Information Flows and Disagreement

Abstract

The aim of this study is to assess the extent to which the degree of heterogeneity of inflation expectations is driven by the flow of information related to current and future price developments. To that end, we follow three routes: i) We propose different measures of information flow that have either a sender or a receiver perspective; ii) We present empirical results for the US and selected EU countries that aim to corroborate the hypothesis that news have the ability to densify expectations, i.e. to reduce forecast heterogeneity; and iii) We augment some otherwise standard models of expectation formation by allowing the individual updating frequency to depend on the observed measure of information flow; since the updating frequency is higher at times of high inflation and decreasing thereafter, this mechanism can contribute to upward biases in inflation expectations over long periods of time.