Inflation Persistence, Noisy Information, and the Phillips Curve

Abstract

vast literature has documented that US inflation persistence has fallen in recent decades, but this finding is difficult to explain in monetary models. Using survey data on inflation expectations, I document a positive co-movement between ex-ante average forecast errors and forecast revisions (suggesting forecast sluggishness) from 1968 to 1984, but no comovement afterward. I extend the New Keynesian setting to include noisy and dispersed information about the aggregate state and show that inflation is more persistent in periods of greater forecast sluggishness.My results suggest that changes in firm forecasting behavior explain around 90% of the fall in inflation persistence since the mid-1980s. I also find that the changes in the dynamics of the Phillips curve (PC) can be explained by the change in information frictions. After controlling for changes in information frictions, I estimate only a modest decline in the slope. I find that a more significant factor in the dynamics of the PC is the shift towards greater forward-lookingness and less backward-lookingness. Finally, I find evidence of forecast underrevision in the post-COVID period, which explains the increase in the persistence of current inflation.

Keywords

JEL codes