Optimal monetary policy with uncertainty

Abstract

This paper investigates the effect of uncertainty on the optimal policy response. Uncertainty plays a central role in both monetarist and neo-Keynesian policy prescriptions. Monetarists advocate cautious policy responses to exogenous shocks because they believe there is a long and variable (uncertain) lag between the policy response and its ultimate impact on the economy. Neo-Keynesians on the other hand believe that imperfect information can cause a dynamic disequilibrium in which shocks are multiplied if they are not offset by policy. A simple dynamic random coefficient model is used to approximate this uncertainty. It is shown that when uncertainty about the impact of policy is dominant the optimal policy converges to a fixed money growth rate, but when uncertainty about the transition dynamics is dominant a very active countercyclical policy is optimal. When both sources of uncertainty are present, as they are in general, the optimal policy response depends on the relative uncertainty and the policy may become more aggressive as uncertainty about the impact of policy increases.