Splitting Up the Roles of CEO and Chairman: Reform or Red Herring? (page 1 of 3)

But Wharton faculty members say there is no evidence that separating these positions, as a general philosophical rule, improves corporate performance.

Many advocates of improved corporate governance assert that the posts of CEO and chairperson should not be held by the same individual because such an arrangement concentrates too much power in the hands of a single executive, leading to potential CEO entrenchment and poor bottom-line performance. The many corporate scandals that have unfolded since 2001 have added impetus to this view, even though there were separate CEOs and chairpersons at WorldCom and Enron when wrongdoing engulfed those companies.

Figures show that a growing number of companies are dividing the jobs of CEO and chairperson, but the trend is not widespread. In March, the Corporate Library, a good-governance advocate, published the results of a survey showing that the same person holds both posts at most American companies -- 377 of the corporations in the Standard & Poor's 500 index. That is 17 fewer companies than the year before.

"In terms of splitting the roles, there is no academic evidence to suggest that it's a good thing," says finance professor Andrew Metrick. "What we do have evidence for is that it's important for the CEO to realize that he is reporting to the board and not controlling the board completely." Many researchers have looked at this issue, Metrick adds, but it has been impossible to reach a firm conclusion that financial performance is improved by splitting the roles. "Most of the dynamics that we'd really like to capture are just not that easy to see in the data. People tend to focus on board independence, but board independence is a very squishy concept."

Metrick says that the emphasis on having separate CEOs and chairpersons is largely a "red herring" because board independence can be accomplished in other ways, such as directors holding meetings without the CEO being present.

"Everyone's struggling to come up with metrics for corporate governance," says accounting professor David F. Larcker. "There are a lot of governance scorecards [compiled by shareholder advocacy groups], but governance is inherently very difficult to measure. One measure that's gotten lots of publicity is splitting the chair and CEO roles. My feeling is, for most cases, it's nice window dressing but not substantively a big deal."


Management professor Michael Useem, director of Wharton's Center for Leadership and Change Management, says a few statistical studies have compared companies where two persons hold the CEO and chair positions with companies where one person holds both posts. This research, which also took into account other factors that can affect financial performance, shows that whether a company does or does not separate the CEO and chairperson titles "has no bearing on corporate financial performance," he notes.

Robert E. Mittelstaedt Jr., vice dean and director of Wharton Executive Education, says experience shows that it is difficult to make the argument that dividing the two roles will result in better performance in all cases. "There's no one black-and-white answer to fit all situations," he explains. "Many people would believe that, if nothing else, you diffuse power to some extent and decrease the probability of a scandal by separating the chairperson and CEO positions. But that doesn't mean you will have great business results." Indeed, by dividing those roles, a company may weaken its ability to develop and implement strategy.