by Itzhak Ben-David (Ohio State University), John R. Graham and Campbell R. Harvey (Duke University)
Graham and Harvey are well known for their CFO survey of the equity risk premium. In this paper, they test whether these executives are miscalibrated, i.e. the distribution of their expectations is not correct. They use a unique panel that includes over 12,500 probability distributions of future stock market returns collected from top financial executives over nearly a decade. They find that executives are severely miscalibrated: realized market returns are within the executives’ 80% confidence intervals only 33% of the time. These executives seem therefore to be overconfident in their views. They show that miscalibration improves following poor market performance periods because forecasters are influenced by extreme negative past returns. Poor returns are salient events that seem to influence these executives. They also find that the degree of miscalibration is strongly correlated with the confidence interval that executives provide to their own-firm project returns. Finally, executives’ miscalibration is correlated with their own-firms’ level of investment.
This is an interesting paper that shows that the risk premium is very difficult to forecast even by professional executives and seems to cast doubt on the efficient market hypothesis since their expectations seem to be systematically miscalibrated and therefore wrong. This is also a cautionary tale for not relying on surveys to forecast macro and financial variables, even when the participants are sophisticated executives. At the same time, their forecasts can guide their investment decisions; public policy makers should take note and provide better public forecasts to complement the private miscalibrated forecasts.
You can read the paper here .