Determinants of the Timing of Quarterly Earnings Announcements

Abstract

Previous empirical research provides descriptive evidence on the timing pattern of earnings announcements but does not attempt to investigate potential explanations. Because some stakeholders are not likely to find it cost-effective to monitor the firm actively, managers have the opportunity to influence the perceptions of relatively uninformed stakeholders through accounting decisions such as the timing of earnings announcements. We provide evidence on this stakeholder explanation of timing decisions by identifying a setting that has the potential to discriminate between this and a confounding explanation for the normal timing pattern suggested in prior studies. Specifically, if managers rushed to report bad news following the October 1987 stock market crash in the belief that the ongoing market chaos reduced the reactions of stakeholders to the news, “normal” timing patterns would be (at least partially) reversed. Our results are generally consistent with prior research in that we document a (somewhat weak) association between earnings news and timing. However, consistent with the stakeholder explanation, we find that the earliest reporting group exhibited, on average, bad news. Thus, we infer that timing decisions for some firms are motivated by a desire to minimize the adverse reaction of stakeholders to bad news. In addition, we report evidence that suggests managers reduced the magnitude of reported earnings following the crash. This evidence corroborates our conclusion that managers are attempting to influence stakeholder perceptions of the firm's earnings performance.