Can Agents Be Better Off with Pay Caps?
基于标准委托代理模型的扩展，研究发现薪酬上限可能适得其反，因为代理人在有薪酬上限时可能过得更好；给定固定努力水平，代理人期望效用可能随奖励上限提高而下降。
Abstract A popular way to discipline the managers of companies or banks that got into trouble during the recent financial crisis has been to impose caps on managers' pay. Using a small extension of the standard principal–agent model, we argue that pay caps might serve the opposite purpose, because the agent might be better off with a pay cap. Specifically, we show that, given a fixed effort level to be implemented, the agent's expected utility can be decreasing in an upper bound for the agent's reward. The effect of pay caps on the general structure of optimal incentive contracts is also characterized. While an improvement of contracting information always helps the principal, it might increase or decrease the marginal cost of imposing pay caps.