International diversification, SFAS 131, and post-earnings announcement drift

Abstract

Using data from 1990-2013, we show 1) the serial correlation of analyst forecast errors increases in the extent of international diversification, 2) PEAD based on analyst forecast errors increases in the extent of international diversification, and 3) the impact of international diversification on the serial correlation of analyst forecast errors and the associated drift is significantly reduced after the implementation of SFAS 131. When we replicate our tests using seasonally differenced earnings, we fail to observe patterns similar to those using analyst forecast errors. Overall, our findings point to the usefulness of accounting information to assist capital markets in pricing earnings of internationally diversified firms.

Valuation Insight: Kang, Kurana, and Wang study to what extent international diversification of firms makes it more difficult for analysts to measure value. Analysts are found to underreact to earnings information of internationally diversified firms so that stock prices continue to drift for weeks after earnings are announced. However, after implementation of SFAS 131 (requiring increased geographic segment reporting) in 1997, analyst valuations have become more efficient.