A Test of the Incremental Power of CEO Productivity in Optimum Incentive Modelling

Employing a unique dataset with varying grant sizes and exercise prices, we test the competing optimal option incentive models of Hall and Murphy (2000, 2002) (HM) and Baker and Hall (2004) (BH) which differ with respect to the impact of CEO productivity on incentive determination and also the treatment of risk aversion and firm risk. Direct testing of the BH model proves more satisfactory than direct testing of the HM model, but given this outcome, it is surprising that the HM model explains a higher proportion of grant abnormal returns than does the BH model. We attribute this outcome to PPS and not the exercise price. Since exercise prices have been shown unrelated to agency problems, we conclude that caution needs to be exercised when prescribing exercise prices to create the optimal incentive, especially for highly risk-averse CEOs and those who are also poorly diversified.