A common dilemma for people who seek advice is that good advice sometimes comes at the cost of revealing negative information about the persons seeking it. In the world of corporate decision making, a CEO who seeks the counsel of the board of directors due to a problem with, for example, a project, is also implicitly conveying that the problem arose under her stewardship. The CEO thus faces the following predicament: She can accurately communicate the problem to the board and therefore get the board’s expert advice on how to proceed, or the CEO can mislead the board by, say, minimizing the problem, prompting less useful advice. The former approach is best for shareholders and the firm. However, the latter strategy is, in a narrow sense, best for the CEO; by being less than forthcoming, the CEO avoids disappointing the board (for the time being) and can hope to preserve her reputation.

The question we consider in our recent study is, how can shareholders achieve honest communication between the board of directors and a CEO who may prefer to distort the true situation? To shed light on this question, we develop a parsimonious economic model between a CEO and a board of directors. The CEO has private information about the status of a project within the firm, such as preliminary progress regarding new product development. The CEO can choose to reveal this information to the board and, in turn, receive the board’s expect advice regarding the best path forward. The key novelty of our setting is that the status of the project also conveys information about the CEO’s ability. The CEO has career concerns; she prefers to keep her job, all else equal, while she also prefers to maximize shareholder value. Consequently, following a bad project outcome, the manager may prefer to sugarcoat the situation and paint a rosier picture. However, that makes the board’s advice  much less useful for the CEO and the firm and can therefore harm shareholder value.

How do shareholders ensure honest communication from the CEO? The main result of our study is that shareholders can adopt a policy of aggressive boards, meaning electing independent board members, who are perhaps less forgiving of a bad CEO.  This result on its face may appear counterintuitive, as the CEO is more open to a board that is more inclined to terminate her employment. The reasoning for this result is as follows. First, an unfriendly board uses a very high standard for retention, requiring CEOs to show strong performance. As a result, CEOs who are a poor fit for the firm realize that they have very little of chance of staying in power. In turn, these “bad” CEOs have less to lose from being honest about intermediate progress, even if the information is negative, and therefore receive the most useful advice from the board. In other words, poor-fit CEOs focus on maximizing shareholder value while they are in power, and less so on preserving their position within the firm. Since the likelihood of staying in power is low due to the high bar for retention, these CEOs are more willing to divulge negative information about projects that could reflect badly on themselves as well. As a result, shareholders benefit in two ways—through value-increasing communication and through the removal of the less-than-ideal CEO.

However, the more aggressive board also has its drawbacks. In particular, the high retention standard that the unfriendly board uses also causes some good CEOs to be fired by mistake. For example, in the event that the CEO’s information is positive, the board continues to remove this CEO if performance measures are not met. Such a draconian policy is necessary, as the board must be harsh in order to induce honest communication in the first place. Despite this cost of errant replacement, we find in the solution (the equilibrium) that shareholders can maximize value through an aggressive board policy, as an aggressive board is both more efficient in removing bad CEOs and inducing honest communication than is a board that is more friendly to the CEO.

Finally, our results offer predictions for empirical analysis and help to explain existing puzzles. For example, previous studies document that CEOs are often replaced in cases where performance measures are met or even exceeded. Our results provide a rational explanation for this, as board members rely on soft information, such as in their interaction with advising the CEO in our model. Indeed, when the CEO reports negative information, the board is far more inclined towards replacement, even if performance numbers turn out to be unexpectedly strong.

This post comes to us from professors Cyrus Aghamolla at the University of Minnesota—Twin Cities and Tadashi Hashimoto at Yeshiva University. It is based on their recent article, “Aggressive Boards and CEO Turnover,” available here.