The Phillips curve and information rigidity in Brazil

Abstract

This work aims at testing the null hypothesis of no sticky information against the alternative of sticky information using Brazilian data. The rejection of the null hypothesis allows us to derive the expected time between information updates. The median of market participants' predictions collected by Gerin/Bacen is used as a proxy to firms' expectation contained in the sticky information version of the Phillips curve. Our estimates imply that inflation expectations in Brazil are updated about once each five quarters, which in part can be attributed to reduced uncertainty about Brazilian inflation in the period of analysis.