CEO succession at many companies occurs in a black box. Shareholders are not privy to boardroom discussions prior to the announcement of a CEO departure, and press releases announcing the change contain boilerplate language that does not make it clear whether the CEO stepped down voluntary or was forced to resign. In this Closer Look, we examine a model — called the Push-out Score that combines observable factors with human evaluation of circumstantial evidence to estimate the likelihood that a CEO was asked to leave. We ask: How accurate is this model and does it provide better insight into whether a CEO was fired? Can this model be used to evaluate the relation between CEO turnover and performance? Can it be used to assess governance quality? Should companies that receive a moderate score take more effort to clarify for investors whether the CEO left voluntarily?
Copyright held by Ian D. Gow, David F. Larcker, and Brian Tayan. Further inquiries about reproduction and use should be directed to the Corporate Governance Research Initiative.
