Pareto-improving asymmetric information in a dynamic insurance market

This paper explored the dynamics of insurance markets under incomplete information. Various information structures are examined, according to the degree of communication between companies. We get equilibrium existence even when adverse selection arises through differentiated learning. This and the Pareto-dominance of private information structures seem to mitigate the prevalent view that adverse selection and competition do not match well in insurance markets; moreover, it provides a new scope for empirical studies. Technically, we extend to dynamics Rothschild-Stiglitz' equilibrium concept, and get to reconsider the "no-malus" property, which we prove to result from the non-consideration of feed-back effects of future on present.