Date of Graduation

8-2019

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Business Administration (PhD)

Degree Level

Graduate

Department

Accounting

Advisor

Vernon Richardson

Committee Member

Michael Crawley

Second Committee Member

Asher Curtis

Third Committee Member

Gary Peters

Keywords

Earnings, Herding, Social Media, Twitter

Abstract

I use data from StockTwits and Twitter to provide evidence that investor attention on social media in the period before earnings is related to short-term overvaluation, consistent with bullish investors herding around common information. In the 2 to 60 days after earnings, returns for companies in the highest quintile of pre-earnings announcement investor attention are 4.2 percent lower than those of companies in the lowest quintile. I find evidence that the negative post-earnings drift result found in this study is related to investors waiting until after earnings are announced to enact costly arbitrage strategies. I further examine intra- and inter-network herding and find evidence that social media influences investors beyond the population of active users. This study contributes to prior literature on herding, social media, and speculation and arbitrage.

Available for download on Sunday, July 19, 2020

Share

COinS