Learning in Relational Contracts

We study relational contracts between a firm and a worker with mutual uncertainty about match quality. The worker’s actions are publicly observed and generate both output and information about the match quality. We show that the relational contracts may be inefficient. We characterize the inefficiency through a holdup problem on the contemporaneous output. In the frequent action limit, these inefficiencies persist if and only if information degrades at least at the same rate at which impatience vanishes. We characterize optimal relational contracts and show that they involve actions that yield both a lower payoff and less information than another action.