中文摘要：

本文研究经理人隐藏行动、努力成本和风险厌恶态度对经理人激励的影响。本文的研究基于标准委托人-代理人模型，但放松其关于经理人行动不能影响企业风险的假定，而采用更接近现实的假定——企业风险至少部分内生于经理人的行动。通过研究，本文将经理人激励明确细分类为努力增进激励（βeffort激励）和追求风险激励（βrisk激励）两类。为实现βeffort激励，经理人报酬只需是企业业绩的线性函数即可；但为实现βrisk激励，经理人报酬应是企业业绩的凸函数。

英文摘要：

This paper will provide some explanations for the existing mixed empirical evidence regarding the relationship between executive incentives and firm risk from the viewpoint of classifying executive incentives. The study is based on the standard principalagent model with a more actual assumption that executives are able to influence or control firm idiosyncratic risk, which relaxes and substitutes the assumption that firm risk is totally independent of executive actions. Through examining the effects of hidden actions, effort cost and risk aversion on executive incentives, the authors of this article deafly classify executive incentives into such two types as Effort-enhancing Incentive （βeffort Incentive） and Risk-seeking Incentive （βrisk Incentive）. In order to provide the βeffort Incentive, it is enough that executive compensation is the linear function of firm performance, which will show the positive relationship between executive compensation and firm performance, and the negative relationship between executive incentives and firm risk. However, to provide the βrisk incentive, executive compensation should be the convex function of firm performance, which will show the positive relationship between executive compensation and firm performance, and the positive relationship between executive incentives and firm risk. In detail, the βeffort incentive is negatively related to both firm systematic risk and firm idiosyncratic risk, but the βrisk incentive is positively related to firm idiosyncratic risk if the executive incentive＇s wealth effect is bigger than its risk-aversion effect. The authors believe that their classifying of executive incentives can help to exploit the general equilibrium relationship between executive incentives and firm risk, and future researches based on the above incentive classification will be important improvements in the field.