Earnings announcement timing, uncertainty, and volatility risk premiums

• Author: Tom Adams,Thaddeus Neururer
• Published date: 01 October 2020
• Date: 01 October 2020
• DOI: http://doi.org/10.1002/fut.22150

J Futures Markets. 2020;40:1603–1630. wileyonlinelibrary.com/journal/fut © 2020 Wiley Periodicals LLC

|

1603

Received: 7 May 2020

|

Accepted: 1 July 2020

DOI: 10.1002/fut.22150

RESEARCH ARTICLE

Earnings announcement timing, uncertainty,

and volatility risk premiums

Tom Adams

1

|Thaddeus Neururer

2

1

School of Business, La Salle University,

Philadelphia, Pennsylvania

2

College of Business Administration,

University of Akron, Akron, Ohio

Correspondence

Thaddeus Neururer, College of Business

Administration, University of Akron, 259

South Broadway Street, Akron, OH 44325.

Email: tneururer@uakron.edu

Abstract

We examine the relationship between firms’quarterly earnings report timing

and uncertainty before quarterly earnings announcements. Prior research

provides conflicting predictions on how investor uncertainty and report timing

are related. Using implied volatilities from equity options and the realized

returns to straddle positions, we find evidence that uncertainty and volatility

risk premiums are higher for firms that report later in the quarter. Further tests

show that the increase in option premiums is unexplained by risk factors

suggesting a mispricing by investors. These results are not associated with

static firm‐level factors and our findings are concentrated in high growth firms.

KEYWORDS

announcement timing, earnings announcements, options, uncertainty, volatility risk premiums

JEL CLASSIFICATION

G13; M40

1|INTRODUCTION

We investigate the level of ex ante uncertainty and the returns to volatility trading around firms’earnings an-

nouncement dates (EADs) as a function of the timing of firms’quarterly earnings announcements. We further in-

vestigate how the options market differentially prices volatility risk around EADs when firms are either scheduled to

report their quarterly results (i.e., their 10‐Q values) versus their annual results (i.e., their 10‐K values). Prior literature

has found that traders embed significant risk premiums into option prices around EADs (e.g., Barth & So, 2014;

Dubinsky, Johannes, Kaeck, & Seeger, 2019). While qualitatively similar to studies examining the returns to stocks

around earnings announcements, the studies focusing on option returns are unique in that the positions in question are

not exposed directly to directional changes in the underlying equity. Instead, investors selling options (e.g., at‐the‐

money [ATM] straddles) are exposed to the risk that the stock will have a significant move, either up or down, along

with the risk of an increase in expected volatility in the post‐EAD period.

1

We develop a two‐sided hypothesis concerning the relationship between reporting timing within a fiscal quarter and

uncertainty and volatility risk premiums.

2

First, we believe that for those firms that report early in the quarter,

uncertainty and volatility risk premiums will be elevated. Prior studies examining the stock returns have shown that

1

An ATM straddle is the simultaneous buying or selling of a put and call contract with the same expiration and strike price where the strike price is

near the current stock price. We describe a straddle position in more detail later in the paper.

2

Other studies will call ‘volatility risk premiums’the similar term ‘variance risk premiums.’There are potential technical differences but, for our

purposes, we do not make the distinction between the two.

there is an earnings announcement premium for stocks that report early in the quarter. For example, Savor and Wilson

(2016) develop a model where investors cannot completely separate the common information in firms’earnings and the

firm‐specific information. Thus, investors overuse the information in firms’reported earnings to update their views of

total market earnings and, consequently, firms are exposed to a higher level of systematic risk relative to statistical tests

and controls. Their tests show that the performance of firms reporting earnings helps predict future aggregate earnings

(also see Campbell, 1993). Additionally, Savor and Wilson (2016) show that early announcers earn higher returns

around their quarterly earnings announcements compared with firms that announce their results later. Similarly,

Hann, Kim, and Zheng (2019) examine the effect of peer firms’reports on other firms’uncertainty levels. Using

implied‐volatility values extracted from equity options, Hann et al. (2019) show that, as expected, when peer firms

announce their quarterly results, uncertainty levels drop for related firms. In other words, information transfers help

resolve uncertainty for industry peers and, thus, uncertainty resolution is not limited to the announcing company.

Consequently, firms that report later in the quarter may have lower uncertainty and volatility risk premiums because

much of the information that is to be released will have already been preempted by the reports of other firms.

On the other hand, delayed reports may generate more uncertainty and higher volatility risk premiums. Late

reporting firms may have difficulties in finishing their reporting process and, consequently, the later reporting may

indicate internal control problems or other reporting issues. This effect is likely exacerbated when the firm is scheduled

to release its annual report as the auditor may be holding the process up to try to rectify any issues encountered during

the audit process. While these issues would likely be firm specific, prior literature shows heightened volatility risk

premiums for firms with higher idiosyncratic risk (e.g., J. Cao & Han, 2013). Thus, even if the potential risks that are

associated with late reporting are firm specific, investors may still demand extra protection in the options market

around the EAD.

Finally, it is also possible that the options market does not fully account for the information spillovers generated by

other peer firms. This would lead to a situation where the returns to selling options (straddles) are higher but standard

proxies for risk are unable to account for the increase in observed returns. In other words, if investors do not fully

impound the information from other firms’information releases, we would observe higher returns to short straddle

positions for firms reporting later in the quarter but implied‐volatility levels would be either unrelated or even lower for

these same firms. This situation would be suggestive of mispricing by the market for expected stock variances.

To investigate how uncertainty levels and volatility risk premiums are associated with the timing of firms’quarterly

earnings reports, we use the OptionMetrics database for the years 1996–2017 inclusive. Using the price database for

option quotes, we develop a data set that details the returns to selling and holding short‐dated delta‐hedged ATM

straddles from −1 to +1 day around the quarterly EAD. These short‐dated option prices are heavily impacted by the

uncertainty of the upcoming earnings release (e.g., Dubinsky et al., 2019) and the straddle returns provide a proxy for

the risk premiums embedded in the options contracts before the EADs.

Our results are summarized as follows. First, when using implied‐volatility levels as a proxy for uncertainty, we find

that implied volatilities are elevated before earnings announcements for those firms reporting later in the quarter. Thus,

even though firms that report later in the quarter will likely have their information partially revealed by the firms that

report earlier in the quarter, investors are still more uncertain about the information releases of late reporting firms.

This result holds in multivariate regressions when we control for multiple other proxies of uncertainty, including

earnings forecast dispersion, trading volume, and historical equity return volatility.

We next apply our tests to straddle returns around the EADs. We also find that straddle returns are elevated for

firms that report later in the quarter. This result holds even if we scale the straddle returns for ex ante variance (i.e.,

option‐implied volatility). These results also hold when we additionally control for other risk and uncertainty proxies.

Consequently, while investors may view firms that report later as riskier and, thus, require extra premiums to sell

options around their EADs, an alternative explanation is that investors consistently overestimate the amount of new

information revealed in the earnings announcements for late reporting firms. This provides a partial explanation for

why our results persistent for the straddles returns even after we control for multiple risk proxies.

We then separate the timing of the reports into annual and quarterly instances. When doing so, we find that

heightened uncertainty levels are primarily associated with annual reports, and the positive association between report

timing and uncertainty largely disappears. In other words, for implied‐volatility levels, the results show that the positive

association between later filings is due to annual reports. Once we control for that difference between the annual and

quarterly reports, we find that firms reporting later do not have conditionally higher levels of uncertainty. In com-

parison, the inclusion of the annual reporting variables does not alter our findings for the straddle returns. Instead, we

1604

|

ADAMS AND NEURURER