Window Dressing the Interim-Earnings Report: An Empirical Assessment for Firms Initially Going Public

Abstract

Several years ago, Green wrote about some of the accounting problems in measuring interim-period earnings for seasonal businesses.' Since then, several empirical studies have examined the usefulness of interim reports in predicting annual results.2 These studies have primarily involved large, established firms whose shares are traded on the New York and other stock exchanges. Since interim-earnings reports are regularly issued by these firms, information is available on the historical relationship between interim and annual earnings (i.e., the extent of seasonality. the nature and magnitude of year-end accruals, etc.). Also, there does not seem to be any rationale for these firms to systematically window dress the specific interim-earnings report examined. Firms making their initial public offering of common stock provide financial information under somewhat different conditions. First, they are required to present in their registration statement annual-earnings reports covering only the five most recent accounting periods prior to the offering. In addition, these firms present earnings data for an interim period if more than 2 or 3 months have elapsed between the previous year end and the offering date. The earnings for this interim period are typically shown compared with earnings of the same period during the previous year. Thus, information regarding the historical relationship between interim and annual earnings for these firms is more limited than for the larger firms previously studied. Moreover, the managements of these firms seem to have a particular incentive to report a favorable earnings performance for the most recent interim period in an effort to influence the offering price or facilitate the offering. Statements regarding the importance of a significant increase in the