This paper investigates whether the voluntary decomposition of consolidated earnings disclosures into industry segments has information content in the sense that such disclosures better enable investors to predict earnings. The broad rationale underlying the experimental design is that if segment disclosure does enable investors to better predict earnings then residual abnormal returns (after controlling for unexpected earnings) surrounding the earnings announcements of firms providing segment disclosures should on average be significantly lower than a matched’ group of firms that do not provide this type of disclosure. Using a short event window design, our results support this view.
|Number of pages||13|
|Publication status||Published - 1994|
- Segmented information