Optimal Regulation without Commitment: Reputation and Incentives

We often observe regulators deviating from stated rules because they cannot resist the ex-post pressure to re-optimize. Examples include the regulation of financial firms (bailouts), fiscal rules, and currency boards. This paper studies the optimal design of regulation in a dynamic model when there is a time inconsistency problem and uncertainty about whether a regulator can commit to follow the rule ex-post. A regulator can either be a commitment type who can always commit to enforce regulation or an optimizing type who sequentially decides whether to enforce or not. This type is unobservable to private agents who learn who about it through the actions of the regulator. Higher beliefs that the regulator is the commitment type (the regulator’s reputation) helps promote good behavior by private agents. Consequently, we show that in a large class of economies, uncertainty about the type of the regulator helps provide incentives to private agents and thus allows for the implementation of good outcomes. Therefore, learning the type of the regulator can be costly. If the initial reputation is not too high, the optimal regulation is the strictest regulation that is incentive compatible for the optimizing type. We show that reputational considerations imply that the optimal regulation is more lenient than the one that would arise in a static environment. We study two applications of this framework. First, we study the design of the optimal inflation target in a dynamic Barro-Gordon model. Second, we study the design of optimal bailout policies when there is an underlying moral hazard problem and social costs associated with default. Contrary to conventional wisdom, we show that bailouts along the equilibrium path are necessary to discipline future risk-taking of financial firms.