Experimental study of implications of SFAS 131 : the effects of the new standard on the informativeness of segment reporting

This paper analyzes whether the new business segment reporting disclosure rules, SFAS 131, will actually provide capital market participants with more predictive ability than the previous rules. For this we conduct three experiments. Two experiments with advanced accounting students as subjects, where the experiments differ in the firm the subjects analyze, and the third with professional financial analysts. In each experiment we provide one group of subjects with accounting reports based on the new standard (New Rules Group, NRG), and another group with reports based on the old standard (Old Rules Group, ORG). We ask both groups to forecast several accounting and market values of a firm. We then compare the performance predictions and analyses of the two groups. Most of the forecasts of the NRG are neither significantly different from those of the ORG, nor significantly more accurate. Subjects also report the variables that they consider important in their analysis. 25% of the NRG students in Experiment I mention the segment data as being central in their decisions and 33% say they used segment or sector data. Among the analysts in Experiment II the corresponding percentages are 0% and 60%, respectively. Also in experiment III, where the subjects rank the top 4 variables they use in their predictions according to importance, segment repots receive a mediocre rank. It therefore appears that the reports according to the new rules, whereas noticeable by the subjects, do not have a major positive impact on their responses. The subjects also exhibit a considerable degree of overconfidence. -- segment reporting ; FASB 131 ; experimental economics ; overconfidence