Print
E-MailQuinn Curtis is the Albert Clark Tate, Jr., Professor of Law at the University of Virginia School of Law; Dain C. Donelson is Associate Professor at the University of Texas at Austin Red McCombs School of Business; and Justin Hopkins is Assistant Professor at the University of Virginia Darden School of Business. This post is based on their recent article, forthcoming in Contemporary Accounting Research.
In our article, we examine how frequently firms restate when they materially misstate their financial statements using stock option backdating as the setting. After identifying firms that materially misstated earnings due to stock option backdating with 95% (99%) probability, we find that only 11.5% (16.1%) of these firms subsequently restated. Restating firms are larger, have greater board independence, higher litigation risk and ROA, a lower market-to-book ratio, less discretionary accruals, and are more likely to have a CFO that was not involved in backdating. Restating firms are also more likely to disclose other adverse news, face securities litigation, and replace their CFO than firms that appear to materially backdate but do not restate. Since nearly nine of ten firms failed to restate, our results should give pause to researchers who use restatements as an indicator of misreporting, and to regulators who levy penalties on those who do self-report.
The empirical evidence examining how often misreporting in financial statements remains concealed is limited because it is difficult to identify undisclosed misreporting. Similarly, why some firms reveal misreporting while others seemingly do not is largely unknown for the same reason. We use the setting of stock option backdating to address these questions by identifying firms that likely issued a material misstatement and examining whether the firm later restated. Using a sample of firms that show strong statistical evidence of backdating, we estimate the likelihood that misreporting will be revealed, and which firm characteristics are associated with the revelation of misreporting. Stock option backdating is a unique setting because it is possible to estimate the likelihood of material financial misreporting independent of its discovery and public revelation. To do so, we follow prior studies (e.g. Bizjak et al. 2009) and identify firms whose stock option issuance patterns provide statistical evidence of backdating.
Our results suggest that the decision to restate is associated with the firm’s information environment. Firms that operate under a bright light are more likely to restate. The likelihood of restating is increasing in the proportion of independent directors, although the presence of directors who were appointed during the CEO’s term tempers this effect. Restatements are also more common if the CFO when the backdating scandal erupted was not the CFO while the firm backdated stock options. This is not true of the CEO, possibly indicating that the CFO has more influence on the restatement decision or faces a harsher penalty from restating. Restatements are also increasing in the size of the firm and litigation risk, consistent with litigation risk encouraging the disclosure of adverse news (Skinner 1994). Finally, restatements are negatively associated with discretionary accruals and market-to-book ratio, and positively associated with ROA, consistent with these characteristics capturing capital market incentives to avoid disclosing adverse news.
As stock option backdating is a relatively unique setting, we next examine whether these conclusions generalize to other settings. We test whether the backdating firms we identify that restated are more likely to also reveal adverse news in other contexts. Evidence of such a pattern would indicate that restating firms operate in an information environment that prompts the discovery and revelation of adverse news. Specifically, we create a panel data set of 503 backdating firms and examine whether they are more likely to disclose adverse news in non-backdating contexts from 2000-2012 (this sample includes only firms that appear to backdate materially). Overall, firms that issue a backdating restatement are more likely to issue a non-backdating restatement and report material weaknesses in internal controls. Thus, we attribute the relation between backdating restatements and other adverse disclosures to managerial incentives and firm characteristics rather than the presence of adverse news.
Finally, we find that firms issuing a backdating restatement are more likely to face litigation and CFO turnover in periods outside the backdating scandal, after controlling for restatement likelihood and other litigation and turnover determinants. This is consistent with these firms’ information environments prompting adverse news disclosure, despite bad outcomes for management.
Overall, these results suggest that transparency with respect to misreporting may be relatively rare. Most misreporting is likely easier to conceal and less likely to attract attention than backdating. Nevertheless, firms with information environments that demand increased transparency, including those with truly independent directors and CFOs and larger firms are more likely to disclose problems.
The complete article is available here.
