There has been a substantial increase, since 2004, in the number of firms that announce annual earnings before audit completion as opposed to after audit completion. In this study, we argue that earnings announced before audit completion are associated with lower financial reporting quality and investor perceptions that earnings are more likely to be overstated. Consistent with this expectation, we document that the market places more (less) weight on good (bad) earnings news for earnings announced after audit completion relative to earnings announced before audit completion. We continue to find this differential market response when we expand the returns window to include the 10‐K filing date, suggesting that the differential response is not driven by investors' temporary concerns about earnings revisions between the earnings announcement and the 10‐K filing date or by differential GAAP disclosures in the earnings announcement, as suggested in prior research. Finally, as a direct test of financial reporting quality, we show that earnings announced with a completed audit are less likely to be restated in the future, are less likely to meet or beat expectations, and are associated with fewer income‐increasing discretionary accruals than those announced with an incomplete audit.

While accounting research has extensively examined initial guidance disclosures, the disclosures that managers make when initial forecasts become materially inaccurate have received much less attention. These updates are unique because managers are communicating that their initial forecasts are no longer correct. In this context, we examine how earnings forecast withdrawals affect managers' reporting reputation, relative to earnings revisions and nondisclosure. While managers face immediate negative market consequences after withdrawals, they enjoy reputational benefits (in the form of improved credibility) in the long run when guidance updates resume. In contrast, reporting reputation does not improve for managers who revise forecasts or for those who choose not to update at all. Difference-in-differences analyses confirm this incremental boost to credibility that is associated with withdrawals. This evidence suggests disclosing what managers do not know may be as important as disclosing what they do know when building a reporting reputation.

We document a growing disparity in earnings disclosure mechanisms. Firms are increasingly disclosing earnings announcements (EA) concurrently with the 10-K filing instead of first issuing a ‘stand-alone’ EA. Firm adoption of concurrent EA/10-Ks is associated with lower investor sophistication, greater impediments to producing timely and reliable earnings information, and greater industry-level concurrent reporting. Concurrent EA/10-Ks differ from stand-alone EAs in that investors anticipate more information in the EA, disclosures are preempted by industry peer EAs, the market reaction is muted even when controlling for EA timing, and post-earnings-announcement drift is greater.

Little is known about how the SEC selects its targets for investigation. We study this subject using a new database of formal SEC investigations. We predict and find that case selection is associated with a firm's (i) likelihood of regulatory noncompliance, (ii) exposure to private sector scrutiny, and (iii) conspicuous public trigger events. The relationship between investigations and regulatory noncompliance and private sector scrutiny preferences is sensitive to SEC constraints, whereas the relationship with triggers is not. We also examine the association between investigation motives and enforcement actions, an important SEC outcome reported to Congress. While regulatory noncompliance-motivated and public trigger-motivated investigations are more likely to result in public charges, specifically when the SEC is constrained, private sector scrutiny-motivated investigations are less likely to result in public charges. Finally, investigation rates of potential targets are associated with the career trajectories of SEC personnel, while investigation outcomes are not.

Stock exchanges are important intermediaries in how firm information enters price. Trading halts are a key tool, often exercised at the exchanges' discretion, to prevent extraordinary price volatility when new information arrives. We investigate how exchanges use discretion and whether the discretion alters the effectiveness of the halts. We provide evidence consistent with halts reflecting the preferences of listed firms rather than the stated exchange objectives (i.e., minimizing excess volatility and off‐equilibrium trades). Furthermore, when exchanges exercise more discretion (unexplained by firm and information characteristics), the halts are less effective. Specifically, halts with more discretion are less likely to resume trading with efficient prices and are more likely to have been called unnecessarily (i.e., little to no price movement during the halt). These findings are consistent with exchanges using halts to cater to listed firms rather than to meet exchange objectives such as minimizing excess volatility or avoiding trades at off‐equilibrium prices.