Monetary Policy, Heterogeneous Expectations and Structural Uncertainty

Abstract

This paper shows that monetary policy does and should respond systematically to time variation in ex-ante uncertainty and heterogeneity in private sector’s views over the business cycle. Empirical tests are initially conducted on the basis of an augmented forward-looking Taylor rule framework, modified to account for learning and robustness. Normative justification is further provided by evaluating the optimal forecast-based monetary policy response under imperfect knowledge given a set of heterogeneous nested reference structural models, estimated to best fit private sector’s forecasts in addition to contemporaneous data.