Shorting activity and stock return predictability: Evidence from a mandatory disclosure shock

• Published date: 01 March 2022
• Author: Paul A. Griffin,Hyun A. Hong,Ivalina Kalcheva,Jeong‐Bon Kim
• Date: 01 March 2022
• DOI: http://doi.org/10.1111/fima.12351

DOI: 10.1111/fima.12351

ORIGINAL ARTICLE

Shorting activity and stock return predictability:

Evidence from a mandatory disclosure shock

Paul A. Griffin1Hyun A. Hong2Ivalina Kalcheva3

Jeong-Bon Kim4

1GraduateSchool of Management, University

of California, Davis,California, USA

2A. Gary Anderson GraduateSchool of

Management, University of California,

Riverside, California, USA

3College of Business, University of Texasat

San Antonio, San Antonio, Texas,USA

4Department of Accountancy, City University

of Hong Kong,Hong Kong, China

Correspondence

Jeong-BonKim, Department of Accountancy,

CityUniversity of Hong Kong, 83 TatChee

Avenue,Kowloon Tong,Hong Kong, China.

Email:bon.jeong.kim@cityu.edu.hk

Abstract

We study the effect of a mandatory improvement in pub-

lic disclosure due to the adoption of International Financial

Reporting Standards (IFRS) on the stock return predictabil-

ity of shorting activity. To assess the impact of the disclo-

sure shock, we measure monthly changes in the demand

for and supply of stocks for shorting and whether those

changes predict negative returns in the following month. We

provide international evidence that the ability of increases

in shorting demand and supply to predict negative returns

declines after the shock. The predictive ability of shorting in

the month before a negative earnings surprise and news of a

firm’s questionable merger and acquisitions transaction also

declines after the shock. These findings imply that the shock

of the mandatory accounting change crowds out some of

short-sellers’ value-relevant information in the equity lend-

ing market. Thus, although the democratization of infor-

mation from a structured accounting change may make

sophisticated investors worse off by reducing their ability

to predict future returns, this change may also benefit all

investors through timely stock price discovery.

KEYWORDS

equity lending market, IFRS, mandatory disclosure shock, shorting

activity, stock return predictability

JEL CLASSIFICATION

G12, G14, G30, M41

© 2021 Financial Management Association International

Financial Management. 2022;51:27–71. wileyonlinelibrary.com/journal/fima 27

28 GRIFFIN ET AL.

1INTRODUCTION

Prior empirical studies find that sophisticated investors such as short-sellers use information from different sources

to predict future equity returns. By accessing and analyzing this information, they contribute to price discovery in

the stock market (Cohen et al., 2007,2009; Desai et al., 2002; Diether et al., 2009; Duong et al., 2017; Engelberg

et al., 2012). Missing from this literature, however, is whether and how an improvementin firms’ mandatory disclo-

sure affects the predictive ability of shorting activities. On the one hand, an improvement in mandatory information

flow could be advantageous to short-sellers. They would have more and better publicly available information than

before, which could improve their ability to predict future stock returns. On the other hand, an improvement in firms’

public disclosure from a mandatory shock could be disadvantageous to sophisticated investors such as short-sellers.

An increase in publicly available information flow could reduce their advantage in predicting future returns by accel-

erating price discoveryfor all investors. It is important to recognize which of these two views prevails for understand-

ing financial market behavior in response to the disclosure shock. We accomplish this by investigating whether the

democratization of information through a mandatory accounting change benefits the majority of investors through

accelerated price discovery (arguably an intended consequence of the shock) or a small group of investors such as

short-sellers by enhancing their ability to predict future stock returns (potentially an unintended consequence).

Toaddress this issue, we take advantage of the change in structured accounting-based publicly available informa-

tion caused by the mandatory adoption of International Financial Reporting Standards (IFRS) in 2005. We investigate

the response of short-sellers to this outside disclosure shock. IFRS required thousands of publicly listed firms in more

than 100 countries to adopt a common set of high-quality accounting standards, all beginning in the same year.How-

ever,not all countries adopted the standard, and not all firms within an adopting country were required to implement

the change. Thus, from a design standpoint, we can view some firms or countries as “treated” to the adoption of high-

quality accounting standards as of 2005 and others as not. Following others (Christensenet al., 2013,2016), this fea-

ture allows us to use a difference-in-differences (DiD) research design to study how the mandatory disclosure shock

affects shorting activities and stock return predictability.We first use theory to explain the ability of shifts in shorting

demand and supply to predict positive or negative stock returns. We then compare the predictive ability of short-

sellers for firms that adopt IFRS (treated firms) with those that do not adopt IFRS (benchmark firms) from a preperiod

before adoption to a postperiod after adoption. This design is not without limitations, however.In a DiD specification,

the assignment of an observation to a treatment group is not random.In this study, some firms or countries could have

“treated” themselves to IFRS because they had more to gain from its adoption, and other firms or countries could have

improved their disclosure without adopting IFRS.1

Our tests build on the theory that outward (inward) shifts in shorting demand and shorting supply represent chan-

nels of information flow useful to predict negative (positive) future returns (Cohenet al., 2007,2009). Outward shifts

in shorting demand capture either an increase in informed trading or additional market frictions and risks associated

with shorting. By contrast, outward shorting supply shifts indicate a loosening of short-sale constraints. Both indi-

cators predict negative future returns. Although the empirical evidence indicates that an outward shift in shorting

demand is the primary channel through which shorting activities are predictive of future negative returns (Cohen

et al., 2007; Diether et al., 2009), outward supply shifts can also predict future negative returns because investors

have access to more shortable shares at a lower cost. Thus, investors can correct anyequity pricing inefficiency (e.g.,

an optimistically biased price) attributable to short sales constraintsin the previous period.

Despite considerable research on the effects of IFRS on stock returns (e.g., Barth et al., 2012,2014; Daske et al.,

2008; Hong, 2013; Hong et al., 2014; Landsman et al., 2012), prior studies do not address whether and how a manda-

tory disclosure shock affects sophisticated investors in the equity lending market. Moreover,this effect is not pre-

dictable from the prior work because an improvement in public disclosure could substitute, complement, or have no

1This issue creates an endogeneity concern for a DiD design when some firms’ or countries’ anticipatory behavior regarding IFRS adoption impacts the

outcomevariable being analyzed (Lechner, 2011).

GRIFFIN ET AL.29

effecton the information short-sellers use to predict future stock returns. Intheory, new publicly available information

can substitute (Boot & Thakor,2001; Diamond, 1985; Diamond & Verrecchia, 1991; Verrecchia, 1982) or complement

the information held by sophisticated investors for trading purposes (Boot & Thakor,2001; Bushman et al., 1996;Di

Maggio & Pagano, 2017; Fischer & Verrecchia, 1999; Indjejikian, 1991; Kim, & Verrecchia,1994,1997).2Consistent

with this theory, we consider three scenarios for how a change in mandatory disclosure could influence the informa-

tion in equity lending markets.

First, a positive shock to the public disclosure could improve information flow due to increased reporting quality,

improved accounting comparability,or reduced information-gathering costs (DeFond et al., 2011;Kim&Shi,2012),

thus reducing the value of short-sellers’ information for predicting future stock returns. Theoretical models show that

high-quality disclosure improves risk-sharing and can lower incentives for sophisticated investors to acquire addi-

tional value-relevant information. High-quality disclosure may,thus, benefit all investors in a general equilibrium (Dia-

mond, 1985; Diamond & Verrecchia, 1991; Verrecchia,1982). Further, Boot and Thakor (2001) show that the incen-

tives of investors to collect information about the firm depend on thekind of publicly available information being dis-

closed. If the information disclosed by the firm is information sophisticated investors havealready, then the informa-

tional advantage of sophisticated over less-sophisticated investors narrows. Thus, if the disclosure shock substitutes

for the information that short-sellers already have and use for trading,the advantage of short-sellers to predict nega-

tive future stock returns and the profitability of shorting should decrease in the postshock period. We denote this as a

crowding-out effect.

The second scenario posits that short-sellers’ ability to predict negative future stock returns increases from the

pre- to the postshock period.3Theoretical models of public disclosure that consider multiple private signals show

that the release of a public signal can make the allocation of private signals more concentrated in equilibrium. Thus,

a small set of traders can hold more than one private signal, whereas the rest remain uninformed (Lundholm, 1991).

Boot and Thakor (2001) explain that complementary public disclosure can strengthen sophisticated investors’incen-

tivesto acquire additional information. Additional disclosure may also attractinvestors with superior information pro-

cessing ability such as short-sellers (Di Maggio & Pagano, 2017). Engelberg et al. (2012) document empirically that

short-sellers’ trading advantage can come from their superior ability to analyze news announcements, although this is

unstructuredsoft information, different from a widely adopted accounting change. Hence, rather than impeding short-

sellers’ profitability, an improved accounting-based disclosure attributable to a disclosure shock could advantage

short-sellers by converting higher-quality disclosure into value-relevant information for shorting purposes. Thus, the

adoptionof IFRS could increase the profitability of short-sellers’ information-based trading strategies in the postshock

period (Kim & Verrecchia, 1994,1997; Loureiro & Taboada, 2015). We denote this as an information-enhancement

effect.

Third, it is also possible that an improvementin disclosure does not affect short-sellers’ information flow. This could

be due to the relative unimportance of the disclosure shock for shorting decisions compared to other information

that short-sellers use. Short-sellers’ other information could be more relevant and timelier than financial reporting

informationand could also be produced for nonfinancial-reporting reasons. In this case, the disclosure shock would not

change short-sellers’ ability to predict negative future stock returns. In light of the different predictions of the three

2Theempirical evidence has not resolved these divergent views, possibly because the amount of firm-level public information exceeding what is required by

lawis endogenous to individual firms (Burns & Kedia, 2006,2008; Kedia & Philippon, 2009). Some empirical studies assume that public disclosure substitutes

for other information useful for sophisticated investors(Barth et al., 2013; Botosan, 1997; Botosan & Plumlee, 2002; Francis et al., 2005;Hail&Leuz,2006;

Leuz& Wysocki, 2016). Others show that public disclosure complements investors’ mix of public and other useful information by documenting an increase in

informationasymmetry at the time of an earnings announcement (Barron et al., 2002,2005; Coller & Yohn, 1997; Krinsky & Lee, 1996; Lee et al., 1993).

3Afurther argument supporting the second scenario relates to the principles-based nature of IFRS, which gives firms greater flexibility in making accounting

choices. Some scholars contend that the emphasis on fair value under IFRS provides managers with greater opportunity to engage in earnings management

(De George et al., 2016). The principles-based standard may allow some firms to optimize their accounting policies to local political and economic consid-

erations.According to Sunder (2007, p. 9), a “Cartesian top-down design” for uniform accounting standards may result in suboptimal financial reporting. By

exacerbating information asymmetries, the top-down design of IFRS may offer more profitable arbitrageopportunities for short-sellers (Ball, 2006;Byard

etal., 2011).