The Information Content of Earnings Announcements in Denmark

Abstract

This paper uses the traditional event study method to examine the information content of annual Danish earnings announcements. Using data from 1999-2001 we find abnormal volatility in the days surrounding the earnings announcements, indicating that the announcements do in fact contain information that is relevant for the stock market. The results do not indicate that the Danish stock market is efficient. The abnormal volatility persists for up to 4 days after the announcement, while significant positive abnormal returns accompany the announcements. This conclusion is shown to be robust to the use of different trading frequency restrictions and existence of concurrent disclosures. Surprisingly, we find a positive relationship between abnormal returns and firm size. This contradicts the findings of previous studies, and indicates that small stock markets behave differently on this aspect. Confirming the results of similar studies in the US, we find that unexpected earnings are best proxied using a model based on consensus analyst forecasts. Finally, while the choice of return allocation procedure has little impact on the results, the treatment of data with respect to concurrent disclosures and trading frequency restrictions has an impact. This provides implications for the comparability of previous studies and methodology of future research.