Boards of Directors as an Endogenously Determined Institution: A Survey of the Economic Literature
Hermalin, Benjamin E., Weisbach, Michael S., Federal Reserve Bank of New York Economic Policy Review
Most organizations are governed by a board of directors. In fact, having a board is one of the legal requirements for incorporation. Many nonincorporated entities also have a governing board of some sort, such as a state university's board of regents. Given the myriad boards in place today, it is reasonable to ask, Why do they exist? What do they do? Can they be "improved"? These questions are at the heart of governance and, to a certain extent, management. As such, they have motivated much of the research on this topic.
This paper surveys the research on boards of directors in the economics and finance literature. Boards of directors are an economic institution that, in theory, helps to solve the agency problems inherent in managing an organization. Although boards satisfy numerous regulatory requirements, their economic function is determined by the organizational problems they help to address. Yet formal economic theory on boards has been quite limited. For example, the characteristics of agency problems that could lead to boards being the equilibrium solution have not yet been specified. Similarly, the conditions under which regulation of boards will lead to improvements are unknown.
Despite the absence of formal theory, we have a strong intuitive sense of the problems facing boards. A major conflict within the boardroom is between the CEO and the directors. The CEO has incentives to "capture" the board, so as to ensure that he can keep his job and increase the other benefits he derives from being CEO. Directors have incentives to maintain their independence, to monitor the CEO, and to replace the CEO if his performance is poor.
To some extent, the vacuum in formal theory has been filled by empirical work on boards. The "cost" associated with this approach, however, is that little of the empirical work on boards has been motivated by formal theory. Rather, it has sought to answer one of three questions:
1. How do board characteristics such as composition or size affect profitability?
2. How do board characteristics affect the observable actions of the board?
3. What factors affect the makeup of boards and how do they evolve over time?
A key issue in this empirical work is how to proxy for the board's degree of independence from the CEO. Much of this work starts from the sometimes implicit assumption that observable board characteristics such as size or composition are related to the level of board independence.1
Research thus far has established a number of empirical regularities. First, board composition, as measured by the insider-outsider ratio,2 is not correlated with firm performance.3 However, the number of directors on a firm's board is negatively related to the firm's financial performance. Second, board actions do appear to be related to board characteristics. Firms with higher proportions of outside directors and smaller boards tend to make arguably better-or at least different-decisions concerning acquisitions, poison pills, executive compensation, and CEO replacement, ceteris paribus. Finally, boards appear to evolve over time depending on the bargaining position of the CEO relative to that of the existing directors. Firm performance, CEO turnover, and changes in ownership structure appear to be important factors affecting changes to boards.
Two important issues complicate empirical work on boards of directors, as well as most other empirical work on governance. First, almost all the variables of interest are endogenous. The usual problems of joint endogeneity therefore plague these studies. For instance, firm performance is both a result of the actions of previous directors and itself a factor that potentially influences the choice of subsequent directors. Studies of boards often neglect this issue and thus obtain results that are hard to interpret.
Second, many empirical results on governance can be interpreted as either equilibrium or out-of-equilibrium phenomena. …