Circadian Rhythm, Analyst Performance and Earnings Forecast

This paper examines how intrinsic circadian rhythm variation in the human body contributes to predictable day and night changes in the characteristics of earnings forecasts made by financial analysts. We find evidence supporting a casual relation between analyst performance and the day and night circadian cycle. Specifically, we find that financial analysts issue less accurate and more pessimistic forecasts at night. We also find that analysts who prefer issuing forecasts at night are less accurate, more pessimistic analysts. Additionally, we find that analyst performance follows the rising and falling trends of circadian rhythm, with analyst optimism and accuracy ascending over the morning but descending over the late afternoon and early evening. And we show that this circadian rhythm effect is significant when controlling for sunlight, disentangling our finding from the prior studies which mainly focus on sunlight induced bias on financial analyst performance. Our findings are robust to the inclusion of a number of control variables and analyst, firm and year fixed effects. By showing the effect of circadian rhythm on analyst performance, we contribute to the growing literature on behavioral and cognitive biases among financial market participants.