Optimal Monetary Policy under Risk and Uncertainty

This paper seeks to characterise optimal monetary policy rules in the presence of risk and uncertainty. I explore a situation in which the true parameters and the true structure of the economy are unknown to the policymaker, and he is reluctant to make a decision based on a single distribution estimate (i.e. he faces Knightian uncertainty). I show analytically that if the policymaker does not know the true structure of the economy he will be more cautious than in the case of only parameter risk. Further, I show that Knightian uncertainty can also lead to an extra precautionary motive when one considers its interaction with parameter risk. In a simple exercise, I provide empirical estimates that demonstrate that adjustments due to parameter and structural risk and Knightian uncertainty can potentially be quite large.