The Role of Commitment in Dynamic Contracts: Evidence from Life Insurance †

We use data from the life insurance industry to study the properties of long term contracts in a world where buyers cannot commit to a contract. We develop a theoretical model that captures the main features of this industry, in particular, how contracts are designed to deal with reclassification risk. The data is especially suited for a test of the theory since it includes information on the entire profile of future premiums. The lack of commitment by consumers shapes contracts in the way predicted by the theory: all types of contracts involve frontloading; this generates a partial lock-in of consumers; contracts that are more front-loaded have a lower present value of premiums over the period of coverage. This is consistent with the idea that more front-loaded contracts retain better risk pools. †We are very grateful to Glenn Daily. We also thank John Cochrane and two anonymous referees for helpful comments, Andrew Funderburk for research assistance, Derek Neal, and Steve Tadelis for detailed comments, and LIMRA Int. and Compulife for generously providing us access to their data. A previous version of the paper circulated under the title “Optimal Dynamic Contracts: Evidence from Life Insurance.” The financial support of the NSF through grants SES 9986287 and SES 9911496 is gratefully acknowledged. ∗University of Wisconsin and NBER. e.mail: [email protected] §New York University. e.mail: [email protected]