Expectations, learning and the costs of disinflation: experiments using the FRB/US model

The costs of disinflation are explored using the Board's new sticky-price rational expectations macroeconometric model of the U.S. economy, FRB/US. The model nests both model consistent and `restricted-information rational' expectations. Monetary policy is governed by interest-rate reaction functions of which two are considered: the well-known Taylor rule and another rule that is more aggressive and richer in its specification, estimated using data for the last 15 years. Agents are required to learn of shifts of the inflation target using linear updating rules. The simulated costs of disinflation are compared with other estimates of sacrifice ratios.