Implementing the Commitment Solution via Discretionary Policy-Making
33 Pages Posted: 31 Mar 2020 Last revised: 23 Jan 2021
Date Written: January 23, 2021
This paper demonstrates that, in a large class of linear-quadratic models with rational expectations, losses due to time-inconsistency problems can be avoided, as the commitment solution can be implemented by a policy-maker who acts under discretion. We focus on two approaches. First, we show that a reputational equilibrium that implements the commitment solution always exists. In this non-Markovian equilibrium, it is optimal for the discretionary policy-maker to react to additional state variables that do not enter the social loss function directly. Second, we show how delegation to a policy-maker with an additional objective for the policy instrument can be used to implement the commitment solution via a standard discretionary Markov equilibrium. This approach can be implemented in practice by using incentive contracts or the announcement of a specific strategy.
Keywords: Discretion, Commitment, LQ RE models, Reputation, Time-Inconsistency, Optimal Delegation
JEL Classification: C61, E52, E61
Suggested Citation: Suggested Citation