Capital market reactions to earnings announcements: empirical evidence on the difference in the information content of IAS-based earnings and EC-Directives-based earnings

Abstract

Listing on a foreign stock exchange and the aim to attract international investors usually forces European quoted companies to adapt information supplied in financial statements to different information needs of international investors. Because of the dominance of the American stock market, this adaptation raises especially the question whether Anglo-American-oriented accounting standards (for instance IAS — International Accounting Standards) convey a higher information content for investors than continental-Europe-oriented accounting standards (for instance EC-Directives). The study examines the information content of earnings announcements, i.e. abnormal returns resulting from un-expected earnings, for a sample of Swiss quoted companies which have changed the accounting standard used for presenting Swiss GAAP consolidated financial statements to either EC-Directives or IAS and can therefore contribute to this discussion. The results of the study suggest that IAS-based earnings announce-ments convey a statistically significant higher information content than earnings announcements based on the Swiss GAAP if a variance-approach is used. For investors in the Swiss capital market, the switch from Swiss GAAP to IAS has therefore increased the information content of financial statements. But comparing IAS-based and EC-Directives-based earnings announcements, the results suggest that for investors IAS-based earnings do not possess a statistically significant higher information content than EC-Directives-based earnings. This result has been achieved despite the fact that for Swiss financial analysts financial statements based on IAS convey a significant higher information content than financial statements based on EC-Directives. Avoiding problems in specifying a model for unexpected earnings by standardizing the mean of the abnormal returns of each event window to a positive value does not lead to a different conclusion if the variance approach is used.