Discretionary Dissemination on Twitter

Abstract

Using an unsupervised machine learning approach to analyze 24 million tweets posted by S&P 1500 firms from 2012 to 2020, we find that firms are more likely to tweet financially related messages around significantly negative or positive news events such as earnings announcements and the filings of financial statements. Such a convex U-shaped relation between the likelihood of posting financial tweets and the materiality of accounting events becomes stronger over time. The evidence suggests that the conclusions drawn on earnings announcements using an early period sample – that firms are less likely to disseminate information on Twitter when the news is bad and material – should be revised. We also show that a machine learning algorithm (Twitter-LDA) is superior to a dictionary approach in classifying short messages like tweets. Our intraday analyses provide consistent evidence.

Keywords: Social Media, Discretionary Dissemination, Disclosures, Twitter

JEL Classification: G14, L30, M14, M15, M40