Welfare‐Improving Asymmetric Information in Dynamic Insurance Markets
构建了一个两期非对称学习模型，发现保险公司间不共享事故信息导致的信息不对称反而能提高福利，这与传统逆向选择文献的结论相反。
This article presents a two‐period asymmetric learning model of insurance markets. When information about past accidents is not shared by insurers, asymmetries of information develop through time. Equilibrium contracts exist, are payoff unique, and display a realistic bonus‐malus pattern. Eliminating asymmetries through information sharing is welfare decreasing, in contrast with past contributions on insurance and adverse selection. When second‐period contract offers cannot be contingent on initial contract choice, a strict increase in welfare is obtained through menus of contracts, although initial contract choice is in itself worthless information.