Analysts’ Responses to Earnings Management

Previous literature studying analysts’ earnings forecasts examines their properties without considering firms’ response to analysts’ forecasts. This study improves upon previous research by considering firms’ earnings management with respect to analysts’ forecasts. I hypothesize that analysts understand firms’ earnings management practices, and incorporate firms’ expected behavior into their forecasts. I demonstrate that for firms with high tendencies and flexibilities to manage earnings downwards, and / or firms with negatively skewed earnings, analysts account for earnings management practices by lowering the optimal forecasts. Comparing analysts’ consensus forecasts with earnings predictions generated by the least absolute deviation method, I find that analysts’ forecasts are systematically below the statistically generated earnings predictions for firm-quarters that have: high accounting reserves available to manage earnings downwards, high unmanaged earnings, low debt to equity ratios, negative forecasted earnings, and negatively skewed unmanaged earnings. These results suggest that analysts forecast below the otherwise optimal level in order to avoid the large optimistic forecast errors that occur when firms who cannot meet forecasts manage earnings downward. The test results also suggests that analysts forecast above the otherwise optimal forecasts when earnings are positively skewed, and / or when firms have high tendencies and flexibilities to manage earnings upwards. This paper also provides preliminary evidence indicating that analysts’ efforts to preserve their forecasts from the effects of downward earnings management practices contribute to the observed asymmetric distribution of earnings surprises.