The Geographic Dispersion of Google Search and the Market Reaction to Earnings Announcements

Abstract

We examine the impact of distance on investor search behavior, and the effect of geographic dispersion of investor search on the stock market response around earnings announcements. We find significant “local bias” in Internet search behavior. While more visible firms have more geographically dispersed search, there is significant additional variation in search dispersion. Motivated by theories of network effects and psychological distance, we predict and find that firms with a higher geographic dispersion of search experience higher abnormal trading volume, lower abnormal bid-ask spreads, and larger earnings response coefficients at the time of earnings announcements, as well as weaker post-earnings-announcement drift. These results hold both cross-sectionally and when examining changes in dispersion or propensity-score matched pairs. In addition, path analysis suggests that both network effects and investor psychology are significant drivers of the return results. Overall, our results suggest that geographic proximity affects search, and that firms with more geographically dispersed search experience better market responses to earnings announcements.

Keywords: geography, Google, investor attention, information asymmetry, earnings response coefficient, post-earnings-announcement drift, investor psychology, network effects, local bias

JEL Classification: G14, M41