Abstract:
This paper investigates the causal relationship between economic policy narratives, derived from President Trump's tweets and tweeting behavior, and stock market uncertainty. To this end, I define different event types based on the occurrence probability of identifted narratives or unusual tweet behaviors. High-frequency market uncertainty responses to different events are recovered using time-series regressions. Events regarding foreign policy, trade, monetary policy, and immigration policy exhibit a signiftcant effect on market uncertainty. Impulse responses become signiftcant between one and three hours after the event occurs, for most of the events. Furthermore, behavior events, such as increases in the tweet or retweeted counts above their average, matter for stock market uncertainty.