Investigates the degree to which the superiority of analysts' corporate earnings forecasts is associated with firm characteristics. Analyst advantage over a time-series model to past earnings variability; Relationship between analyst advantage and the amount of coverage in the `Wall Street Journal' newspaper.

[This research investigates the degree to which the superiority of analysts' earnings forecasts (relative to a univariate time-series model) is associated with certain firm characteristics. The analysts' information advantage is characterized as being related to private information-gathering incentives, and to the amount of information disseminated about the firm. The objective is to determine whether analyst forecast superiority is related to firm characteristics not examined in previous research. Specifically, the investigation relates the analyst advantage over a time-series model to past earnings variability and the extent of coverage in The Wall Street Journal. Statistical controls were employed for the market value of the firm's common stock, the firm's number of lines of business, and the time lapse between the end of the previous fiscal quarter and the release of the earnings forecast. The methods of data analysis consist of estimating OLS regressions, heteroscedasticity-consistent estimators, and bootstrapping techniques. The results indicate, first, that the analyst advantage in forecast accuracy over a time-series model is materially related to the historical variability in the earnings time series. Second, no positive relation is evident in our data between the analyst advantage and firm size, a result that is at variance with some previous research. Third, the analyst advantage is positively related to the amount of coverage in The Wall Street Journal Index, which is consistent with the intuitive notion of prior research that analysts' forecasts improve as more information becomes available. Finally, an attempt was made to ensure that the results were not caused by violations of classical regression assumptions. This was accomplished by explicitly correcting for a nonconstant variance, and by allowing for cross-correlation using bootstrapping. The asymptotic results are very similar to the bootstrapping results, but neither adjustment has altered the primary findings using OLS.]

We document that the effect of Regulation Fair Disclosure (FD) on public management earnings forecasts (MFs) is asymmetric. Our results suggest that FD increased managers' use of MFs as a downward-guidance mechanism to help achieve meeting or beating earnings expectations. This effect is more pronounced when existing analyst forecasts are optimistic and when firms had selective disclosure policies pre-FD. We also find that the increased use of MFs as downward guidance leads to post-FD reductions in MF quality (accuracy and informativeness) for the downward guiding MFs that are most likely meet/beat motivated, while quality improves for upward-guiding MFs. Finally, our evidence suggests that results from prior research about FD-induced changes in information environment variables, such as analyst forecast quality and investor trading activities, depend on whether the firm issues MFs and whether those MFs are downward guiding. Data Availability: All data are available from public databases identified in the paper.