BETA
This is a BETA experience. You may opt-out by clicking here

More From Forbes

Edit Story

CEO Pride Goeth Before A Fall, Study Finds

This article is more than 8 years old.

If former Lehman Brothers CEO Dick Fuld had taken a test that business school professors use to measure executive confidence, called Core Self-Evaluation, or CSE, he would likely have gotten a very high score, says Ciaran Heavey, a professor at the University of Dublin. Heavey is the co-author of a new paper entitled “Exuberance in the Corner Office: CEO Core Self-Evaluation and the Rise and Fall of Irish Firms, 2005-2009,” which will be presented at the annual meeting of the Academy of Management in early August.

According to the paper, which examines the CEO confidence and performance of 156 privately-held Irish companies in 2006 and 2007, when the economy was booming, and in 2008 and 2009, as it was crashing, firms where CEOs had high CSEs performed much worse during the bust than did firms with lower-CSE chief executives. “A mixture of high CSE and aggressive expansion constitutes a potentially dangerous combination,” says the paper. “We surmise that high-CSE CEOs engage in a flurry of initiatives during boom times, many of them speculative and poorly vetted. Once the environment collapses, many of these poorly-reinforced initiatives tumble—like a house of cards.”

Fuld’s performance, says Heavey, offers a “classic case” of a high-CSE executive over-extending his 158-year-old firm’s reach and then failing to rescue the company from disaster. Lehman Brothers had a 40-to-1 debt-to-equity ratio because of its huge positions in the subprime mortgage market. After the economy started contracting in 2008, the firm declared bankruptcy, throwing 26,000 people out of work, costing investors millions and, many believe, worsening the economy’s downward spiral. “That’s the risk of having a highly confident CEO,” says Heavey. “Because they have such a strong belief in what they’re doing, they underestimate their company’s ability to handle a situation like the one Lehman faced.”

Heavey and his team had a unique opportunity to study the effects of CEO confidence on company performance in a recession, because in 2005 he had gathered data on Irish CEOs for a previous study on whether executives with high confidence levels pushed their firms to be more entrepreneurial (they did). Having gathered that initial data, he was able to go back and look at those firms’ performance when the economy tanked.

Heavey and his colleagues were surprised by their findings. “If you look at all the academic literature, it talks about executive confidence in a very positive sense,” he says. That implies that confident CEOs will do a good job leading their firms during challenging times. “But what we saw instead was that the companies led by CEOs with higher levels of confidence actually did the worst,” he says. “That’s contrary to the common belief that confidence is universally good.”

Specifically, companies with CEOs who scored in the highest third in self-confidence saw their returns on assets fall from an average of 12% in 2006 to near zero in 2009. In firms with moderately confident chiefs, returns also fell, but not nearly as far, from 10% to 4%. Firms with low-confidence CEOs also suffered, but not as much as high-confidence CEOs. Their returns dropped from 6% during good times to 2% in the bust.

To measure CSE, the researchers asked the CEOs to rate a dozen statements on a scale of 1 (very strongly agree) to 7 (very strong disagree). The statements included “I am confident I get the success I deserve in life,” “I rarely have doubts about my competence,” and “I rarely feel pessimistic.” The mean score for the group was well-above people in the general population. Those in the upper third had extremely high scores.

“We expected that highly confident CEOs would have been able to take control of events and be more resilient when there were downturns in the economy,” says Heavey. “But what we saw instead was that the companies led by CEOs with higher levels of confidence actually did the worst.”

Heavey's co-authors on the study are Brian Fox and Zeki Simsek from the University of Connecticut and Donald Hambrick from Pennsylvania State University.

Also on Forbes: