The SEC has long asserted that earnings management practices result in adverse consequences for investors. We examine whether SEC oversight affects firms' accounting quality in terms of earnings management trade‐offs. We expect that increased firm‐specific regulatory scrutiny, in the form of an SEC comment letter, will induce management to switch from accrual‐based earnings management (AEM), which is a main focus of the SEC, to real‐activities‐based earnings management (REM), which is not likely to be commented on in the SEC's review process. Consistent with our predictions, we find that AEM is lower and REM is higher following the receipt of a comment letter, relative to non‐comment‐letter years and a propensity‐score‐matched sample of non‐comment‐letter firms. However, we do not find a significant difference in total earnings management (i.e., the sum of AEM and REM), suggesting that the higher REM acts as a substitute for lower AEM activity. We further find that our results are driven by accounting comments relating to estimates and accruals and not by classification‐only comments, which suggests that a comment letter that does not question specific issues associated with estimates and accruals is not a strong enough signal to induce the firm to change earnings management behavior. Additionally, the shift to REM is attenuated for firms with high institutional ownership. These results collectively suggest that the comment letter process effectively constrains AEM but has the unintended consequence of firms, on average, switching to REM.

No abstract

No abstract

In this study, we examine whether investors' actions to acquire accounting information are predictive of future firm performance because these actions partially reveal investors' private expectations of this performance. Using a database of EDGAR downloads, we find some evidence that information acquisition of accounting reports by EDGAR users is, on average, predictive of future firm performance. We then determine the identity of the EDGAR user and examine whether different users' private expectations will be relatively more predictive of subsequent performance. We find that the information acquisition activities of more sophisticated institutional users (e.g., hedge funds, investment banks) are more strongly associated with future performance than are those of less sophisticated retail users. Finally, we find that information acquisition by sophisticated institutions is a leading indicator of their equity holdings. In summary, this study provides evidence of predictive information embedded in sophisticated investors' actions to acquire accounting information.

No abstract