Monetary Policy Under Uncertainty: Min-Max vs Robust-Satisficing Strategies

Abstract

We study monetary policy under uncertainty. A policy which ameliorates a worst case may differ from a policy which maximizes robustness and satisfices the performance. The former strategy is min-maxing and the latter strategy is robust-satisficing. We show an “observational equivalence” between robust-satisficing and min-maxing. However, there remains a “behavioral difference” between robust-satisficing and min-maxing. Policy makers often wish to respect specified bounds on target variables. The robust-satisficing policy can be more (and is never less) robust, and hence more dependable, than the min-max policy. We illustrate this in an empirical example where monetary policy making amounts to selecting the coefficients of a Taylor-type interest rate rule, subject to uncertainty in the persistence of shocks to inflation.