New research finds that material restatements of earnings that involve correcting accounting irregularities and not just honest errors, lead to a loss in company credibility with investors lasting substantially longer than previously documented.

Investors, according to the new study, have a diminished response to earnings reports of such companies for an average of close to three years.

The study, which appears in the current issue of The Accounting Review, a journal of the American Accounting Association, found that companies with a material restatement experience a significant decline in the “earnings response coefficient,” a measure of investors’ response to earnings reports, for 11 quarters after the restatement is announced.

“This is a considerably longer period than the three quarters reported in prior research, suggesting that the decline in credibility and information content of earnings after restatements is not short-lived," said the researchers, Xia Chen and Qiang Cheng of Singapore Management University and Alvis K. Lo of Boston College, in their paper.

The study found that the magnitude of the drop in earnings response coefficient was more than half the ERC in the pre-restatement period, largely driven by announcements of quarterly results that are more likely to be subject to credibility concerns, such as those reporting good news and those reporting larger accruals of non-cash accounting items that are more subject to manipulation than cash.

"Investors are more suspicious of good news than of bad news disclosed by the firm," wrote the researchers. "Prior research suggests that high-accrual firms have lower earnings quality."

But the study also found that companies that reported material restatement were able to restore more credibility with investors when they took steps such as removing the CEO and CFO, dismissing the outside auditing firm, and replacing the audit committee chairman. They experienced a relatively shorter-lived drop in their ERC, ending in the first year after the announcement of a restatement.

Approximately 27 percent of this group of companies dismissed both the CEO and CFO within a year of the announcement, while 30 percent replaced the chair of the audit committee and 15 percent dismissed the external auditor. Companies that took one or another of these measures  to improve reporting credibility did not sustain a reduced ERC beyond the fourth or fifth quarter following their restatement announcements.

The paper analyzes investors’ responses to 1,208 restatement announcements made by public companies during the period January 1997 through June 2006. The researchers found that 343 restatements were classified as material on the basis of meeting one of three conditions: the firm explicitly used variants of "fraud" or "irregularity" in the restatement; there was a related investigation by the Securities and Exchange Commission or the U.S. Department of Justice; or there was a non-SEC independent investigation into the misstatement, such as a probe launched by a company's audit committee.

The researchers found a sharp difference in investor response to material restatements compared to others. For the former, the announcement led to an immediate drop in stock price averaging 7.2 percent for the three days following the restatement announcement, while the drop in the latter was only 1.8 percent. The difference in the duration of mistrust was equally dramatic: for material restatements, the decline in the ERC from what it was before the restatement announcement lasted for an average of 11 quarters, while for non-material restatements the ERC declined for only one quarter.

In short, investors are relatively forgiving when it comes to honest errors but not when it comes to material irregularities.

Not only does the drop for companies with material restatements last a longer time, but it is of greater magnitude as well. The ERC dropped 56 percent, from a coefficient of about 3.3 during the five quarters preceding the restatement announcement day to an average of about 1.45 for the 11 quarters post-announcement. The professors found that "the reduction in the ERC shows a decreasing trend, being highest in the first year and the lowest in the third year after the restatement announcement, consistent with material-restatement firms gradually regaining investors' trust over time."

In conclusion, the researchers asserted that in addition to reconciling results of previous research, the study "has important implications for investors, regulators, firms and accounting researchers," helping all "to better understand the adverse consequences of aggressive financial reporting and the impact of remedial actions taken by firms to regain investors' trust."

The study, the study entitled "Is the Decline in the Information Content of Earnings Restatements Short-Lived?" appears in the January/February issue of The Accounting Review, published six times a year by the American Accounting Association.