车险市场分离道德风险和逆向选择的实证分析

Separating Moral Hazard from Adverse Selection:An Empirical Analysis

本文的目的是考察国内车险市场中是否存在信息不对称现象，并试图分离道德风险和逆向选择的影响，即分离出二者对信息不对称的贡献。利用国内一家大型财险公司某省份2008～2009年商业第三者责任险数据，论文通过Probit模型验证了信息不对称的存在；同时，论文利用分组控制住了逆向选择（风险异质性）的影响，并利用对照组剔除了学习效应的贡献，证明商业第三者责任险市场中不存在道德风险；因此该市场的信息不对称完全是由逆向选择贡献的。

This article investigated whether there was information asymmetry in China＇ s auto insurance market, and tried to separate the impacts of moral hazard from those of adverse selection on information asymmetry. Using panel data from the 2008 -2009 commercial third party liability insurance data of a provincial branch of a large do- mestic property ＆ casualty insurance company, the article confirmed the existence of information asymmetry through the Probit model. By controlling the impacts of adverse selection （risk heterogeneity） and eliminating the contribu- tion of learning effects based on appropriate grouping, the article found no evidence of moral hazard in the commer- cial third party liability insurance market. Therefore, it concluded that information asymmetry was completely gener- ated by adverse selection.