This paper evaluates the empirical relationship between top executive turnover and firm performance. Based on a sample of the 460 largest UK listed companies during the period 1990-1998 we establish an inverse and robust statistical relation between the probability of a management change and a firm's performance: top executives are fired for poor performance. This can result from internal monitoring of management by the board or block share holders. Second, the data indicate that only very poor levels of performance affect significantly the turnover likelihood: corporate performance must fall dramatically to force a senior executive job separation. Third, the likelihood of managerial turnover for poor performance has not changed over time: today's senior managers face the same disciplining effects as those senior managers in earlier years. Finally, there seems to be no evidence that managerial stock ownership, measured as the proportion of ordinary shares owned by top managers, enables them to become entrenched.