Unusual Financial Communication: ChatGPT, Earnings Calls, and Financial Markets

Abstract

We investigate unusual communication patterns between firm executives and financial analysts. Leveraging ChatGPT’s general knowledge about the structure of earnings conference calls, we identify 25 dimensions of unusual financial communication, which help explain how financial markets respond beyond the impact of traditional sentiment and content analyses. The markets respond negatively to both unusual communication styles and content, leading to decreased stock returns, increased trading activity, higher volatility, option-implied uncertainty, and downward revisions of next-quarter earnings forecasts by analysts. Our study demonstrates how unusual communication can be an important impediment to how market participants process and impound information.

Keywords: ChatGPT, Earnings Calls, Communication, AI, Financial Analysts, Uncertainty

JEL Classification: C45, C88, D80, G11, G12, G14, M41