Expectations, Learning and the Design of Monetary Policy Rules

Robert Tetlow and Peter von zur Muehlen - Federal Reserve Board

This paper investigates the choice of monetary policy rules by a monetary authority that is dealing with an ill-informed public. We show that if the private sector must learn the true structure of the rule throught direct observation and estimation, it is very often the case that a simple, "suboptimal" rule will outperform the more complicated rule that would be optimal under full information. This is demonstrated through simulations using a small sticky-price rational expectations model under various conditions of prior beliefs on the part of the private sector. The results lend support for notion that simplicity and transparency should be objectives in the design of policy rules.

Scheduled for Session 1.3 Optimum Policy Making Under Uncertainity