The Causal Impact of Media in Financial Markets

Abstract

It is challenging to disentangle the causal impact of media reporting from the impact of the events being reported. We solve this problem by comparing the behaviors of investors with access to different media coverage of the same information event. We use zip codes to identify 19 mutually exclusive trading regions, corresponding to 19 large U.S. cities and local newspapers (e.g., the Houston Chronicle). For all earnings announcements of S&P 500 Index firms, we find that local media coverage strongly predicts local trading, after controlling for characteristics of the earnings surprise, firm, local investors, and reporting newspaper(s). Moreover, the local trading-local coverage effect: 1) depends precisely on the specific timing of local reporting (e.g., one day after the earnings announcement, two days afterward, etc.) and 2) disappears entirely during extreme weather events, which leaves media content unchanged, but disrupts transmission to investors.

Keywords: public news, media coverage, media effect

JEL Classification: G14, G12