Communicating with Vocal Emotions
Talk by Oleksandr Talavera (Birmingham) as part of the Research Seminar Series of the IOS Economics Department.
This paper explores the interpersonal vocal communication of managers with analysts during earnings conference calls. We apply a novel machine learning technique on a sample of more than 20,000 earnings conference calls of S&P 500 firms to generate voice emotions measures. Focusing on analyst-manager conversations, we find evidence of reciprocal interactions when managers respond with a similar vocal emotion to analysts. We also document that managers, who dialogue with a female analyst, exhibit a more positive vocal response. Moreover, female managers and younger managers are more likely to display negative vocal emotions than their male and older colleagues. Further analysis shows that investors and analysts incorporate emotionally charged information in trading and earnings forecasts.