Optimal Policy Perturbations


Model mis-specification remains a major concern in macroeconomics, and policy makers must often resort to heuristics to decide on policy actions; combining insights from multiple models and relying on judgment calls. Identifying the most appropriate, or optimal, policy in this manner can be challenging however. In this work, we propose a statistic -the Optimal Policy Perturbation (OPP)- to detect "optimization failures" in the policy decision process. The OPP does not rely on any specific underlying economic model, and its computation only requires (i) forecasts for the policy objectives conditional on the policy choice, and (ii) the impulse responses of the policy objectives to shocks to the policy instruments. We illustrate the OPP in the context of US monetary policy decisions. In forty years, we only detect one period with major optimization failures; during 2010-2012 when unconventional policy tools should have been used more intensively.