Signal in the Noise: Trump Tweets and the Currency Market

Abstract

In this paper, we conduct a textual analysis of Trump tweets. Our method extracts the signal from the noise, by identifying the subset of tweets that contain information on macroeconomic policy or trade content. Informative tweets result in a USD appreciation and a decline in intraday volatility, reflecting Trump's optimistic views on the U.S. economy. These effects persist after controlling for macroeconomic announcements. We rationalize our findings within a model of Bayesian traders that interpret Trump tweets as a public signal in the FX market. Currency returns are driven by a bias between the public signal and speculators' expectations.

Keywords: Foreign exchange market, textual analysis, Trump, X (Twitter)

JEL Classification: F31, G12, G15