Strong managers, weak boards?

Many governance reform proposals are based on the view that boards have been too friendly to executives, for example, by awarding them excessive pay. Although boards are often on friendly terms with executives, it is less clear that they have systematically failed to function in the interests of shareholders. Understanding board monitoring requires a theory of boards that takes into account how firms provide incentives for their Chief Executive Officer's (CEOs) through other means. We develop a model in which a CEO's ownership stake and private benefits have opposite effects on his willingness to share private information with an independent board of directors. To encourage the CEO to communicate, the board may optimally commit to a low monitoring intensity when either CEO ownership is low or private benefits are high. Our model suggests that the existing cross-section evidence on the correlation between board composition and CEO ownership and tenure needs re-evaluation. Using a new proxy for board monitoring, we provide new evidence that this cross-sectional correlation appears to be non-monotonic, with board independence first decreasing and then increasing in CEO ownership and tenure. We discuss the implications of our model for the design and evaluation of governance structures.