Monitoring Managers: Does It Matter?

In the paper, Monitoring Managers: Does It Matter? which was recently made publicly available on SSRN, we investigate how boards of directors monitor management, under what circumstances they fire CEOs, and whether these actions improve performance. Boards of directors are tasked with ensuring that firms are run by competent managers who act in their shareholders’ interest by providing appropriate incentives and through “active monitoring,” that is, collecting information about the firm’s operations or the manager’s ability and firing the manager if necessary. Much of the economic literature on corporate governance and boards studies the provision of incentives and pays less attention to monitoring. In this paper, we ask if boards with large shareholders indeed engage in active monitoring and whether such monitoring in turn improves performance.

Our tests are based on detailed and unique data for an 18-year panel of 473 private-sector firms in 19 transition economies in Central and Eastern Europe and Central Asia that were financed by 43 private equity funds following the collapse of the Soviet Union. These rich data allow us to document what types of information boards collect and how they weight each type when taking a key decision: whether to fire the CEO.

We show that boards engage in monitoring in the sense of collecting ‘hard’ (i.e., verifiable) and ‘soft’ (i.e., nonverifiable) information about the firm’s operations and the CEO’s ability. We also show that boards act on their information: They fire the CEO in response to negative hard information (i.e., poor performance relative to agreed targets) and when their soft information suggests that he is incompetent.

Learning about the CEO’s ability requires that the board can filter out noise when evaluating the CEO. We show that monitoring enables the board to distinguish between bad luck or honest mistakes on the one hand and behavior or decisions that would raise concerns about the CEO’s ability and so about the company’s future performance on the other. Specifically, we find that boards in our data fire the CEO only in response to the latter concerns rather than for poor performance that was the result of bad luck or of a decision that was wrong ex post but reasonable ex ante.

Our setting exploits an exogenous legal change in a board’s ability to act on the information it collects. Not surprisingly, we find that boards that acquire the legal power to fire a CEO without needing approval from shareholders at large are more likely to do so. More interestingly, we also find that the sensitivity of CEO turnover to soft information doubles when boards are allowed to act on their information, while the sensitivity to hard information remains unchanged. This is an important result because it suggests that active monitors can take actions based on soft information that may be difficult to convey convincingly to shareholders who are not represented on the board. Thus, if the board lacks the power to act on what it observes, it is possible that incompetent CEOs are allowed to stay in post for too long.

Finally, we use the law changes to investigate if CEO turnover improves performance. While it would be surprising if it didn’t, there is scant prior evidence of a causal effect. Using the law changes allows us to solve this problem. We find that firing incompetent CEOs leads to large improvements in performance. This last piece of evidence closes the loop. We show that boards engage in monitoring, act on the information collected, and that such actions improve performance.

The full paper is available for download here.