Are happy CEOs better at forecasting?
Study finds that joyful CEOs are more likely to issue forecasts, less likely to miss their forecast targets, and exhibit lower optimistic bias in their forecasts
This is the first study to look at CEO happiness as a unique personality trait and evaluate its impact on the forecast properties of corporate earnings. To do so the authors, Karel Hrazdil, Xin Li, and Nattavut Suwanyangyuan used the IBM Watson Tone Analyzer to estimate CEOs’ levels of joy. They kept all CEO responses with at least 100 words (the minimum quantity required by the Watson Tone Analyzer) and used these responses as inputs for the software for each conference call in which the CEO responds to analysts’ questions. The Watson Tone Analyzer then calculates “Joy” as a percentile score for Watson’s sample population. Their sample comprised 6127 conference calls over the span of 12 years (2002-2013).
According to the study, happy CEOs are more likely to issue and meet forecasts, and their forecasts are less biased and more accurate. For firms with more joyful CEOs who issue forecasts, the study also finds that analysts revise their forecasts upwards, and produce forecasts that are less dispersed and more accurate.
This is yet another great example of how natural language processing can help us infer levels of joy based on recorded information from over 6000 conference calls and relate it to future corporate performance.