A0162
Title: Going viral: Inflation narratives and the macroeconomy
Authors: Max Weinig - Universität Hamburg (Germany) [presenting]
Ulrich Fritsche - Universitat Hamburg (Germany)
Abstract: In recent years, there has been increasing interest in the analysis of narratives in macroeconomic research. Our paper contributes to this research by proposing a way to extract identified economic narratives from media reports. Therefore, this paper applies a keyword-assisted topic model (keyATM) and a semisupervised semantic scaling technique (LSS) to measure inflation narratives as tone-adjusted time series to a large news corpus covering five years of news coverage. This approach enables us to measure the prevalence and spread of inflation narratives over time and to examine the role of these narratives in aggregate macroeconomic expectations. Using Granger causality tests and local projections, we provide empirical evidence on the dynamics between inflation narratives and inflation expectations. Moreover, the paper highlights the vast heterogeneity across short-term and mid-term expectations as well as across socioeconomic groups.
