Inattentive agents and inflation forecast error dynamics: A Bayesian DSGE approach

This paper studies a dynamic stochastic general equilibrium model with inattentive agents to explain the dynamic patterns of inflation forecast errors. Empirical findings suggest that inflation forecasts implied by the model are quite similar to survey expectations. This model dominates a full-information rational expectations model in its ability to account for why actual inflation is systematically associated with inflation expectations and survey forecast errors, and for generating excessively persistent forecast errors. Historical decomposition analysis reveals that supply shock and measurement errors in inflation are found to be dominant forces driving variations in inflation forecast errors, while secular shifts in inflation are generated mainly by supply and inflation target shocks.