“Ask
chief executives why their companies are performing so well, and
they’ll typically credit a brilliant strategy coupled with hard-nosed,
diligent execution. When you ask Lars Sørensen of Novo Nordisk what
forces propelled him to the top of HBR’s 2015 ranking of the
best-performing CEOs in the world, he cites something very different:
luck.”
So begins our recent profile of the best performing CEO of 2015. Sørensen’s modesty is refreshing, but is it accurate?
A series of recent papers help answer that question, by quantifying
the roles of luck, ability, and experience in CEOs’ success. Together
they suggest two conclusions: first, no single trait or skill seems to
explain CEO performance; and second, luck plays a very large role.
The CEO, then, is not the primary driver of a firm’s success. Industry, for instance, matters far more. But what factors determine which CEOs rise and fall, and who makes it to the top job in the first place?
It shouldn’t take a careful empirical study to convince you that CEOs don’t get where they are on the basis of ability alone. If that were true, the C-suite would not be so dominated by white men. Nonetheless, a new working paper by Renee Adams at the University of New South Wales, Matti Keloharju at Harvard Business School, and Samuli Knupfer at BI Norwegian Business School helps to quantify the role of ability in becoming CEO.
The researchers used data from the Swedish military to compare Swedish men who ended up becoming CEOs to men who didn’t. Military service was mandatory in Sweden between 1970 and 1996, and the government attempted to measure young men’s aptitude using a series of cognitive tests, as well as an interview. The researchers were able to match this data with other sources to determine the professions these men eventually entered, and how well they were paid.
Their results are what you might expect: CEOs are, on average, smarter than most people. But they’re hardly exceptional. The median CEO of a large firm in Sweden scored in the top 17% in “cognitive ability,” a measure based on four tests given by the military to assess inductive reasoning, verbal comprehension, spatial ability, and technical comprehension.
(The researchers don’t argue that such ability is necessarily innate, as opposed to shaped by one’s upbringing; they just note that the scores predate any leadership experience.)
CEOs of large firms scored higher on this measure than CEOs of small firms and family firms. But large firm CEOs didn’t score much higher than other professionals, like doctors, lawyers, or engineers.
Success in the military interview was a slightly better predictor of becoming a CEO, compared to cognitive tests. As the researchers explain, “Conscripts obtain a higher score in the interview when they demonstrate that they have the willingness to assume responsibility, are independent, have an outgoing character, demonstrate persistence and emotional stability, and display initiative.” Perhaps this represents a measure of the things CEOs need beyond smarts; perhaps it reflects bias in the interview process.
Overall this evidence suggests that CEOs are a relatively talented bunch, but not much more so than other skilled professionals, or for that matter than the executives who report to them.
“These results suggest that the high pay CEOs enjoy does not arise from scarce supply of the three traits we study,” the researchers conclude, noting that Swedish CEOs receive a 1200% pay premium compared to the average worker. “While a favorable mix of traits may make it easier to climb on the corporate ladder, it by no means assures a position as a chief executive of a major company.”
What else shapes the careers of successful CEOs? It’s hard to avoid Sørensen’s conclusion that luck is the single biggest factor. Consider a 2014 study by Dirk Jenter at Stanford and Fadi Kanaan at MIT: they looked at more than 3,000 CEO turnovers between 1993 and 2009 and found that CEOs were frequently fired for factors well outside their control. Recall that in the CEO effect studies, industry and macroeconomic factors tended to matter more for financial success than who was running the company. Yet, Jenter and Kanaan found that CEOs were significantly more likely to be dismissed during recessions or when their industry is suffering, despite the fact that these trends have little to do with managerial skill. (Lest you start to feel bad for CEOs, if anything they are more likely to be compensated for good luck than fired for bad.)
Perhaps the most sweeping indictment of the idea that CEOs’ careers are shaped by skill comes from a 2014 paper by Markus Fitza at Texas A&M. He argues that even that 2% to 22% of firm performance that studies have attributed to CEOs could be largely the product of luck. To understand why, you have to dive into the methodology of all this research, but I promise it will be worth it.
Studies of the CEO effect usually look at how much CEO turnover correlates with a company’s financial performance, after accounting for other factors like industry performance, overall economic performance, and the like. If a firm does badly for four years under one CEO, then another CEO comes in under the same economic conditions and the firm does well for the next four years, that gets counted as the CEO effect.
But that’s not necessarily proof of CEO skill. Imagine that when CEOs take the top job, they get to pick between two strategies, and once they pick, that strategy is in place for their full tenure. One strategy leads to success, the other to failure, but no one knows which strategy is which. Perhaps the first CEO in the example above just randomly happened to pick the bad strategy, and the next CEO happened to pick the good one. In that case, the CEO effect would just be measuring luck.
That example is purposely oversimplified, but CEOs clearly do face uncertainty when choosing a strategy. If a new CEO coming in and changing course is more like a dice roll than an indication of competence, then measurements of the CEO effect reflect luck rather than skill.
Fitza’s contribution is to estimate just how confident we can be that the CEO effect is due to skill rather than luck. To do so he simulates how big a CEO effect you might see in the data if company performance were entirely random. He shows that, because of how few data points exist about any single CEO, randomness can create the illusion of a CEO effect, even when no skill is involved. Fitza calculates that the portion of the CEO effect that we can be confident is related to skill is between 4% and 5% of total company performance. (The study that found the 2% to 22% range tried a similar approach and came to the conclusion that in some industries the CEO effect was large enough to rule out luck, while in others it was not.)
What should we make of all this?
It’s safe to say that CEOs are, overall, a talented bunch, but that’s not what separates them from other professionals, nor is it the main reason their firms succeed or fail. Certainly it doesn’t come close to explaining why they’re so well paid. Put another way, CEOs matter, just less than many people think. Instead, luck, and yes, bias, play a far larger role in determining who ends up leading companies, and whether they are fired or end up industry leaders.
There’s perhaps no better proof of this point than the third factor the researchers looked at in the study of male CEOs in Sweden. In addition to the interview and the cognitive tests, the researchers considered height. Taller men, they found, were more likely to become CEO, though the effect wasn’t as strong as for the other two factors. Even though CEOs weren’t at the very top of the spectrum in intelligence, when the researchers combined cognitive ability, interview results, and height into one score, they found that CEOs of large Swedish firms were in fact in the top 5%. Becoming CEO doesn’t necessarily hinge on being the best and brightest, though that helps. It’s also important that society thinks you look the part.
Walter Frick is a senior associate editor at Harvard Business Review.
No comments:
Post a Comment