A Study Of Foreign Earnings Management Using An Empirical Distribution Approach

Effective January 1, 1998, SFAS 131 changed the way managers voluntarily disclose business, geographic and operating segments of their companies. Reporting geographic segments is specifically related to the reporting of overseas operations of multinational companies. This study analyzes the possibility of earnings management in foreign earnings and how SFAS 131 might have changed the patterns of foreign earnings distribution. I use an empirical distribution approach outlined by Burgstahler and Dichev (1997) in examining whether foreign earnings are managed to avoid earnings decreases and losses. Results show that foreign earnings are managed to avoid losses but not earnings decreases. In addition, SFAS 131 does not significantly reduce earnings management in foreign earnings. What is more important, however, is whether firms have reported an increased number of segments in the post-SFAS131 period. The empirical evidence is consistent with the fact that managers who voluntarily disclose information are less likely to engage in earnings management. I also test earnings management related to domestic earnings. The study of domestic earnings is necessary because both foreign and domestic earnings are components of total earnings and can be used for earnings management purposes. In addition, because Beaver et al. (2007) suggest income taxes partially cause the discontinuity observed in Burgstahler and Dichev (1997), I use pretax foreign, pretax domestic, after-tax foreign and after-tax domestic earnings for all analyses in order to examine the income tax effect. I do not find taxes, especially foreign taxes, to be a significant factor in my study. I further replicate Burgstahler and Dichev (1997) using the sample period of my study. Consistent with my findings, evidence of avoidance of losses is significant but not for avoidance of earnings decreases. Future research is necessary to address what causes the changes.