Six studies, 1,722 people. Exploitative leadership defined and measured. https://doi.org/10.1177/0149206317707810


📃 Lilienfeld, Miller and Lynam (2018) The debate over diagnosing public figures nobody has examined. https://doi.org/10.1177/1745691617727864


📃 Miller and colleagues (2017) Narcissism as a dimension, with grandiose and vulnerable forms. https://doi.org/10.1146/annurev-clinpsy-032816-045244


📃 Goncalo, Flynn and Kim (2010) Narcissists were not more creative, and were better at persuading people they were. https://doi.org/10.1177/0146167210385109


📃 Nevicka and colleagues (2018b) Followers low in self-esteem found narcissistic leaders more abusive, and burned out more. https://doi.org/10.3389/fpsyg.2018.00422


📃 Kaiser, LeBreton and Hogan (2015) 320 managers, 4,906 raters. Middling dark-side scores went with the best leadership behaviour. https://doi.org/10.1111/apps.12024


📃 Judge, Piccolo and Kosalka (2009) The bright and dark sides of leader traits. https://doi.org/10.1016/j.leaqua.2009.09.004


📃 Resick and colleagues (2009) 75 chief executives in baseball. Personality, leadership style and results. https://doi.org/10.1037/a0016238


📃 Harms, Spain and Hannah (2011) Leader development and the dark side of personality. https://doi.org/10.1016/j.leaqua.2011.04.007


📃 O'Reilly and colleagues (2014) Narcissistic chief executives and what they are paid. https://doi.org/10.1016/j.leaqua.2013.08.002


📃 O'Reilly, Doerr and Chatman (2018) Narcissistic chief executives meant more and longer lawsuits, and less appetite for expert advice. https://doi.org/10.1016/j.leaqua.2017.08.001


📃 Liu and colleagues (2017) 211 leaders, 1,205 staff. Leaders who felt unfairly treated turned self-interested, and staff went quiet. https://doi.org/10.1037/apl0000237


📃 Nevicka, Baas and Ten Velden (2016) After a knock to the ego, narcissism fuelled effort on tasks that showed off ability. https://doi.org/10.1111/jopy.12223


📃 Bushman and Baumeister (1998) Narcissism plus an insult produced aggression towards the source; self-esteem did not matter. https://doi.org/10.1037/0022-3514.75.1.219


📃 Morf and Rhodewalt (2001) A grandiose yet fragile self, in constant need of outside confirmation. https://doi.org/10.1207/s15327965pli1204_1


📃 Kausel and colleagues (2015) Narcissists dismissed advice because they rated advisers low, and more so when made to justify themselves. https://doi.org/10.1016/j.obhdp.2015.07.006


📃 Hogeveen, Inzlicht and Obhi (2014) Power reduced the brain's automatic mirroring of other people's actions. https://doi.org/10.1037/a0033477


📃 Akstinaite, Robinson and Sadler-Smith (2020) Linguistic markers of hubris in chief executives. https://doi.org/10.1007/s10551-019-04183-y


📃 Park, Westphal and Stern (2011) Flattery and agreement made chief executives overconfident and slower to change course. https://doi.org/10.1177/0001839211429102


📃 Chen, Crossland and Luo (2015) Firms with overconfident chief executives were slower to learn from forecasting mistakes. https://doi.org/10.1002/smj.2291


📃 Owens and Hekman (2012) 55 interviews. What humble leaders do, and why it helps the people around them. https://doi.org/10.5465/amj.2010.0441


📃 Claxton, Owen and Sadler-Smith (2015) Unbridled intuition as an overlooked source of hubris. https://doi.org/10.1177/1742715013511482


📃 Benson and Campbell (2007) Two samples of managers. Dark-side traits and performance followed an inverted U. https://doi.org/10.1111/j.1468-2389.2007.00384.x


📃 Grijalva and Newman (2015) Updated meta-analysis. Narcissism the strongest dark-trait predictor of counterproductive work. https://doi.org/10.1111/apps.12025


📃 Nandkeolyar and colleagues (2014) Avoidance coping made an abusive supervisor's effect on performance worse. https://doi.org/10.1037/a0034262


📃 Hamstra and colleagues (2021) Three samples. Narcissistic managers seemed less trustworthy and staff went silent, unless they seemed sincere. https://doi.org/10.1111/joop.12337


📃 Sedikides and Campbell (2017) The arc of a narcissistic leader: disruption, clash, then exit or negotiated controls. https://doi.org/10.1177/1745691617692105


📃 O'Reilly and Chatman (2020) How much transformational leadership overlaps with grandiose narcissism. https://doi.org/10.1177/0008125620914989


📃 Craig and Amernic (2018) Looking for language markers of hubris in letters to shareholders. https://doi.org/10.1007/s10551-016-3100-3


📃 Tepper, Simon and Park (2017) Review of the causes and consequences of abusive supervision. https://doi.org/10.1146/annurev-orgpsych-041015-062539


📃 Grapsas and colleagues (2020) Narcissism as a moment-by-moment pursuit of status. https://doi.org/10.1177/1745691619873350


📃 Kakkar and Sivanathan (2017) Economic uncertainty lowered people's sense of control and raised the appeal of dominant leaders. https://doi.org/10.1073/pnas.1617711114


📃 Smith and colleagues (2018) Review of the upsides of dark traits and the downsides of bright ones. https://doi.org/10.1177/0149206317733511


📃 Ellen and colleagues (2019) Three samples. People good at managing their own resources were protected from a narcissistic boss. https://doi.org/10.1007/s10551-017-3666-4


📃 Hochwarter and Thompson (2012) Three samples. Staying actively engaged neutralised a narcissistic supervisor's effect. https://doi.org/10.1177/0018726711430003


📃 Carnevale, Huang and Harms (2018) 262 pairs. Leader narcissism harmed staff only where the leader did not consult them. https://doi.org/10.1016/j.obhdp.2018.04.003


📃 Fehn and Schütz (2021) What harmed followers was rivalry as they saw it, not as the leader rated it. https://doi.org/10.1007/s10551-020-04604-3


📃 Bernerth (2022) 1,017 people under 424 leaders. Narcissism hurt relationships, and the leaders' own well-being. https://doi.org/10.1007/s10551-020-04595-1


📃 Hayward, Rindova and Pollock (2004) How press celebrity can persuade a chief executive of their own genius. https://doi.org/10.1002/smj.405


📃 Burkhard and colleagues (2023) Meta-analysis of 199 studies. Chief executive overconfidence was, on average, good for performance. https://doi.org/10.1177/01492063221110203


📃 Billett and Qian (2008) First deals were neutral, later ones negative: success credited to skill built overconfidence. https://doi.org/10.1287/mnsc.1070.0830


📃 Pavićević and Keil (2021) A slower pre-deal process cut what overconfident chief executives overpaid. https://doi.org/10.1002/smj.3319


📃 Brown and Sarma (2007) Chief executive dominance mattered as much as overconfidence in the decision to acquire. https://doi.org/10.1016/j.jeconbus.2007.04.002


📃 Jordan and colleagues (2020) A boss's rage hurt less when that boss also backed their people politically. https://doi.org/10.1108/jmp-08-2019-0474


📃 Cragun, Olsen and Wright (2020) Review and meta-analysis. Five ways to measure chief executive narcissism, and mixed findings. https://doi.org/10.1177/0149206319892678


📃 Chatterjee and Pollock (2017) How narcissistic chief executives shape the professional world around them. https://doi.org/10.5465/amr.2015.0224


📃 Dutton and colleagues (2001) How managers sell issues to top management, and the moves that matter. https://doi.org/10.2307/3069412


📃 Dutton and colleagues (2002) How people read the context before deciding whether to raise an issue. https://doi.org/10.1287/orsc.13.4.355.2949


📃 Sonnentag and Fritz (2015) Mentally leaving work behind as the heart of recovering from job stress. https://doi.org/10.1002/job.1924


📃 Hülsheger and Schewe (2011) Meta-analysis of emotional labour. Faking feelings carried a cost to well-being. https://doi.org/10.1037/a0022876


📃 Kernis and Sun (1994) Narcissism and reactions to feedback from other people. https://doi.org/10.1006/jrpe.1994.1002


📃 De Wit and colleagues (2017) Leaders who saw power as responsibility took more advice than those who saw it as opportunity. https://doi.org/10.1002/job.2171


📃 O'Reilly and Hall (2021) 452 people. Grandiose narcissists ignored experts, decided worse and blamed others. https://doi.org/10.1016/j.paid.2020.110280


📃 Kim and colleagues (2021) Middle managers facing a narcissistic chief executive felt uncertain and stepped back or performed. https://doi.org/10.1037/apl0000965




P.P.S. Why Do I Even Have the Nerve to Write This?


Fair question.


I am someone who has spent years working alongside remarkable executives, navigating chaos, translating vision into structure and figuring things out, sometimes beautifully and sometimes the very hard way. I have had the privilege of being the steady right hand to leaders who move fast, take risks and expect a lot. That environment has shaped me more than any classroom ever could.


I also happen to have an MBA, and I am studying psychology because people fascinate me. How we work. Why we disconnect. What actually holds us together when the pressure rises.


But really, none of that is the point.


I am here because I have lived the erosion and rebuilt from it. I have seen the cost of carrying too much and the relief that arrives when you finally stop disappearing inside your own competence.


So I share what I have learned in case it helps someone else. Take what is useful. Leave the rest. And remember this one truth that professionals like you often forget:


You are probably doing far better than you give yourself credit for.


Comments

Commenting on this post isn't available anymore. Contact the site owner for more info.
Post: Blog2_Post
bottom of page