A new perspective on post‐earnings‐announcement‐drift: Using a relative drift measure

• Date: 01 October 2019
• Author: Kevin Ow Yong,Joonho Lee,Michael Clement
• DOI: http://doi.org/10.1111/jbfa.12401
• Published date: 01 October 2019

DOI: 10.1111/jbfa.12401

A new perspective on

post-earnings-announcement-drift:

Using a relative drift measure

Michael Clement1Joonho Lee2Kevin Ow Yong3

1McCombs School of Business, University of

Texasat Austin, Austin, TX, USA

2College of Business Administration,California

State PolytechnicUniversity, Pomona, CA, USA

3Singapore Institute of Technology,Singapore

Correspondence

KevinOw Yong, Singapore Institute of Technol-

ogy,10 Dover Drive, S138683, Singapore.

Email:kevin.owyong@singaporetech.edu.sg

Abstract

Prior research finds that there is a delayed reaction to both analyst-

basedearnings surprises and random-walk-based earnings surprises.

Focusing on the market reaction from the post-announcement win-

dow, prior studies show that analyst-based drift is larger than

random walk-based drift. This finding is counter-intuitive if we

believe large, sophisticated investors tend to trade on analysts’

forecast earnings news and thus react faster and more completely

than smaller and less sophisticated investors react to random walk

earnings news. In this study, we construct a relative measure of

post-earnings-announcement drift (PEAD) (i.e., drift as a propor-

tion of total market reaction to earnings news) which we refer to

as the ‘drift ratio’, and we provide evidence, consistent with our

intuition, that analyst-based drift ratio is smaller (not greater)

than random-walk-based drift ratio. We find that this difference is

more pronounced in more recent periods and for firms with more

sophisticated investors. Our approach to measure the PEAD is more

intuitive than that in traditional PEAD literature. Our results thus

complement existing research findings by utilizing the drift ratio

measure to generate new insights about the drift phenomenon.

KEYWORDS

analyst-based earnings surprise, anomalies, information uncer-

tainty, post-earnings-announcement drift, random-walk-based

earnings surprise

JEL CLASSIFICATION

D82, G14, M41

1INTRODUCTION

Post-earnings-announcement drift (PEAD) is the tendency for a stock’s cumulative abnormal returns (CAR)to drift in

the direction of an earnings surprise following an earnings announcement. In PEAD studies, earnings surprises can be

J Bus Fin Acc. 2019;46:1123–1143. wileyonlinelibrary.com/journal/jbfa c

2019 John Wiley & Sons Ltd 1123

1124 CLEMENT ET AL.

estimated assuming investors’ earnings expectations were formed from analysts’ forecasts (hereafter ‘AFsurprises’

give rise to ‘AFdrift’) or from random-walk-based earnings forecasts (hereafter ‘RW surprises’ give rise to ‘RW drift’).

Prior research documents that AF drift and RW drift are two distinctly different PEADs, exhibitinga markedly different

magnitude and pattern of returns (e.g.,Doyle, Lundholm, & Soliman, 2006; Livnat & Mendenhall, 2006). In recent years,

Ayers, Li, and Yeung (2011) investigated whether the two PEADs are attributable to distinct sets of investors who

use different expectation models when they react to earnings news. Specifically, they found that small traderstend

to systematically trade in the direction of the RW surprises after earnings announcements, and large traders tend to

trade in the direction of AF surprises. Their findings imply that the RW drift and AF drift are driven bydifferent groups

of investorsand that the relative level of investor sophistication of different investors matter in how they interpret RW

surprises versus AF surprises.

One potentially counter-intuitive finding in traditional PEAD literature which has not received much attention is

that AF drift is greater than RW drift. This finding is counter-intuitive if we believe large, sophisticated investorstend

to trade on AF earnings news and thus react faster and more completely than smaller and less sophisticated investors

react to RW earnings news. In other words, AF drift should be smaller than RW drift. We note that this potentially

counter-intuitive finding is due to the traditional research design for PEAD studies which focuses only on returns in

the post-announcement window (i.e., drift window).

In this paper, we use a relative measure of drift to further examinePEAD. We define our drift measure, which we

refer to as the ‘drift ratio’, as the market returns in the drift window divided by the sum of the market returns both in

the earnings announcement window and in the drift window.1An advantage of this approach is that it helps to con-

trol for the size of the total price movement to the earnings announcement. By using the drift ratio, we are able to

examine the properties of the two drifts differently than previous studies because the relativedrift measure captures

investors’under-reaction relative to the total market response to the earnings announcements. We believe that the rel-

ative drift measure provides new insights into the PEAD phenomenon. For example,we are able to examine different

groups of investors’responses in terms of their drift ratios. If one group of investors is more sophisticated than another

group, we can reasonably expect that these investors’ trading patterns differ not only in the drift window but also in

the announcement window. Consequently,they should exhibit different drift ratios.

Consistent with prior research, we find that AF drift is greater than RW drift when using traditional drift measures

to measure the magnitude of PEAD. We also find that, in the 3-day window surrounding earnings announcements,

the market reaction to AF surprises is greater than the reaction to RW surprises. Using the relative drift measure, we

find that the AF drift ratio is smaller than the RW drift ratio, implying a relatively less delayed investorreaction to an

AF earnings surprise. This is consistent with the notion that more sophisticated investors are primarily trading on AF

surprises and respond faster to these earnings surprises than less sophisticated investors respond to RWsurprises.

Wealso conduct analyses to examine the time-series variation in both RW and AF drifts identified by studies such as

Ayerset al. (2011). If more sophisticated investors are primarily trading on AF surprises, their trading patterns toward

AF surprises should improve relatively faster overtime than less sophisticated investors’ trading patterns toward RW

surprises. The results in this study confirm our intuition using the relative drift measure. Specifically,we find evidence

that investors’immediate reactions during the earnings announcement window have increased monotonically over the

years for both AF and RW surprises, whereas investors’ delayed reactions to both AF and RW surprises havesignifi-

cantly declined in recent years. We find that these changes are more significant for AF surprises than RW surprises.

Specifically,we find that the AF drift ratio has decreased by more than half over our sample period, while the RW drift

ratiohas not changed significantly overthe same period. The mean AF drift ratio between 1984–1992 is 0.55, whereas

1Forexample, a drift ratio of 0.50 implies that 50% of the total market response to the earnings surprise is realized in the drift window. Hence, a smallerdrift

ratioimplies a relatively less delayed investor reaction. At the extreme, a drift ratio of 1 implies that the market reacts solely to firms’ earnings surprises in the

driftwindow which we define as the period from two days after the earnings announcement through one day after the next quarterly earnings announcement.

Conversely,a drift ratio of 0 implies that the market fully reacts to the earnings surprise during the 3-day announcement window which surrounds earnings

announcementday.