Aktiemarkedets effektivitet og forventninger omkring årsregnskabsmeddelelser

This paper uses the traditional event study method to examine the efficiency of the Danish stock market with respect to annual report announcements. In addition it examines the association between unexpected earnings per share (eps) and abnormal returns, and how the use of different procedures to handle »thin« stock markets affects the results. Using data from 1999-2001, we find that annual report announcements do in fact contain information that is relevant for the stock market. The results, however, do not indicate an efficient market. The market reacts slowly to the information, and does not have correct average expectations. These results are shown to be robust to the existence of other stock exchange announcements surrounding the annual report announcement. The findings indicate that the market has become less efficient than it was in the period 1993-1996 based on a similar study conducted by Lønroth et al. (2000). Our study also shows that the expected association between abnormal returns and unexpected eps, measured as realized eps minus analysts' forecasts, does not exist. This not only contradicts the results of Lønroth et al. (2000), but also those of many American studies in the earnings literature. Finally, it is found that the results greatly depend on what selection criteria are placed on data with respect to trading frequency, and it is only when these selection criteria are minimal, that the use of different procedures has an impact on the results.