An Empirical Analysis of the Decline in the Information Content of Earnings Following Restatements

Abstract

Regulatory officials and market analysts have speculated that the loss of credibility in subsequently reported financial information is a long-lasting consequence of earnings restatements. I measure the information content of earnings using a standard earnings-returns framework over several years surrounding restatements to examine characteristics of the decline in the information content of earnings. Results indicate that although the information content of earnings declines following restatements, the loss is temporary. In particular, the earnings response coefficients for earnings announcements surrounding restatements exhibit a U-shaped pattern in which they are no longer significantly lower in the post-restatement period over an average of four quarters. The extent to which the earnings of restatement firms suffer a loss of information content varies across several dimensions. First, the duration of the loss is greater for firms that restate earnings to correct revenue recognition errors and for restatements that result in a large decline in the stock price at the announcement date. Second, there is not a loss in the information content of earnings for firms that make changes to their financial reporting governance structures following restatements. Overall, the evidence in this paper is consistent with a short-term decline in investor confidence regarding financial reporting following restatements, but shows that suspicion regarding the information loss of post-restatement earnings in the long-term is unwarranted.