No abstract

In this article I examine the response of investors and analysts of nonannouncing firms to the earnings report of the first announcers in the industry. The error in the earnings forecast of the first announcer is found to be informative about the errors in the contemporaneous earnings forecasts of subsequent announcers in the industry. However, investors and analysts do not appear to fully incorporate the information from the first announcers’ news in their revised earnings expectations for subsequent announcers. This apparent underreaction to the first announcers’ news leads to predictable stock returns for subsequent announcers in the days following the first announcement. Results of this study can be seen as further evidence of investor and analyst underreaction to publicly available information.

In the last decade, an emerging body of research looking at self-selected, corporate news events concludes that equity markets appear to underreact. Recent theoretical papers have explored why or how underreaction might occur. However, the notion of underreaction is contentious. Concern has focused on two issues-spurious results from unusual time periods and/or misspecified return benchmarks or methods. In this paper, we revisit the issue of underreaction by focusing on one of the most simple of corporate transactions, the stock split. Prior studies that report abnormal return drifts subsequent to splits do not appear to be spurious, nor a consequence of misspecified benchmarks. Using recent cases, we report a drift of 9% in the year following a split announcement. We consider fundamental operating performance as a source of the underreaction. Splitting firms have an unusually low propensity to experience a contraction in future earnings. The evidence suggests that investors underreact to this information. Analyst earnings forecasts are comparatively too low at the time of the split announcement and appear to revise sluggishly over time, a result consistent with underreaction by markets to corporate news events.

Prior research documents mean reversion in firm profitability and growth under the implicit assumption that profitability and growth of all firms revert to a common benchmark at the same rate. However, a large body of academic research suggests that there are systematic interindustry differences (e.g., industry barriers to entry) that differentially affect firm performance based on industry membership. We evaluate the relative forecast accuracy of mean reverting models at the industry and economywide levels and find that industryspecific models are generally more accurate in predicting firm growth but not profitability. * Georgetown University; †University of Miami; ‡Indiana University. We appreciate the comments of seminar participants at the McDonough School

No abstract