the information content and predisclosure information. This contradicts previous studies, and it is

interpreted as evidence of a low level of pre-announcement information. Confirming the results of
similar studies, the paper finds that unexpected earnings are best proxied using a model based on
consensus analyst forecasts.
Originality/value – This paper contributes to the existing literature by analyzing the information
content of earnings announcements in a small stock market with accounting standards that are
congruent with the International Accounting Standards.
Keywords Earnings, Financial forecasting, Denmark
Paper type Research paper
1. Introduction
One of the most compelling and intriguing research questions of our time is how
information is reflected in the price of stocks. In this paper we provide new evidence in
this area by analyzing Danish earnings announcements (EAs). First, we examine
whether the Danish stock market reacts to EAs in an efficient manner that is consistent
with the EAs containing relevant information. We then attempt to explain the market’s
reaction using the level of predisclosure information and amount of surprise contained
in the EA.
This study contributes to the existing literature in two respects. First, we analyze
the information content of EAs in a small stock market where the accounting
standards are congruent with the International Accounting Standards (IAS). The
manner in which small stock markets react to earnings announcements is interesting,
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1743-9132.htm
The author is grateful for helpful comments from Jan Bartholdy, Ken L. Bechmann, Bent Jesper
Christensen, Kasper Hansen, Johannes Raaballe, an anonymous referee, and workshop
participants at the 2004 European Accounting Association Conference and the University of
Aarhus.
IJMF
4,1
4
International Journal of Managerial
Finance
Vol. 4 No. 1, 2008
pp. 4-36

q Emerald Group Publishing Limited
1743-9132
DOI 10.1108/17439130810837366
since there are several aspects where it is likely that small and large stock markets
differ with respect to their information environment.With regard to pre-announcement
information it can be argued that the smaller size leads to a less developed market with
less investor sophistication and therefore less pre-announcement information. On the
other hand, one could argue that the smaller size leads to a more transparent market
with more pre-announcement information. Additionally, it is possible that the speed
with which the new information is incorporated into prices is affected by the size of the
stock market. Again, predictions regarding both a decrease and an increase in the
speed of adjustment can be set forth. Therefore in both cases the question of which
effect is the dominant one becomes an empirical issue.
The information content of EAs in a small stock market has previously been studied
in Kallunki (1996) in the context of the Finnish stock market. However, there are two
rather unique institutional features of the Finnish stock market that make it a less
suitable candidate for isolating the effect of stock market size on the information
content of EAs. First, as mentioned in Kallunki (1996), Finnish accounting standards
are very different from the IAS. Secondly, as also noted by Kallunki (1996),
short-selling is not possible in the Finnish stock market, and this has implications for