Understanding the Market Reaction to Audio Cues in Earnings Conference Calls

• Applicant: Professor Dr. Oliver Hinz
• Subject Area: Management and Marketing
• Term: since 2022
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 512670450

In regular earnings conference calls firms aim to inform analysts and investors about their economic conditions and about their outlook on future business success. These calls intend to reduce information asymmetries between management and (prospective) stakeholders of the firm. While reported data and facts are publicly available to everyone, contemporary AI methods make it possible to obtain additional pieces of information about the current and future state of the firm. Using earnings conference call transcripts, previous research has shown that natural language processing methods help explain the market reaction to earnings conference calls and predict future firm performance. With this proposed project, we aim to go one step further and leverage not only the text transcripts but also the audio tracks with state-of-the-art AI methods. Put differently, we do not only try to understand the impact of “what has been said” but also “how it has been said” in these calls. In doing so, we study whether this piece of information, which we call Audio Cues, is predictive for future firm performance and to what extent investors incorporate it into prices.

• DFG Programme: Research Grants
• Co-Investigators: Dr. Kevin Bauer; Professor Dr. Alexander Hillert