| Scope and method of study. The purpose of this dissertation is to assess the usefulness of geographic segment earnings disclosures. In this dissertation, disclosures of geographic segment earnings will be considered useful (1) if there is evidence of use by investors in valuing securities of multinational firms and (2) if earnings disclosed by geographic segment are useful in explaining betas and risk-adjusted security returns of multinational firms' securities. In the first instance, raw returns are regressed on geographic segment earnings using long-window associations and leading-period returns. If the market prices earnings of all geographic segments similarly, then geographic segment earnings provide no information to the market beyond that provided by consolidated earnings. In the second instance, a returns-generation model of multinational firms' securities is developed. Based on the model, a firm's domestic accounting beta should be positively related to its security beta, and the firm's foreign accounting beta should be positively related to its risk-adjusted security returns. Measures of domestic and foreign accounting betas are generated using geographic segment earnings disclosures. Finding a positive association would be an indication that disclosures of geographic segment earnings provide relevant accounting data.;Findings and conclusions. The results provide some evidence that geographic segment earnings are used to value securities. However, this relationship is unstable, depending largely upon time period and sample composition. Also, the results show that unless geographic segments are consistently defined for three or more years, they are not used by the market to value securities. The results further show that the domestic accounting beta is positively related to beta but this relationship is not more significant than the relationship between beta and total accounting beta. There appears to be no relationship between foreign accounting beta and risk-adjusted security returns. The results of this dissertation suggest that current disclosure of geographic segment earnings disclosures are inadequate and that standard setters should revise current geographic segment disclosure standards. |
