A Note on the Dynamics of Incentive Contracts

Laffont and Tirole (1988) show that when the uncertainty about the agent's ability is small, the equilibrium must involve a large amount of pooling, but whether the continuation equilibrium induced by a optimal first-period menu of contracts is partition or not remains unclear. They construct a nonpartition continuation equilibrium for a given first-period menu of contracts and conjecture that this continuation equilibrium need not be suboptimal for the whole game under small uncertainty. We show that, irrespective of the amount of uncertainty, this nonpartition continuation equilibrium generates a strictly smaller payoff for the principal than a different menu of contracts with a partition continuation equilibrium. In this sense, Laffont and Tirole's menu of contracts, giving rise to a nonpartition continuation equilibrium, is not optimal. An intuition behind this result is provided that may shed some light on the problem of dynamic contracting without commitment.