基于公共偏好理论的激励机制研究——兼顾横向公平偏好和纵向公平偏好

Incentive Mechanisms Based on Fairness Preference Theory——Give Consideration to Both Vertical and Horizontal Fairness Concerns

基于公平偏好理论,本文改进了"经济自利人"的假设,并拓展了委托代理的分析框架,以此对传统HM模型有所发展。本文同时考虑横向和纵向的公平偏好,完善了单边维度公平偏好下的委托代理分析。研究表明,伴随着代理人公平偏好程度的增大,代理人努力水平的提升和代理成本的降低揭示了公平偏好理论对于优化委托代理合作的重要作用,这对现实的委托代理合作有很强的指导意义。

The principal agent theory focuses on designing incentive mechanisms to reduce moral risk and losses for both principal and agent under the condition of asymmetric information.The traditional principal agent theory is based on the ＂ self-interest＂ hypothesis and has many limitations.The psychology game experiment of experimental economics shows that humans have limited self-interest and often consider interests of others.The growing theory of fairness concerns indicates that fairness is an important factor to influence people＇s thinking and behavior.This paper incorporates the fairness concerns theory into the framework of principal-agent and further develops the traditional principal agent theory.This paper addresses vertical fairness and horizontal fairness concerns,improve the hypothesis of ＂ economic self-interest person＂,and modify agents＇income function.In addition,this paper expands the analysis framework of principal-agent from single to double agents in order to analyze the principal-agent incentive mechanism.In the condition of symmetric information,the balanced solution of the fairness-concerns double agent model is only influenced by the degree of vertical fairness concerns and coefficient of effort＇s cost,rather than by risk avoidance and wage contract.In the case of asymmetric information,different optimal distribution coefficients of output can be obtained by considering different dimensions of fairness concern.In another word,we could not obtain complete conclusion by only considering single dimension fairness concerns.In this paper,we incorporate vertical fairness concerns and horizontal fairness concerns into the model.We equalize the degree of horizontal and vertical fairness concerns for data analysis.Under the condition of asymmetry information,the study finds that the optimal output sharing coefficient of the fairness-concerns double agent model is uni-directionally changing with the degree of fairness-concerns.When the fairness concerns tend to become infinite,the optimal output sharing coefficient would be the same as the results of symmetric information condition.The direction of unidirectional is determined by the traditional factor,not the degree of fairness concerns.The level of agents＇ efforts is changing the same direction with the degree of fairness concerns,which means we could improve the level of agents＇efforts by raising the degree of fairness concerns.Obviously,this is another way to motivate agents besides dividing profit between principal and agent.Agency cost has positive correlation with output variance and risk avoidance,but negative correlation with the degree of fairness concerns.This means that the agents＇fairness concerns could significantly help reduce agency costs.With fairness concerns,the agency cost is less than traditional HM model.When the degree of fairness concerns tends to become infinite,the agency cost is close to 0.The view of fairness concerns can help enhance cooperation between principal and agents.By applying vertical fairness concerns and horizontal fairness concerns into the fairness-concerns double agent model,this paper improves the traditional HM model,and shows fairness concerns is an influential factor for cooperation level.Due to asymmetric information,the principal-agent collaboration could not reach the optimal level in reality.This paper offers solutions to this problem by raising the agents＇degree of fairness concerns in order to reduce behavior deviation.The process of designing incentive mechanism must be in the frame of vertical fairness concerns and horizontal fairness concerns.