Narrative-driven fluctuations in sentiment: Evidence linking traditional and social media

This paper studies the role of narratives for macroeconomic fluctuations. We micro-found narratives as directed acyclic graphs and show how exposure to different narratives can affect expectations in an otherwise standard macroeconomic model. We capture such competing narratives in news media's reports on a US yield curve inversion by using techniques in natural language processing. Linking these media narratives to social media data, we show that exposure to a recessionary narrative is associated with a more pessimistic sentiment, while exposure to a nonrecessionary narrative implies no such change in sentiment. In a model with financial frictions, narrative-driven beliefs create a trade-off for quantitative easing: extended periods of quantitative easing make narrative-driven waves of pessimism more frequent, but smaller in magnitude.