The Listenability of Disclosures and Firms’ Information Environment

Abstract

This paper examines how the listenability of corporate vocal disclosures affects firms’ information environment. Listenability refers to the ease with which humans can comprehend spoken language. We introduce a new measure of listenability by analyzing audio recordings of 56,989 quarterly earnings calls using an advanced, supervised machine learning model trained on 680,000 hours of labeled audio data. We find that enhanced listenability of the presentation portion of earnings conference calls is associated with higher quality manager-analyst interactions during the Q&A session of the call and better information production by analysts after the call. Further analyses reveal that higher listenability is associated with stronger price responsiveness, higher stock price informativeness, faster price formation, and lower bid-ask spread following the call. These results are consistent with clear and effective corporate oral communications facilitating a better information environment.

Keywords: listenability, vocal disclosure, analyst forecast, information environment

JEL Classification: G14, G18, M41