Who could resist an article with this title—Are Narcissistic CEOs All That Bad?—from the Rock Center for Corporate Governance at Stanford University? Our lived experience with narcissists tends to suggest that they can have a corrosive effect, so we presume that narcissistic CEOs would have a negative effect on their companies as well. Maybe not so much, according to the authors. While it’s often believed that “CEO narcissism is highly prevalent, and considerable research suggests that narcissism is associated with worse outcomes,” the authors suggest that that research relies on “indirect evidence thought to be indicative of narcissism.” Instead, for this paper, the authors base their views on more direct assessments of personality and come up with some “unexpected associations.”
The authors describe narcissism (based on “Narcissistic Personality Disorder” from the Mayo Clinic as well as Exhibit 1 to the paper) as a personality disorder “characterized by an inflated sense of self-importance, an excessive need for attention and admiration, and lack of empathy.” Not exactly the personality type that you would associate with good corporate governance. But, the authors observe, other traits common to narcissists are also traits often attributed to executives successful at climbing the corporate ladder: “self-confidence, risk tolerance, a focus on goal achievement, and more extraverted personalities.” Sometimes, though, it’s tough to distinguish between healthy self-confidence and narcissism.
The authors describe all kinds of psychological inventories that are used to assess narcissism, but they usually require completion of questionnaires, which can be a challenge in a corporate setting. Instead, researchers have looked to “unobtrusive” measures that rely on observations, such as attention to speech patterns and video observation. Some studies that have relied on unobtrusive observations have found that “narcissistic CEOs exhibit more extreme performance outcomes, make more aggressive strategy bets, make worse acquisitions, have lower quality earnings, and receive excessive pay”—hardly favorable outcomes.
But would the result be the same if the assessment were based on a “validated instrument”? That’s what the authors set out to examine by looking at a sample of 179 CEOs who were “formally assessed by long-time directors who know and have worked closely with that CEO.” Noting that “[r]esearch finds that self-reported and third-party observed personality
test results tend to be similar,” the authors contend that directors “who work closely with an executive over many years and observe firsthand how this individual makes decisions, interacts with others, treats subordinates and also treats internal and external constituents should be well-situated to make an informed assessment of that person’s personality.” In this sample, directors reported knowing the CEO for an average of 6.1 years (5 median), with 16% claiming to know the CEO “extremely well,” 48% “very well,” and 31% “moderately well.” Those whose level of familiarity with the CEO was considered inadequate were excluded.
The authors used a personality inventory called the “Big Five Model,” which tests for extraversion, agreeableness, conscientiousness, emotional stability and openness. Most CEOs in the sample scored high on conscientiousness and openness. While there was more variability for the other dimensions, the authors found that only 14% of the CEO sample were introverted, 12% disagreeable and only 9% neurotic (meaning low in emotional stability). Under the Big Five Model, “studies have shown an association between grandiose narcissism and individuals scoring high in extraversion and low in agreeableness.” Using this association, the authors found that 8% of their CEO sample satisfied these criteria.
The authors also examined the results of a Narcissistic Personality Inventory, which rates an individual’s level of narcissism on a 7-point scale. This test measures for attributes of authority, self-sufficiency, superiority, exhibitionism (need to be the center of attention), exploitativeness, vanity and entitlement. While it is estimated that about 5% of the U.S. is narcissistic, the NPI showed that the incidence of narcissism among CEOS might be three times higher—“upwards of 18 percent of CEOs might be considered narcissist.” Nevertheless, only 2% of the sample scored over 6 on the 7-point NPI.
The CEOs were then assessed on a “humility scale.” Surprise, the median score for the CEOs in the sample was very humble (6 out of 7), with only 2% registering below 3 on the humble scoreboard. However, the authors acknowledge, assessing whether the CEOs are just feigning humility can be a challenge, since that’s a talent at which narcissists (“covert narcissists”) are particularly adept.
But what about the impact on governance? The authors looked at company factors such as stock-price performance, ESG ratings, governance quality and compensation. With regard to stock price performance, the authors’ findings were consistent with the predicted outcomes. They found that, for companies with CEOs with higher narcissism levels (NPI above 3.5), the median annualized stock-price performance was 4% compared with 27% for CEOs with lower narcissism scores (NPI below 3.5). In addition, companies with more narcissistic CEOs underperformed the S&P 500 by 12% annualized, compared to outperformance of 11% annualized for less narcissistic CEOs.
But the answer was surprisingly different for ESG ratings. Presumably, narcissists would concern themselves primarily with financial gain, not environmental and social issues. However, the authors found that the most recent median Refinitiv ESG score for companies with more narcissistic CEOs was 54, compared with only 45 for less narcissistic CEOs. Are narcissist CEOs concerned with social responsibility? Perhaps, but, citing several studies, the authors suggest that narcissistic CEOs may just want to attract positive media attention with some useful virtue-signaling. Alternatively, it may be that ESG ratings are not very accurate.
And there was another surprise when it came to good governance. The authors examined the presence of charter or bylaw provisions generally thought to be shareholder-friendly and to reflect good governance. The authors found that, with one exception, companies led by narcissistic CEOs were more likely to have charters and bylaws with these good governance features—they were more likely not to have a staggered board, more likely to permit shareholder action by written consent and more likely to require only simple majorities to amend the company’s charter and bylaws. What’s the big exception? Multi-class share structures with unequal voting rights. The authors found that narcissistic CEOs are “slightly more likely to lead companies with unequal share voting rights.” But do narcissistic CEOs actually favor good governance and protection of shareholder rights? Maybe, but, the authors speculate, it may also be true that “narcissistic CEOs are better able to ‘control’ the board or reduce its influence, thereby permitting governance friendly features that do not substantively lead to an increase in governance quality.” (Or maybe having multi-class share structures with unequal voting rights provides the CEOs with enough control to avoid concern with other features such as simple majorities and written consent.)
The last factor the authors examined was CEO pay—and no shockers there. The authors’ data showed that “narcissistic CEOs receive median total compensation that is 33 percent higher ($10.34 million) than CEOs that are not narcissistic ($7.79 million).” Before you jump to conclusions, however, consider that, at the same companies, the second most highly paid executives in the company also were paid significantly more, which could have a somewhat neutralizing effect; the ratios between the CEO and the second highest paid executives were quite close for companies with narcissistic CEOs (2.1) compared to less narcissistic CEOs (2.3). So is the compensation distributed more equally among executives to achieve a level of fairness and equity or could the second-level executives be paid at a higher level to “coopt their loyalty rather than to reward them for performance”?
Accordingly, among the factors studied by the authors, only stock price was shown to be adversely associated with CEO narcissism; although not exactly benefits, all the others factors proved to be slightly positive or neutral. So narcissistic CEOs may not be so bad after all. But then, the authors looked only at stock price and governance issues—not at corporate culture or strategic innovation, CEO leadership or employee satisfaction, also important issues. Happily, while there may be more narcissistic CEOs than narcissists among the general population, the authors assure us that, contrary to conventional wisdom, the “overwhelming majority of CEOs appear to have a fairly healthy personality profile of agreeableness, sociability, conscientiousness, and humility.” In the end, the authors leave us with these questions: “Just how much contribution does an individual CEO make to company performance? How does personality contribute to this relationship?” And then, for directors, how much do they even discuss “the personality of their CEO and its impact on the firm? What action do directors take, if any, when they identify personality tendencies that might be of concern, such as narcissism, disagreeableness, or low emotional stability?”