An Integrated Analysis of the Association between Accrual Disclosure and the Abnormal Accrual Anomaly

Abstract

We analyze whether the nondisclosure of accrual information at earnings announcements might contribute to the abnormal accrual anomaly. Prior studies draw inferences about the pricing of (abnormal) accruals at earnings announcements and around the filing dates through analyses of the long-term stock performance of high and low (abnormal) accrual firms. In this study, we take a more integrated approach by directly analyzing the pricing of abnormal accruals around the earnings announcement, around the SEC filing, and over the year subsequent to the SEC filing, conditional on whether a firm discloses accrual information at the earnings announcement. We find no evidence of accrual mispricing for firms that disclose accrual information at the earnings announcement. For these firms, the market is able to differentiate the discretionary from the nondiscretionary components of the earnings surprise. In contrast, as expected, the market fails to distinguish between the discretionary and the nondiscretionary components of the earnings surprise for firms that do not disclose accrual information at the earnings announcement. These firms experience a stock price correction around the filing date. However, the correction is only partial, resulting in a post-filing drift.