Why do firms use incentives that have no incentive effects?

This paper illustrates why firms might choose to implement stock option plans or other pay instruments that reward "luck." I consider a model where adjusting compensation contracts is costly and where employees' outside opportunities are correlated with their firms' performance. The model may help to explain the use and recent rise of broad-based stock option plans, as well as other financial instruments, even when these pay plans have no effect on employees' on-the-job behavior. The model suggests that agency theory's often-overlooked participation constraint may be an important determinant of some common compensation schemes, particularly for employees below the highest executive ranks.