Streaks in Earnings Surprises and the Cross-section of Stock Returns

Management Science, Forthcoming

34 Pages Posted: 16 Feb 2009 Last revised: 14 Nov 2013

Roger Loh

Singapore Management University - Lee Kong Chian School of Business

Mitch Warachka

University of San Diego; Claremont McKenna College - Robert Day School of Economics and Finance

Date Written: October 27, 2011

Abstract

The gambler's fallacy (Rabin, 2002) predicts that trends bias investor expectations. Consistent with this prediction, we find that investors underreact to streaks of consecutive earnings surprises with the same sign. When the most recent earnings surprise extends a streak, post-earnings announcement drift is strong and significant. In contrast, the drift is negligible following the termination of a streak. Indeed, streaks explain about half of the post-earnings announcement drift in our sample. Our results are robust to more general definitions of trends than streaks and a battery of control variables including the magnitude of earnings surprises and their autocorrelation. Overall, post-earnings announcement drift has a significant time-series component that is consistent with the gambler's fallacy.

Keywords: Trends, Streaks, Gambler's Fallacy, Post-Earnings Announcement Drift

JEL Classification: G11, G12, G14

Suggested Citation

Loh, Roger and Warachka, Mitch, Streaks in Earnings Surprises and the Cross-section of Stock Returns (October 27, 2011). Management Science, Forthcoming. Available at SSRN: https://ssrn.com/abstract=1343451 or http://dx.doi.org/10.2139/ssrn.1343451

Roger Loh (Contact Author)

Singapore Management University - Lee Kong Chian School of Business ( email )

Lee Kong Chian School of Business
50 Stamford Rd
Singapore, 178899
Singapore

HOME PAGE: http://www.mysmu.edu/faculty/rogerloh/

Mitch Warachka

Claremont McKenna College - Robert Day School of Economics and Finance ( email )

500 E. Ninth St.
Claremont, CA 91711-6420
United States

University of San Diego ( email )

5998 Alcala Park
University of San Diego
San Diego, CA 92110-2492
United States

Paper statistics

Downloads
510
Rank
43,445
Abstract Views
2,314