Menu Expand

Risk Effect versus Delayed Price Response: The Case of the Post-Earnings-Announcement Drift in Germany

Cite JOURNAL ARTICLE

Style

Burghof, H., Johannsen, M. Risk Effect versus Delayed Price Response: The Case of the Post-Earnings-Announcement Drift in Germany. Credit and Capital Markets – Kredit und Kapital, 42(1), 55-91. https://doi.org/10.3790/kuk.42.1.55
Burghof, Hans-Peter and Johannsen, Matthias "Risk Effect versus Delayed Price Response: The Case of the Post-Earnings-Announcement Drift in Germany" Credit and Capital Markets – Kredit und Kapital 42.1, 2009, 55-91. https://doi.org/10.3790/kuk.42.1.55
Burghof, Hans-Peter/Johannsen, Matthias (2009): Risk Effect versus Delayed Price Response: The Case of the Post-Earnings-Announcement Drift in Germany, in: Credit and Capital Markets – Kredit und Kapital, vol. 42, iss. 1, 55-91, [online] https://doi.org/10.3790/kuk.42.1.55

Format

Risk Effect versus Delayed Price Response: The Case of the Post-Earnings-Announcement Drift in Germany

Burghof, Hans-Peter | Johannsen, Matthias

Credit and Capital Markets – Kredit und Kapital, Vol. 42 (2009), Iss. 1 : pp. 55–91

Additional Information

Article Details

Author Details

Prof. Dr. Hans-Peter Burghof, Universität Hohenheim, Lehrstuhl für Bankwirtschaft und Finanzdienstleistung, Schloss Osthof 510 F, D-70599 Stuttgart.

Matthias Johannsen, Universität Hohenheim, Lehrstuhl für Bankwirtschaft und Finanzdienstleistung, Schloss Osthof 510 F, D-70599 Stuttgart.

Abstract

Risk Effect versus Delayed Price Response: the Case of the Post-Earnings-Announcement Drift in Germany

This paper presents supporting evidence for the post-earnings-announcement drift using annual data on 850 firms listed on the Frankfurt stock exchange for the years 1990 to 2003. Standardized unexpected earnings and unexpected earnings based on the security return model yield significant abnormal returns to the drift trading strategy of about 3% to 6% over 59 to 109 days. In an analysis of covariance the variables size and book-to-market ratio are insignificant in explaining the drift. Further, a control variable for a momentum effect is highly significant, and the inverse Mills ratios to control for a survivorship bias are significant for periods starting from about 160 days. All variables combined cause the drift to become insignificant for the standardized unexpected earnings model and reduce the significance of the drift given by the security return model. The insignificance of the variables size and book-to-market ratio is confirmed by repeating the analysis within sub-samples. The results suggest that there is a delayed response to earnings-related information on the German stock market. (JEL G14, M41)