The determinants and trading consequences of automated information acquisition

Date

2023-12

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

I examine the determinants and trading consequences of automated acquisition of financial reports. Prior literature predominately studies human-based information acquisition, yet capital market participants are increasingly relying on automation to acquire financial reports. Using a novel approach to identify automated downloads of 10-Ks, I find strong evidence that automated and human downloads are positively correlated, suggesting that humans’ own interests influence how machines are programmed. I also find that machines selectively download the 10-Ks for firms with higher expected information frictions and/or misvaluation (e.g., smaller, value firms with complex filings). Using staggered XBRL adoption as an instrument for automated downloads, I find that automated downloads increase trading volume, particularly for firms with information frictions. However, I do not find evidence that automated downloads impact price movements or price efficiency. Taken together, the evidence suggests that automated acquisition of 10-Ks leads to differential interpretations, resulting in more trading but not necessarily more informed trading.

Department

Description

LCSH Subject Headings

Citation