Does a powerful CEO help or hurt a company? It depends on whether anyone's watching the wolf.
When it comes to corporate governance, what's best for a company may also be what's most difficult to put into practice, according to a new study examining the impact of "dominant" chief executives on corporate strategy and financial performance.
The study, titled "Dominant CEO, Deviant Strategy, and Extreme Performance: The Moderating Role of a Powerful Board," was conducted by three Canadian business professors: Jianyun Tang of Memorial University, and Mary Crossan and W. Glenn Rowe, both with the University of Western Ontario's Richard Ivey School of Business. The study was recently published online by the Journal of Management Studies. It is expected to appear in the print edition of the journal later this year.
According to the authors, the study attempts to shed light on the tendency by the media and others to portray dominant CEOs as either heroes or corporate destroyers. The study defines dominant CEOs as those who are more likely to make unilateral decisions in the face of opposition from other top executives, while less dominant CEOs are those more likely to compromise.
The study found that firms with dominant CEOs tend to have a performance that is different from that of the industry norm, resulting in either great success or spectacular failure. "What we found is it's not really that they are bad or good but that they tend to have extreme performance," explains Dr. Tang, professor of strategic management at Memorial.
It cites Kenneth Lay, head of disgraced energy-giant Enron Corp., as an example of a dominant CEO who achieved spectacular failure, and Bill Gates, head of Microsoft Corp., as an example of someone whose company has achieved great success.
The study also found that powerful boards of directors can reduce the likelihood of a CEO harming a firm because they are more likely to screen out and oppose unsound business decisions. "In other words, having dominant CEOs is risky, but powerful boards help control the downside risk while leaving the upside potential relatively open," the authors write.
However, the study yielded another surprising result. It found that firms that achieved the best financial performance were those with less dominant CEOs coupled with less powerful boards. In this situation, the CEO's power was likely to be balanced by other senior executives, rather than the board.
What the study seems to suggest is that when it comes to corporate governance, there are two favoured options: a dominant CEO coupled with a powerful board, and a less dominant CEO coupled with a less dominant board. "Power has to be balanced one way or another," either between the CEO and the board, or the CEO and senior executives, Dr. Tang says. The study also highlights the need for boards to be aware of the downside risks of having a dominant CEO and provide more stringent oversight when the situation occurs, he says.
But, he adds, in practice, companies seldom achieve this balance. "That's not actually that surprising because a lot of CEOs have control over their boards," he says. "That's the difficulty with corporate governance. It's not easy to ask someone to set up a mechanism to oversee themselves."
Things have improved somewhat in recent years, he adds. In the 1990s, CEOs used to have more say over board appointments. Now many companies have independent nomination committees. "But I think in most cases CEOs still have a lot of influence," he says.
The researchers based their findings on the performance of a sample of 51 publicly-traded U.S. computer companies from 1997 to 2003. The authors selected the computer sector because it is a fast changing industry in which top executives tend to have a lot of discretion and their impact was more likely to be detected.
Special to The Globe and Mail