Earnings quality and short selling: Evidence from real earnings management in the United States

• Published date: 01 October 2017
• Date: 01 October 2017
• DOI: http://doi.org/10.1111/jbfa.12264

DOI: 10.1111/jbfa.12264

Earnings quality and short selling: Evidence from

real earnings management in the United States

KoEun Park

Collegeof Management, University of Mas-

sachusettsBoston

Correspondence

KoEunPark, Department of Accounting and

Finance,College of Management, University

ofMassachusetts Boston, 100 Morrissey Blvd.,

Boston,MA 02125, United States.

Email:KoEun.Park@umb.edu

Abstract

Prior research provides evidence consistent with managers using

real earnings management (REM) to increase earnings. This study

examines whether short sellers exploit the overvaluation of firms

employing REM. I find that firms with more REM have higher subse-

quent short interest. The positive relation between REM and short

interest is more pronounced in settings where the costs associated

with accrual-based earnings management are high, such as when

a firm has low accounting flexibility or faces greater scrutiny from

a high quality auditor. I also find some evidence that short sellers

respondto REM more than to other fundamental signals of firm over-

valuation. My inferences are robust to the use of propensity score

matching. Collectively, my evidence suggests that short sellers not

only trade on REM information, but they also trade as if they under-

stand the substitutive nature of alternative earnings management

methods. This study provides additional insight into the important

role that short sellers play in monitoring managerial operating deci-

sions and overallearnings quality.

KEYWORDS

corporate governance, earnings quality, real earnings management,

short selling

1INTRODUCTION

Shortsellers have been viewed in the academic literature as well-informed and sophisticated investors (e.g., Diamond &

Verrecchia,1987).1In particular, several studies provideevidence that short sellers identify overvalued firms that have

engaged in accruals management.2Yetmanagers also engage in real earnings management (REM), a practice that has

become prevalent and that brings more severeconsequences than accrual-based earnings management (e.g., Cohen &

Zarowin, 2010; Cohen, Dey,& Lys, 2008). Since REMʼs negative implications for future performance tend to be incor-

porated into stock prices with a delay,3short sellers likelyhave incentives to target firms engaging in REM. Despite

1Although the media and some regulators claim that short selling causes an unwarranted downward spiralin stock price (e.g., Karpoff & Lou, 2010; Drake,

Myers,Myers, & Stuart, 2015), these claims are largely based on anecdotal evidence.

2See,for example, Desai et al. (2006), Karpoff and Lou (2010), and Hirshleifer et al. (2011).

3See,for example, Li (2012).

1214 c

2017 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/jbfa JBus Fin Acc. 2017;44:1214–1240.

PARK 1215

the substitutive relation between accrual-based earnings management and REM (e.g.,Zang, 2012), we know very little

aboutwhether short sellers also monitor REM or ignore it, thereby potentially encouraging managers to switch to REM.

My goal is to provide evidenceof the level of sophistication in short sellersʼresponse to managersʼabnormal operating

decisions. More specifically,I examine whether and how short interest is related to a firmʼs REM as compared to other

fundamental signals identified by prior studies. Doing so provides a more complete picture of short sellersʼmonitoring

of the overallquality of corporate earnings.

Short selling refers to the sale of a stock by an investor who does not own it but borrows it from other investors

in anticipation of profiting from a price decline. Given the high costs associated with short positions, short sellers

have strong incentives to identify overpriced firms, thereby facilitating the incorporation of unfavorable informa-

tion into market prices. A number of studies suggest that short sellers are sophisticated investors by providing evi-

dence that their positions predict future returns (e.g., Asquith, Pathak, & Ritter,2005; Desai, Ramesh, Thiagarajan, &

Balachandran, 2002) and that short-selling constraints result in prices that do not fully reflect negative information

(e.g.,Boehme, Danielsen, & Sorescu, 2006; Jones & Lamont, 2002). Short sellers tend to target firms that are overpriced

relative to fundamentals (e.g., Curtis & Fargher,2014; Dechow, Hutton, Meulbroek, & Sloan, 2001). In particular, sev-

eral studies examinewhether short sellers consider overpricing associated with poor earnings quality in their decision

process by focusing on accruals. For example, Hirshleifer,Teoh, and Yu (2011) provideevidence that short interest is

positively associated with accruals, suggesting that short sellers exploitthe accrual anomaly.

Earnings quality has been identified as a critical element in capital markets. Prior research suggests that managers

manipulate reported earnings by changing accounting methods or estimates used to represent their operating activi-

ties.4However,short-term-focused financial reporting behavior is not limited to manipulating accounting practices. To

artificially boost short-term reported earnings, managers can change the timing or structuring of real operations. This

practice, also known as REM, has received considerable attention (e.g., Cohen & Zarowin, 2010; Cohen et al., 2008;

Roychowdhury,2006; Zang, 2012). For example, managers may cut prices or extendmore lenient credit terms to accel-

eratesales into the current period. They may overproduce to decrease cost of goods sold (COGS) to meet their earnings

targets in the current period. Given the different nature between accrual-based earnings management and REM, prior

evidence of short sellersʼtradingon accruals information does not necessarily provide an answer about whether short

sellers also monitor and respond to managersʼabnormal operating decisions.

The focus of this study is to explore whether short sellersʼtrading is related to a firmʼs REM. I argue that direct

investigation into the relation between REM and short interest warrants further research for three major reasons.

First, unlike accrual-based earnings management, REM has negative implications for future cash flows and firm value

because these practices alter real operations (e.g., Cohen & Zarowin, 2010; Cohen et al., 2008; Ewert & Wagenhofer,

2005; Leggett, Parsons, & Reitenga,2009; Mizik, 2010; Mizik & Jacobson, 2008). Second, while investors can be aided

by audit reports to discover accrual-based earnings management, it is more difficult for the averageinvestor to detect

REM(e.g., Kim & Sohn, 2013). If investors fixate on firmsʼreported earnings while failing to fully respond to the negative

impact of REM activities, firms with high levels of REM will be overvalued(e.g., Li, 2012). The overpricing can motivate

short sellers to profit from subsequent stock price declines of high REM firms. Third, Zang (2012) finds that managers

trade off between accrual-based earnings management and REM. Managers facing short sellersʼscrutiny on accruals

management might switch to REM if short sellers do not respond to REM. Investigating the relation between REM

and subsequent short interest conditioning on a firmʼs ability to engage in accrual-based earnings management is an

approachthat better reflects short sellersʼsophistication and monitoring role with regard to overall corporate earnings

quality.

Following prior studies, I use monthly short interest (e.g., Chi, Pincus, & Teoh, 2014; Curtis & Fargher, 2014;

Hirshleifer et al., 2011) and examine real activities management that increases earnings by offering price dis-

counts, overproducing, and cutting discretionary expenses(e.g., Roychowdhury, 2006; Zang, 2012).5My baseline tests

4This practiceis known as accrual-based earnings management. Dechow, Ge, and Schrand (2010), Fields, Lys, and Vincent (2001), Healy and Wahlen (1999)

andKothari (2001), among others, provide a survey of the literature on accrual-based earnings management.

5Idefine REM as deviations in real activities from normal business practices for the primary purpose of inflating short-term earnings (Roychowdhury, 2006).