Anomalies and News

• Author: R. DAVID MCLEAN,JOSEPH ENGELBERG,JEFFREY PONTIFF
• Date: 01 October 2018
• Published date: 01 October 2018
• DOI: http://doi.org/10.1111/jofi.12718

THE JOURNAL OF FINANCE •VOL. LXXIII, NO. 5 •OCTOBER 2018

Anomalies and News

JOSEPH ENGELBERG, R. DAVID MCLEAN, and JEFFREY PONTIFF∗

ABSTRACT

Using a sample of 97 stock return anomalies, we find that anomaly returns are 50%

higher on corporate news days and six times higher on earnings announcement days.

These results could be explained by dynamic risk, mispricing due to biased expecta-

tions, or data mining. We develop and conduct several unique tests to differentiate

between these three explanations. Our results are most consistent with the idea that

anomaly returns are driven by biased expectations, which are at least partly corrected

upon news arrival.

ACADEMIC RESEARCH SHOWS THAT a large number of observable firm character-

istics can predict the cross-section of stock returns (see Fama (1998), Nagel

(2013), and McLean and Pontiff (2016)). This “anomalies” research goes back

to at least Ball and Brown (1968) and Blume and Husic (1973), yet more than

four decades later, academics still disagree on what causes this return pre-

dictability.

There are three popular explanations for cross-sectional predictability.First,

return predictability could be a result of cross-sectional differences in risk,

which are reflected in discount rates (e.g., Fama (1991,1998)). Under this ex-

planation, cross-sectional return predictability is expected because return dif-

ferences reflect ex ante differences in the discount rates used to value the stocks.

Second, return predictability could reflect mispricing (e.g., Barberis and

Thaler (2003)). For example, the marginal investor may have biased ex-

pectations about cash flows, and the anomaly variables are correlated with

∗Joseph Engelberg is at University of California at San Diego. R. David McLean is at George-

town. Jeffrey Pontiff is at Boston College. Engelberg and Pontiff have no conflicts of interest to

disclose. McLean acknowledges receiving financial support in the form of a grant from the Social

Sciences and Humanities Council of Canada (SSHRC). We thank Pierluigi Balduzzi, Sohnke Bar-

tram, Mark Bradshaw, Gene Fama, Juhani Linnainmaa, Yan Liu, Fabio Moneta, Peter Nyberg,

Ken Singleton (Editor), David Solomon, Philip Strahan, Tuomo Vuolteenaho,and Bohui Zhang for

helpful comments as well as seminar participants at the AFA meetings, Acadian Asset Manage-

ment, Arizona State, the Asian Bureau of Finance and Economic Research, Arrowstreet Capital,

Auburn, Berkeley,Boston College, Claremont McKenna, Cornell, Cubist Systematic Strategies, De-

Paul, ESCP,Georgetown, Michigan State, Northeastern, SFS Cavalcade, Symposium on Intelligent

Investing, Temple, the TinbergenInstitute, UC Irvine, UCLA, UC Riverside, UC Santa Cruz, Uni-

versity of Central Florida, University of Miami Behavioral Finance Conference, University of South

Carolina, University of South Florida, UT Smoky Mountain Conference, University of Toronto,

University of Utah, the University of Washington, and Yale. We thank Lauren Vollonfor excellent

research assistance.

DOI: 10.1111/jofi.12718

1971

1972 The Journal of Finance R

these mistakes across stocks. Under this view, when new information arrives,

investors update their beliefs, which in turn corrects prices and leads to return

predictability.

Third, return predictability could be due to data mining. As Fama (1998)

points out, academics have likely tested thousands of variables, in which case

it would not be surprising to find that some of them predict returns in sample,

even if in reality none of them do.1

In this paper, we differentiate between these three explanations of cross-

sectional return predictability by comparing predictability on days when firm-

specific information is publicly released to days when we do not observe news.

We use the 97 anomaly variables studied in McLean and Pontiff (2016), as each

of these variables has been reported to predict the cross-section of stock returns

in a published academic study.Days with firm-specific information releases are

defined as earnings announcements days or days with a Dow Jones news item.

We find that anomaly returns are 50% higher on corporate news days and

six times higher on earnings announcement days.2We find similar effects on

both the long and short sides, that is, anomaly-shorts have lower returns and

anomaly-longs have higher returns on information days. These effects appear

to be related to firm-specific news, as anomaly returns are not higher on days

with macroeconomic news. The findings are also not explained by a day-of-the

week effect or by extreme returns causing news, as anomaly returns are not

elevated on extreme return days that do not also have news. We discuss how

our results relate to each of the three explanations of cross-sectional return

predictability below.

A. Systematic Risk

A standard, static risk-factor model (e.g., Fama and French (1993), and

Carhart (1997)) constructs a stock’s expected return as a product of its system-

atic risk exposures (“betas”) and their corresponding risk premiums. In these

factor models, a firm-specific news event does not change a stock’s expected

return because it is unrelated to the time-invariant betas and risk premiums.

Our finding of predictably higher anomaly returns on information days is at

odds with static risk-factor models.

However, our results could be consistent with dynamic-risk models, which

allow for time-varying risk premia and time-varying betas. Papers in this spirit

include Patton and Verardo (2012), who find that a stock’s beta with respect

to the market portfolio is higher on earnings announcement days and explain

1Recognition of a “multiple testing bias” is stressed more recently in the finance literature by

Harvey, Lin, and Zhu (2016), McLean and Pontiff (2016), and Linnainmaa and Roberts (2018).

2Stock returns are unconditionally higher on earnings announcement days (Lamont and

Frazzini (2007)). Savor and Wilson (2016) attempt to explain this fact. This is not the effect that

we document nor the one we want to explain; our main specifications control for this effect through

the use of earnings announcement dummy variables. We find that anomaly-long (anomaly-short)

returns are higher (lower) on earnings and news days while controlling for the fact that stock

returns are higher on earnings announcement days.