Can Shareholders Be Wrong?

Two views of board/CEO relationship persist. One is the common view that boards that are “entrenched” — or insulated from shareholder action — will ignore a struggling CEO to the detriment of shareholder interests. Such lax oversight is deemed to have played a role in allowing the recent spate of corporate scandals and record-setting bankruptcies to take place. The opposing view holds that boards that respond quickly to public sentiment could become agents of shareholders who are less informed. “[That view] is very much built on the model of a short-sighted shareholder agitator,” says Ray Fisman, associate professor at the Columbia University Business School and coauthor of the working paper Governance and CEO Turnover: Do Something or Do the Right Thing. However, Fisman says, caving in to vocal shareholders’ demands might not be in the best interest of all shareholders in the long run because shareholders may be too quick to misattribute a short-term stumble to a failure in the executive suite, when there may be extenuating circumstances or causes.

So which view captures what’s really happening in a typical boardroom? Fisman and his coauthors Matthew Rhodes-Kropf, an associate professor at Columbia Business School, and Rakesh Khurana, an associate professor at the Harvard Business School, examined 436 of the largest U.S. companies for which complete data could be assembled. They analyzed company data for the period from 1980 through 1996 and identified 728 CEO successions. On the basis of a literature search, they determined that CEOs were forced out in 180 of those successions.

For each company, the researchers measured the degree of board entrenchment by checking data on corporate bylaws from the Washington, D.C.–based Investor Responsibility Research Center. They looked for statutes that are generally considered to be unfriendly to shareholder rights, such as staggered boards or poison pills, which corporate governance advocates deem to be obstructive because they protect directors from shareholder action.

The data showed differences between companies with entrenched boards and others. For example, entrenched boards are less likely to fire a CEO, and when they do, their companies tend to experience greater improvement in operating income after the firing. However, although these findings suggest that entrenchment matters, they don’t shed light on whether cozy boards tend to be indolent or whether vocal shareholders tend to be shortsighted — because both views predict the same results. After all, a board cozy with management would tend to keep its CEO unless the potential gains for firing were enormous, so performance after such a firing would be expected to pick up.

The most intriguing finding emerges when the authors looked at a nonevent: the retention of a CEO when corporate performance lags. Here, the two views diverge in their predictions: On one hand, if entrenched boards are beholden to CEOs and will replace them only under extreme circumstances, then the companies are likely to continue to founder. On the other hand, if entrenched boards know something about the CEO, the company and its long-term prospects that vocal shareholders don’t, they might retain the CEO only if they deem the CEO to be capable, and performance would be expected to turn around.

That is indeed what the research revealed. When a CEO was retained even though performance declined beyond a threshold statistically deemed to signify a high propensity to fire, the companies with entrenched boards outperformed others in the subsequent two years.

Although this supports what the paper describes as the “misguided shareholder view of entrenchment,” the authors point out that not all vocal shareholders are shortsighted and not all entrenched boards act to the shareholders’ long-term benefit. But findings do highlight the fact that entrenchment itself might not always be a bad thing. After all, the board’s role is more than just oversight, Fisman points out, and entrenchment’s effect on the board/CEO relationship is not entirely clear.

What does this tell the burgeoning group of corporate governance ratings companies? “You can only concentrate on things you can observe and measure,” says Fisman. “If we had some harder-edged measure of the soft notion of board/management relations, it would be of tremendous utility.” In the meantime, perhaps shareholders shouldn’t be so quick to conclude that a board that isn’t taking public action is in the CEO’s back pocket.

For more information, contact the authors at rf250@columbia.edu, rkhurana@hbs.edu or mr554@columbia.edu.