This paper explores one reason why a corporate board often fails to replace a substandard CEO. I consider the situation in which the incumbent CEO and direscotrs make decisions in the absence of the new CEO. I show that in this situation, which is common in practice, the board and the CEO end up maximizing the expected utilities of the negotiating parties that do not include the expected utility of the potential CEO. This sometimes results in the retention of an inefficient CEO. Moreover, I argue that this same logic provides a theoretical explanation for how a new CEO is chosen in relation to both the voluntary and enforced replacement of an existing CEO. Specifically, the equilibrium succession policy may depart from the optimum succession policy; that is, the optimum from the shareholders' perspective.