Review of Quantitative Finance and Accounting, 11 (1998): 165–182
© 1998 Kluwer Academic Publishers, Boston. Manufactured in The Netherlands.
Information Asymmetry Around Earnings
TERI LOMBARDI YOHN
Assistant Professor, Georgetown School of Business, Georgetown University, Washington D.C. 20057
Abstract. This study examines bid-ask spreads to determine how the anticipation and release of earnings
announcements affect information asymmetry in the stock market. I use regression analysis and ﬁnd that bid-ask
spreads are negatively related to public information availability and positively related to earnings variability and
the market reaction to prior unexpected earnings. The results suggest that ﬁrms for which earnings is expected
to yield a relatively larger stock market reaction have greater information asymmetry than ﬁrms for which
earnings are expected to yield a smaller market reaction.
I also ﬁnd that bid-ask spreads gradually increase in the four days prior to earnings announcements, and
increase sharply the day prior to, the day of and the day after the earnings announcements. Bid-ask spreads seven
to ten days after earnings announcements are not signiﬁcantly different from bid-ask spreads seven to ten days
prior to earnings announcements.
Key words: Bid-ask spreads, information asymmetry, earnings announcements
This paper studies how earnings announcements affect information asymmetry in the
stock market. Recent theoretical research shows that the anticipation of earnings an-
nouncements increases information asymmetry in the market as traders search for private
information in an attempt to proﬁt from upcoming announcements. In this paper, I test if
information asymmetry is cross-sectionally related to the anticipated stock market reac-
tion to earnings. I argue that traders will be more likely to search for private information
about ﬁrms whose earnings are expected to yield a larger market reaction. This is because
larger market reactions to earnings will allow traders to earn higher proﬁts on their private
information. I hypothesize, therefore, that information asymmetry will differ across ﬁrms
and will correlate positively with the traders’ anticipated market reaction to earnings.
This relation should hold at any point in time as long as the potential returns to private
information collection differ across ﬁrms. The differences in information asymmetry
should, however, become particularly pronounced in anticipation of the earnings release
date (McNichols and Trueman (1994), Demski and Feltham (1994) and Daley, Hughes
and Rayburn (1995)). My empirical tests focus on this event period.
In this paper, I test if bid-ask spreads in the nine days prior to earnings announcements
are cross-sectionally related to the expected market reaction to earnings.
I include three
factors to capture the expected reaction to earnings: public information availability, earn-
ings predictability and the stock market reaction to prior unexpected earnings. I argue that
@ats-ss10/data11/kluwer/journals/requ/v11n2art3 COMPOSED: 07/02/98 8:26 am. PG.POS. 1 SESSION: 29