The findings suggest that the initial restatement may be a way of “cleaning house;” i.e., the new auditor requiring a restatement of financial reports issued during the tenure of the previous auditor. Investors may find this information valuable as they assess the potential outcomes of auditor switches. Significant corporate changes could distract firms from fully identifying all material misstatements in one restatement. Managers and directors should pay careful attention to the remediation of financial reporting weaknesses during these times. Restatement severity decreases among later restatements, but the market reactions to the first through third restatement announcements are similar in magnitude. These findings should be useful to investors and other market participants for understanding the implications of repeat restatements.

Files, R., Sharp, N. Y., & Thompson, A. M. 2014. Empirical Evidence on Repeat Restatements. Accounting Horizons 28 (1): 93-123.

Restatements often have costly consequences for companies and investors including increases in the cost of capital, litigation, and SEC enforcement proceedings, as well as auditor resignations, executive turnover, reduced investor confidence, and negative market reactions at the announcement date and beyond. The authors extend research on accounting restatements by investigating a unique set of restatement firms that exhibit a serious breakdown in the financial reporting process: namely, those firms that restate previously issued financial statements more than once within a relatively short period of time. The primary purpose of this paper is to provide empirical evidence on the characteristics of repeat restatements and the firms that experience them. The authors also evaluate the market consequences of repeat restatements, including their impact on stock returns and executive turnover.

Investigating repeat restatement firms is important for two main reasons. First, these companies have continuing difficulty correcting poor financial reporting and, as a result, expose their stakeholders to the negative consequences of multiple restatements. Second, despite the large literature on restatements, prior research has not specifically examined firms that restate more than once. Thus, little is known about repeat restatements firms, whether or how these firms differ from single restatement firms, or the specific factors that are associated with the likelihood of repeat restatements.

The authors use a sample of 2,212 restatements announced between 2002 and 2006 reported in Audit Analytics. In the sample, 1,003 restatements are announced by firms reporting only one restatement and 1,209 restatements are reported by 616 repeat restatement firms.

• 38 percent of the companies in the sample restate more than once, and the second restatement is announced, on average, 18.5 months after the first restatement.
• 31 percent of repeat restatement firms restate three or more times and 10 percent restate four or more times.
• Relative to the first restatement, firms’ second restatements are less likely to involve revenue or decrease reported net income, involve fewer accounts, and are more likely to be technical, suggesting that restatement severity attenuates as more restatements are announced.
• Repeat restatements are more likely among firms disclosing internal control (IC) weaknesses in the final year of the misstatement period.
• Repeat restatement firms, on average, experience announcement returns of -1.55 percent, -1.34 percent, and -1.54 percent for the first, second, and third restatements, respectively. This suggests that repeat restatements convey incremental, negative news to market participants.

In the authors' examination of firms that restate the same fiscal period they find:

• 57 percent of the first restatements in the sample are followed by second restatements with overlapping misstatement periods.
• Firms that employ a Big N auditor or those that switch auditors between the end of the misstatement period and the announcement of the initial restatement are less likely to report overlapping restatements.

http://www.auditingresearchsummaries.org