Differential Effects of Different Earnings Management Methods on Analysts' Forecast Behavior

초록

How do analysts react to the information asymmetry brought about by earnings management? Increased information asymmetry leads investors to demand more information from firms, resulting in a larger number of analysts following. On the other hand, it can also be interpreted as a higher level of risk for analysts and led to hesitation for analysts to follow the firms concerning their costs and benefits. Based on these contrary ideas, this paper examines the influence of information asymmetry measured via two different earnings management methods on analysts’ decision to follow a firm and on the quality of analysts’ forecast. We find that analysts avoid forecasting the firms with high level of real-activity-based earnings management while accrual-based earnings management does not have a significant impact on analyst following. This result implies analysts’ interests overwhelm investors’ demands on average. Furthermore, in testing the firms with slightly positive profits, real-activity-based earnings management increases the number of analyst following, on the contrary. This reversion suggests that investors’ demands grow larger in the suspected firms for earnings management and it overrides analysts’ risk in earnings forecast. We also analyze the quality of analysts’ forecast information to earnings management. Consistent with our expectation, real-activity-based earnings management impairs the forecast accuracy, shifts the consensus forecasts of the analysts upwards, and increases the dispersion among analysts. These findings verify the reason why analysts are reluctant to forecast the firms with real-activity-based earnings management.