Gao, Kaijuan; Ding, Mu an
A icle
Sho -sale e inancing and p ice adjus men speed o bad
news: E idence om a quasi-na u al expe imen in China
China Jou nal o Accoun ing Resea ch
P o ided in Coope a ion wi h:
Sun Ya -sen Uni e si y
Sugges ed Ci a ion: Gao, Kaijuan; Ding, Mu an (2019) : Sho -sale e inancing and p ice adjus men
speed o bad news: E idence om a quasi-na u al expe imen in China, China Jou nal o Accoun ing
Resea ch, ISSN 1755-3091, Else ie , Ams e dam, Vol. 12, Iss. 4, pp. 379-394,
h ps://doi.o g/10.1016/j.cja .2019.11.001
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/241805
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by-nc-nd/4.0/
Sho -sale efinancing and p ice adjus men speed o
bad news: E idence om a quasi-na u al expe imen in
China
Kaijuan Gao
a
, Mu an Ding
b,⇑
a
Hunan Uni e si y, China
b
Shanghai Uni e si y o Finance and Economics, China
ARTICLE INFO
A icle his o y:
Recei ed 15 Oc obe 2018
Accep ed 7 No embe 2019
A ailable online 5 Decembe 2019
ABSTRACT
Sho selling may accele a e s ock p ice adjus men o nega i e news. Howe e ,
he li e a u e p o ides mixed e idence o his p edic ion. Using sho -sale efi-
nancing and a s agge ed diffe ence-in-diffe ences (DID) model, his pape
explo es he effec o sho selling on s ock p ice adjus men . Ou esul s show
ha (1) sho -sale efinancing imp o es he speed o s ock p ice adjus men o
nega i e news. This esul holds a e we con ol o endogenei y. (2) The pos-
i i e ela ionship be ween sho -sale efinancing and s ock p ice adjus men
speed is significan in subsamples o s ocks wi h highe ea nings managemen
o lowe accu acy o analys o ecas s, indica ing ha fi ms wi h mo e opaque
in o ma ion a e mo e likely o be a ge ed by sho selle s. In subsamples o
s ocks wi h a highe owne ship concen a ion o lowe owne ship by ins i u-
ional in es o s, sho selling is mo e likely o inc ease he speed o s ock p ice
adjus men , indica ing ha owne ship s uc u e may influence nega i e news
mining. (3) As sho -sale efinancing exace ba es he abso p ion o bad news
by s ock p ices, i inc eases c ash isk. This s udy en iches he esea ch on
he economic consequences o sho selling and p o ides empi ical e idence
suppo ing egula ions on sho selling in China.
Ó2019 Sun Ya -sen Uni e si y. P oduc ion and hos ing by Else ie B.V. This
is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecom-
mons.o g/licenses/by-nc-nd/4.0/).
1. In oduc ion
Capi al ma ke s use efficien p icing o asse s o achie e an op imal alloca ion o esou ces. In an effec i e
capi al ma ke , he s ock p ice guides he flow o esou ces o he mos p omising p ojec s, and manage s use
h ps://doi.o g/10.1016/j.cja .2019.11.001
1755-3091/Ó2019 Sun Ya -sen Uni e si y. P oduc ion and hos ing by Else ie B.V.
This is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/).
⇑
Co esponding au ho .
E-mail add ess: [email p o ec ed]om (M. Ding).
China Jou nal o Accoun ing Resea ch 12 (2019) 379–394
HOSTED BY Con en s lis s a ailable a ScienceDi ec
China Jou nal o Accoun ing Resea ch
jou nal homepage: www.else ie .com/loca e/cja
he s ock p ice o make decisions (Du ne e al., 2004). Howe e , hese ules ely on efficien asse p icing. As
an impo an financial inno a ion p oduc , sho selling may imp o e p ice disco e y in a capi al ma ke by
acili a ing he abso p ion o nega i e news in o s ock p ices (Mille , 1977). Sho selling has exis ed in coun-
ies wi h de eloped capi al ma ke s o a long ime. The Chinese capi al ma ke has a a he sho his o y. I s
egula ions and a ionali y o i s in es o s need imp o emen , and i s condi ions o sho selling a e imma-
u e. Howe e , a e mo e han 20 yea s o de elopmen , he Chinese capi al ma ke has de eloped conside -
ably. On Ma ch 31, 2010, China officially pilo ed a sho -sale ading mechanism, hoping ha i would help
s ock p ices be e in eg a e in es o in o ma ion and be e eflec in insic alue.
P e ious s udies ha e examined he effec o sho selling on he efficiency o s ock p icing. Mos s udies
using global da a indica e ha sho selling p omo es s ock p ice abso p ion o nega i e in o ma ion and
imp o es ma ke efficiency (Diamond and Ve ecchia, 1987; Hong and S ein, 2003; B is e al., 2007;
Boehme and Wu, 2013). B is e al. (2007) sugges ha emo ing sho -sale es ic ions makes s ock p ices
mo e sensi i e o nega i e news. Al hough B is e al. (2007) con ol o coun y-specific cha ac e is ics using
c oss-lis ing and e en s udies, he selec ion o c oss-lis ing and policy implemen a ion is no andom. The
app oach in B is e al. (2007) does no comple ely esol e he endogenei y p oblem by only compa ing be o e
and a e a shock. I is no ewo hy ha he conclusion in B is e al. (2007) is based on e idence om a specific
coun y and he e o e may no be applicable o o he coun ies wi h diffe en ins i u ional en i onmen s.
China has he la ges eme ging capi al ma ke . P e ious s udies on whe he sho selling affec s p icing effi-
ciency ha e shown mixed esul s. Chang e al. (2014), Xiao and Kong (2014), and Li e al. (2015) find ha
sho selling can imp o e ma ke efficiency. Xu and Chen (2012) p o ide e idence ha sho selling does
no imp o e he speed o p ice adjus men o nega i e news using one-yea da a. Howe e , Tang e al.
(2016) ex end he sample om 2007 o 2012 and sugges ha sho selling does indeed imp o e he speed
o s ock p ice adjus men o nega i e news by in es iga ing he ela ionship be ween ma gin ading and sho
selling.
We p opose ha opposi e conclusions wi h he same ins i u ional backg ound may be ela ed o he mea-
su es o efficiency and samples used. Fi s , p e ious s udies a e mainly based on he in o ma ion con en and
ola ili y o liquidi y o s ock p ices (Chang e al., 2014; Xiao and Kong, 2014; Li e al., 2015). Howe e , sho
selling in China is diffe en om ha in o he coun ies (Gao and Lin, 2018). The China Secu i ies Regula o y
Commission pe mi s bo h sho selling and ma gin ading o only 950 pilo fi ms. Sho selling deli e s neg-
a i e in o ma ion, whe eas ma gin ading deli e s posi i e in o ma ion. The e o e, he effec o sho selling
may be compounded by ha o ma gin ading. Second, egula o s o en ake liquidi y, ola ili y, u no e
and o he indica o s as c i e ia
1
o choosing pilo fi ms o sho selling and ma gin ading. As a esul , pilo
fi ms may be diffe en om con ol fi ms (Su and Ni, 2018). Thi d, due o he lack o sho able secu i ies, sho
selling in China is a e. The e o e, i is easonable o d aw he conclusion ha sho selling does no acili a e
mo e efficien p ice disco e y in his specific capi al ma ke .
In Feb ua y 2013, a pilo p og am called efinancing was officially launched in China. In he efinancing
p ojec , China Secu i ies Finance Co po a ion Limi ed (CSFC) bo ows s ocks om lis ed companies’ sha e-
holde s and o he lende s and makes hem a ailable o secu i ies companies, which make hem a ailable o
in es o s o sho selling. The in oduc ion o his ading mechanism has acili a ed sho selling by expand-
ing he numbe o s ocks ha secu i ies companies can lend o hei clien s. Fig. 1 shows he ime end o
sho -sale efinancing olume as well as sho selling om 2010 o 2019. As can be seen in Fig. 1, a e
sho -sale efinancing was officially allowed in 2013, he olume o sho selling inc eased exponen ially. This
1
The s anda d o a sho selling a ge is defined in he ‘‘de ailed ules o implemen a ion,”so ed om la ge o small acco ding o
weigh ed e alua ion alue, and comp ehensi ely conside s he condi ions o he indi idual s ock and ma ke including he ollowing
ac o s. (1) The s ock has been on he exchange o o e 3 mon hs. (2) The fi m has no less han 100 million sha es o no less han 500
million yuan o ma ke alue o s ocks pu chased h ough ma gin ading, and no less han 200 million sha es o no less han 800 million
yuan o ma ke alue o s ocks sold h ough sho selling. (3) The numbe o sha eholde s shall no be less han 4,000. (4) None o he
ollowing si ua ions ha e occu ed in he pas h ee mon hs: (i) he a e age daily u no e was less han 15% o he a e age daily u no e
o he benchma k index, and he a e age daily u no e amoun was less han RMB50 million; (ii) he de ia ion alue be ween he a e age
daily change and he a e age daily change o he benchma k index was mo e han 4%; o (iii) he fluc ua ion ange was mo e han 5 imes
ha o he benchma k index. (5) S ock issuing companies ha e comple ed he e o m o non- adable sha es. (6) S ock ansac ions a e no
subjec o special ea men by he exchange. (7) O he condi ions as p esc ibed in he ‘‘de ailed ules o implemen a ion.”
380 K. Gao, M. Ding / China Jou nal o Accoun ing Resea ch 12 (2019) 379–394
indica es ha he launch o he sho -sale efinancing p og am indeed boos ed he g ow h o sho selling.
Du ing he ea ly pe iod a e he launch o he p og am, secu i ies companies only selec ed some sho able
secu i ies as pilo secu i ies, so he pilo sho able secu i ies and o he sho able secu i ies a e he ea men
g oup and con ol g oup, espec i ely. As a esul , he diffe ences be ween he ea men and con ol g oups
in liquidi y, ola ili y and u no e a e small. Acco dingly, his allows us o examine he eal economic con-
sequences o sho selling. Addi ionally, a e efinancing ansac ions s a ed on Augus 27, 2012, ma gin
ading has been allowed o bo h he con ol and ea men g oups. Howe e , only he ea men g oup is
a ailable o sho -sale efinancing om Feb ua y 28, 2013. The e o e, ou sample, o a ce ain ex en ,
may emo e he con ounding effec o ma gin ading.
Using a s agge ed diffe ence-in-diffe ences (DID) model and sho -sale efinancing in China, his pape
explo es whe he sho selling affec s he speed o s ock p ice adjus men o nega i e in o ma ion. The esul s
show ha sho -sale efinancing imp o es p ice adjus men o nega i e news. The esul holds a e con olling
o endogenei y, indica ing ha sho selling imp o es he speed o p ice adjus men o nega i e in o ma ion.
We also find ha he posi i e ela ionship be ween sho -sale efinancing and he speed o p ice adjus men o
nega i e in o ma ion is significan in a subsample o s ocks wi h highe ea nings managemen o lowe accu-
acy o analys o ecas s, indica ing ha fi ms wi h mo e opaque in o ma ion a e mo e likely o be ollowed by
sho selle s. In fi ms wi h highe owne ship concen a ions o lowe pe cen ages o ins i u ional in es o s,
sho selling is mo e likely o inc ease he speed o p ice adjus men o nega i e in o ma ion, indica ing ha
owne ship s uc u e may influence nega i e news mining o sho selling. In he end, we find ha as sho -sale
efinancing exace ba es he abso p ion o bad news in o s ock p ices, s ock p ice c ash isk inc eases. In es o
s uc u e may explain his esul . Sho selling in China has only been pe mi ed and p ac iced ecen ly; in
addi ion, mos in es o s a e indi iduals and ins i u ional in es o s, which exhibi he d beha io (Xu e al.,
2013). These ac o s may cause excessi e sho selling due o o e confidence in p i a e in o ma ion o due
o an excessi ely pessimis ic mood. Such sho selling beha io can cause nega i e emo ions in he ma ke
o inc ease (Scheinkman and Xiong, 2003), igge sha e p ices alling and inc ease s ock p ice c ash isk.
Ou pape con ibu es o he li e a u e in se e al ways. Fi s , using an exogenous e en , we find ha sho
selling in he Chinese capi al ma ke imp o es he speed o p ice adjus men o nega i e news. Al hough B is
e al. (2007) use na ional le el da a and sugges ha elaxing he es ic ion on sho selling makes he pilo
fi ms’ s ock p ices mo e sensi i e o nega i e news; howe e , esul s based on da a om one coun y may
no be applicable o coun ies wi h diffe en ins i u ional en i onmen s. In addi ion, p io esea ch on his
opic in he Chinese con ex has no eached a consis en conclusion (Chang e al., 2014; Xu and Chen,
2012; Tang e al., 2016).
Second, we disco e he influence o he in o ma ion en i onmen and owne ship s uc u e on he economic
consequences o sho selling, which en iches he li e a u e on sho selle s’ bad news mining based on he
cha ac e is ics o lis ed companies.
0
200000
400000
600000
800000
1000000
1200000
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019
Volume o Sho -sale Re inancing uni : sha es
Volume o Sho Selling uni : million sha es
Fig. 1. Time end o sho -sale efinancing and sho selling olume.
K. Gao, M. Ding / China Jou nal o Accoun ing Resea ch 12 (2019) 379–394 381
Thi d, we find ha sho selling in China inc eases s ock p ice c ash isk. Whe eas his conclusion is con-
sis en wi h Callen and Fang’s (2015) s udy on US-lis ed companies, i is inconsis en wi h ano he s udy using
a Chinese sample (Chu and Fang, 2016). Using sho selling olume in China, Chu and Fang (2016) find ha
al hough ma gin ading inc eases s ock p ice c ash isk, sho selling educes s ock p ice c ash isk o some
ex en . We p opose ha sho selle s choose companies wi h mo e accumula ed nega i e news o sho sell, so
he findings based on sho selling olumes may ha e endogenei y issues. This pape sol es his p oblem by
using whe he a fi m is subjec o sho -sale efinancing o s udy he effec o sho selling on s ock p ice c ash
isk.
Fou h, ou conclusions ha e implica ions o egula o s. Sho selling in China helps accele a e p ice
adjus men o nega i e news, bu i can also inc ease s ock p ice c ash isk due o sha e p ice s imula ion. Chi-
na’s policymake s should be ca e ul in egula ing sho selling ac i i ies, make igid ma ke ope a ional ules,
make ull and effec i e use o sho selling’s ad an ages and a oid excessi e ola ili y caused by sho selling.
The emainde o his pape is o ganized as ollows. In Sec ion 2, we e iew he li e a u e and gi e esea ch
hypo heses. Sec ion 3 epo s he da a sou ces, model in e p e a ion, a iable analysis and summa y s a is ics
o he sample. Sec ion 4 epo s and discusses he empi ical esul s and handling o he endogenei y p oblem.
In Sec ion 5, we u he analyze he effec o in o ma ion anspa ency and owne ship s uc u e on he sho
selling effec and explo e he economic consequences o sho -sale efinancing on c ash isk. Sec ion 6 p esen s
he conclusion.
2. Ins i u ional backg ound, li e a u e e iew and hypo heses
Ma gin ading and sho selling e e o b oke s lending money o in es o s o buy secu i ies o lending
secu i ies o in es o s o sell in an icipa ion o a p ice d op, espec i ely; he bo owe s mus deposi colla e al.
On Ma ch 31, 2010, a pilo p og am o allow ma gin ading and sho selling was officially launched in China,
allowing in es o s o bo ow unds and secu i ies om b oke s o ma gin ading and sho selling and end-
ing he ‘‘unila e al ma ke ”in which sho selling had been p ohibi ed o mo e han 20 yea s. The sho sell-
ing mechanism has become an impo an e o m o he Chinese s ock ma ke ; i has a ac ed ex ensi e
a en ion and b ough new oppo uni ies o in es o s’ ading me hods. The de elopmen o sho selling
in China is diffe en om ha in a ma u e capi al ma ke . The pilo has been expanded, s ep by s ep, om
90 fi ms a ailable o sho selling a he beginning o 950 fi ms on Decembe 12, 2016. Al hough sho selling
may b ing significan s ock p ice disco e y and ma ke efficiency, i s scale was es ic ed in he ea ly s ages due
o insufficien secu i y sou ces, limi ed capi al and a high en y h eshold. In addi ion, due o high ansac ion
cos s, ansac ion ime and place cons ain s and o he sho selling cons ain s, ma gin ading has he dom-
inan ma ke ole. Basically, he une en scale o he wo ypes o ading means ha hey ha e an asymme ic
effec on ma ke efficiency, so he e is s ill oom o de elopmen o he sho selling mechanism.
To imp o e he ma gin ading and sho selling mechanisms and b oaden he sou ces o unds and secu-
i ies a ailable o ma gin ading and sho selling, CSFC implemen ed a efinancing mechanism on Augus
27, 2012, allowing secu i ies ading h ough efinancing. On Feb ua y 28, 2013, CSFC officially implemen ed
sho -sale efinancing. Sho -sale efinancing means ha a secu i ies finance company lends i s secu i ies o
ano he secu i ies company o ha company’s secu i ies lending business. Fig. 2 shows he ime end o
he numbe o pilo secu i ies and he ansac ion olume o sho -sale efinancing. As o Decembe 12,
2016, when he numbe o pilo secu i ies expanded o 950, equal o he numbe a ailable o ma gin ading,
he olume o sho -sale efinancing achie ed apid g ow h. The la ge inc ease in sho -sale efinancing a e
2014 was due o he s a e council publishing ‘‘Opinions o he S a e Council on Fu he P omo ing he Sound
De elopmen o Capi al Ma ke s”in May 2014. Secu i ies lending and sho selling ha e achie ed leap og
de elopmen h ough he pe ec ing o he capi al ma ke sys em and op imized sho selling mechanism.
Sho -sale efinancing is conduci e o imp o ing secu i ies ma ke efficiency, mee ing in es o equi emen s
o di e sified in es men s and isk managemen , es ablishing and s eng hening he in es men concep o
he secu i ies ma ke and imp o ing secu i ies companies’ abili y o manage income, inno a ion and isk.
The g adual expansion o he numbe o pilo fi ms no only sho ens he gap be ween he ea men and con-
ol g oups in his s udy bu also a oids he possibili y o ma gin ading affec ing he esul s and hus p o ides
an ideal na u al expe imen o explo e he economic consequences o sho selling.
382 K. Gao, M. Ding / China Jou nal o Accoun ing Resea ch 12 (2019) 379–394

The sho -sale efinancing p og am has impo an p ac ical significance. Fi s , he pilo p og am inc eased
sho selling by Chinese in es o s and p o ided hem wi h mo e p ofi oppo uni ies, which is conduci e o
in es o s implemen ing di e sified in es men s a egies h ough efinancing o dispe se isk. Second, he pilo
p og am changed in es men beha io and educed isk in he Chinese secu i ies ma ke . When sho selling is
es ic ed due o capi al ma ke imma u i y and a ious sys em sho comings, inside ading and s ock
manipula ion lead o con inuous s ock p ice inc eases. When ha happens, a la ge numbe o in es o s o en
ollow he end o seek excess p ofi s. Howe e , he in oduc ion o sho -sale efinancing has ocused sho
selle s’ a en ion on o e alued s ocks and op imized in es men beha io in he ma ke . Finally, he efinanc-
ing pilo has p omo ed he e u n o secu i ies o hei in insic alue and educed he p ice bubble in he ma -
ke . S ocks wi h high sha e p ices a e mo e likely o a ac sho selle s o secu i ies lending. Sho selle s
deli e ele an in o ma ion o he s ock ma ke by sho ing secu i ies, hus inhibi ing he p ice bubble and
causing s ock p ices o e u n o hei undamen al alue, which has also been explained by p io esea ch
(Su and Ni, 2018). In p ac ice, ‘‘black swan”e en s, such as Shuanghui clenbu e ol and Yili melamine, ha e
a significan effec on s ock p ice. Black swan e en s may inc ease sho -sale efinancing, which will enhance
he p ice disco e y unc ion o he s ock ma ke . Fig. 3 shows he end o sho -sale efinancing ading in
Hebei I on and S eel om 2014 o 2017. A e 2014, he e was excess capaci y and nega i e g ow h a Hebei.
A la ge numbe o employees we e laid off, and b ibe y occu ed in 2017. This black swan e en had a signi -
ican nega i e effec on Hebei, leading o a sha p inc ease in i s sho selling ading olume in 2015 and 2017.
This example shows ha a black swan e en can indeed inc ease sho selling ading olume.
P e ious s udies on sho selling ha e ocused on he effec o sho selling on s ock p icing efficiency bu
ha e no eached a consis en conclusion. Mille (1977) was he fi s o p opose ha unde sho selling es ic-
ions, s ock p ices only eflec he a i ude o op imis ic in es o s and a e o e alued because pessimis ic in es-
o s canno sho sell he s ock. Diamond and Ve ecchia (1987) sugges ha sho selling es ic ions educe
he efficiency o s ock p icing, especially when bad news canno be eflec ed imely in s ock p ices. Howe e ,
Diamond and Ve ecchia (1987) also poin ou ha in es o s adjus s ock p ice expec a ions in conside a ion
o sho selling es ic ions, so he es ic ions do no esul in o e alued s ock p ices. This iew was sup-
po ed by Xu and Chen (2012) using ma gin ading and sho selling da a in China. They ound ha he p ice
o a ge fi ms makes no diffe ence o he con en o nega i e p i a e in o ma ion o o he speed o adjus men
o ma ke downswings. The influence on p icing efficiency is s ill limi ed mainly because he mechanism is lim-
i ed by ac o s du ing he pilo ha led o low ading olume. A e he implemen a ion o sho selling,
esea ch on sho selling g adually inc eased. P e ious s udies ha e used empi ical da a o p o e ha sho
selling es ic ions lead o he o e alua ion o s ock p ices (Chang e al., 2007;Figlewski, 1981) and e ified
0
100
200
300
400
500
600
700
800
900
1000
0
5
10
15
20
25
2013-02-28
2013-04-24
2013-06-24
2013-08-14
2013-10-15
2013-12-04
2014-01-27
2014-03-26
2014-05-21
2014-07-14
2014-09-03
2014-11-03
2014-12-24
2015-02-17
2015-04-17
2015-06-10
2015-08-03
2015-09-25
2015-11-24
2016-01-25
2016-03-23
2016-07-01
2016-08-23
2016-10-24
2016-12-14
2017-02-13
2017-04-07
2017-06-02
2017-07-25
2017-09-14
2017-11-13
No. o i ms
Volume o Sho -sale Re inancing (1,000,000)
T ansac ion Da e
T ansac ion Volume No. o Pilo Secu i ies
Fig. 2. Time end o sho -sale efinancing ansac ion olume and numbe o fi ms.
K. Gao, M. Ding / China Jou nal o Accoun ing Resea ch 12 (2019) 379–394 383
he hypo hesis o s ock p ice o e alua ion p oposed by Mille (1977).Ai ken e al. (1998), using he e en
esea ch me hod, sugges ha s ock p ices would all a e sho sales in he Aus alian ma ke . Die he
e al. (2009) also ound ha s ock e u ns dec ease a e sho sales. Desai e al. (2002) confi m he decline
o s ock p ices a e sho sales using NASDAQ da a om 1988 o 1994 and p opose ha he mo e sho sell-
ing ansac ions, he lowe he s ock’s e u n. Chang e al. (2014) find ha when lis ed companies became pilo
fi ms, hei s ock e u n became nega i e, indica ing ha sho selling es ic ions led o he o e alua ion o
s ock. The Chinese capi al ma ke is no ye ma u e, and he in oduc ion o he sho selling mechanism was
la e, so ele an esea ch has only ecen ly de eloped. Gu and Hao (2011) find ha sho selling cons ain s
lead o p icing de ia ion using a s ochas ic alua ion model and panel da a. Li e al. (2014) find ha compa ed
wi h s ocks a ailable o sho , s ocks es ic ed om sho sales showed highe yields a e he shock. Li e al.
(2015) p opose ha ma gin ading could imp o e p icing efficiency by imp o ing s ock liquidi y, educing
in o ma ion asymme y and inc easing sha eholding wid h. Some esea ch p oposes ha sho selling may
educe s ock p ice ola ili y (Hong and S ein, 2003; Xiao and Kong, 2014). Chu and Fang (2016) p opose ha
sho selling imp o es p icing efficiency bu ha he possibili y o educing ola ili y is small while he unde -
lying s ock has li le isk o c ashing. Su and Ni (2018) s udy he ela ionship be ween sho selling cons ain s
and s ock p ice changes om he pe spec i e o sho -sale efinancing and find ha a educ ion in sho selling
cons ain s can no only educe ma ke asymme y and ola ili y bu also educe ma ke s abili y and agg a-
a e ma ke slump.
Some pape s dis inguish posi i e news om nega i e news by ocusing on sho selling and s ock p icing
efficiency and s udy he adjus men speed o s ock p ice o diffe en kinds o news. Diamond and Ve ecchia
(1987) heo e ically p opose ha sho -selling cons ain s p e en nega i e news om being eflec ed in he
s ock p ice in a imely manne , hus educing he a e o in o ma ion abso p ion by he s ock p ice. B is
e al. (2007) use global da a o compu e c oss-au oco ela ions be ween lagged one-week ma ke e u ns
and indi idual s ock e u ns and confi m ha elaxing sho -selling cons ain s is conduci e o imp o ing
he speed o adjus men o bad news. Howe e , conclusions d awn a a na ional le el may no apply in
a diffe en ins i u ional en i onmen . Saffi and Sigu dsson (2011) use he co ela ion coefficien o ma ke
e u ns on lagged s ock e u ns, he delay o s ock p ice eac ion o ma ke in o ma ion, and he dis ibu-
ion o s ock e u ns as measu es o s ock p icing efficiency. They show ha sho selling can accele a e he
0
1
2
3
4
5
6
7
8
9
10
2014-08-19
2014-09-17
2014-10-22
2014-11-19
2014-12-17
2015-01-16
2015-02-13
2015-03-20
2015-04-20
2015-05-19
2015-06-16
2015-07-15
2015-08-12
2015-09-11
2015-10-16
2015-11-13
2015-12-11
2016-10-27
2016-11-24
2016-12-22
2017-01-20
2017-02-24
2017-03-24
2017-04-25
2017-05-24
2017-06-23
2017-07-21
2017-08-18
2017-09-15
2017-10-20
2017-11-17
2017-12-15
Volume o Sho -Sale Re inancing (10,000)
T ansac ion Da e
Fig. 3. Time end o sho -sale efinancing olume o Hebei i on and s eel.
384 K. Gao, M. Ding / China Jou nal o Accoun ing Resea ch 12 (2019) 379–394
speed o s ock p ice adjus men o in o ma ion, bu hey do no dis inguish be ween good and bad news,
which would affec sho selling diffe en ly. Howe e , Xu and Chen (2012) use one-yea da a o pu o wa d
ha sho sales do no imp o e he speed o s ock p ice adjus men o nega i e news, which can be
explained by he sho sample pe iod. Addi ionally, in 2011, when sho selling was in i s ini ial pe iod,
i could no play i s significan ole in such a sho pe iod. The posi i e ela ionship be ween sho selling
and speed o s ock p ice adjus men o nega i e news could no be ob ained due o he es ic ions on
unde lying s ock ype and quan i y, ansac ion cos s and ba ie s o en y. Tang e al. (2016) use an
enla ged sample ange, 2007–2012, and find ha sho selling can indeed imp o e he speed o s ock p ice
adjus men o nega i e news. Despi e using a PSM + DID me hod o a obus ness es , he pa icula i ies
o secu i ies and he influence o ma gin ading could no be emo ed, so endogenei y p oblems emain.
The e o e, i is necessa y o conside endogenei y and explo e he effec o sho selling on he speed o s ock
p ice adjus men o nega i e news.
Sho selling can affec he speed a which s ock p ices adjus o nega i e news. In sho selling ading,
in es o s who hold nega i e news expec a s ock p ice o all in he u u e, so hey pay a ma gin o a b oke
o bo ow he s ock and sell i . When he s ock p ice alls, he in es o buys he s ock and epays i o he
b oke hey bo owed i om, ea ning money om he s ock p ice alling. Sho selling es ic ions esul
in pessimis ic in es o s who canno sho sell lea ing he s ock ma ke (Mille , 1977); he e o e, nega i e news
is no eflec ed imely in s ock p ices. Fama (1965) p oposed ha in an efficien ma ke , s ock p ices eflec
his o ical and cu en in o ma ion abou a company’s alue, good o bad, and s ock p ices adjus acco ding
o ma ke in o ma ion imely and comp ehensi ely. When sho selling is es ic ed, sho selle s canno eac
o nega i e in o ma ion h ough ading, esul ing in asymme ic p ice adjus men s o posi i e and nega i e
in o ma ion and delayed p ice adjus men s o nega i e news. The ban on sho selling also educes he effi-
ciency o in o ma ion ansmission and exp ession in he ma ke (Diamond and Ve ecchia, 1987), leading
o a dec ease in he p opaga ion speed o bad news, whe eas he in oduc ion o sho selling accele a es
he speed o adjus men o nega i e in o ma ion h ough ading. Thus, allowing sho selling accele a es
in es o s’ sea ch o nega i e news and he a e a which s ock p ices abso b i .
Based on ha , we p opose he ollowing hypo hesis:
Sho selling accele a es he speed o s ock p ice adjus men o nega i e news.
3. Resea ch design
3.1. Sample cons uc ion and da a sou ces
We choose he pilo fi ms o sho selling om 2010 o 2017 as he sample. The esea ch da a mainly come
om he CSMAR da abase. Table 1 p o ides he annual adjus men s in ma gin and sho selling pilo fi ms
and he expansion and adjus men s o sho -sale efinancing. On Decembe 5, 2011; Janua y 31, 2013;
Sep embe 16, 2013; Sep embe 22, 2014; and Decembe 12, 2016, he fi ms a ailable o ma gin and sho
selling secu i ies we e expanded, ul ima ely eaching 950. On Feb ua y 28, 2013; Sep embe 16, 2013; June
20, 2014; Ap il 30, 2015; and Decembe 12, 2016, he numbe o sho -sale efinancing pilo fi ms inc eased,
s a ing a 287 and ending a 950.
Table 1
Time able o adjus men s o sho selling and sho -sale efinancing fi ms.
Yea Sho selling in Sho selling ou To al sho selling Refinancing in Refinancing ou To al efinancing
2010 96 6 90 0 0 0
2011 190 2 278 0 0 0
2012 0 0 278 0 0 0
2013 482 60 700 299 12 287
2014 218 19 899 356 18 625
2015 0 8 891 269 3 891
2016 77 18 950 77 18 950
K. Gao, M. Ding / China Jou nal o Accoun ing Resea ch 12 (2019) 379–394 385
3.2. Model and a iables
We use he me hod in B is e al. (2007) o measu e he speed o s ock p ice adjus men o nega i e news:
qdi i¼q0iq1i;ð1Þ
in which q0
i
and q1
i
a e he c oss-au oco ela ions be ween ma ke weekly e u ns lagged one week and indi-
idual s ock weekly e u ns, espec i ely, when he ma ke e u n alls o ises, eflec ing diffe en p ice delays
when he ma ke mo es in diffe en di ec ions, namely he speed o s ock p ice adjus men o nega i e news.
B is e al. (2007) poin ou ha downside c oss-au oco ela ion (q0
i
) is high when he e is a sho -selling
es ic ion, as s ock p ices canno quickly espond o he ma ke ’s downwa d fluc ua ion. In con as , upside
c oss-au oco ela ion (q1
i
) is low i s ock p ices can adjus quickly o good news when he ma ke fluc ua es
upwa d. When sho selling is allowed and ma ke ic ion is emo ed, he asymme ic es ic ion be ween
ma gin ading and sho selling disappea s, esul ing in a significan d op in qdiff
i
. Thus, he diffe ence in
c oss-au oco ela ion qdiff
i
can be used o measu e he speed o s ock p ice adjus men o posi i e and nega i e
news.
F om he sho selling fi ms used as he sample, he sho -sale efinancing fi ms a e he ea men g oup and
he es a e he con ol g oup. A e a ea men fi m can be aded h ough sho -sale efinancing, Shock is 1;
o he wise Shock is 0. We exclude he ollowing om he sample: (1) financial companies; (2) companies wi h
nega i e equi y; (3) companies wi h annual s ock ading days unde six mon hs; and (4) companies wi h miss-
ing a iables.
Xu and Chen (2012) sugges ha he P/E a io is ela ed o s ock p icing efficiency om he pe spec i e o a
p icing bubble. Wi h e e ence o Tang e al. (2016), company size (na u al loga i hm o o al asse s a he end
o he yea ) and ma ke - o-book alue a e significan ly co ela ed wi h he speed o s ock p ice adjus men o
news and a e selec ed as con ol a iables. Hou and Moskowi z (2005) decompose he ac o s ela ed o p ice
delay in o adi ional liquidi y indica o s ( ading days, e c.) and in es o ecogni ion o a en ion (numbe o
sha eholde s, e c.) and belie e ha in es o a en ion may explain s ock p ice delays be e han adi ional
liquidi y indica o s. In addi ion, we e e o Hou and Moskowi z (2005) o selec ou con ol a iables, such
as he numbe o employees, ma ke be a coefficien , financial le e age and o he a iables ela ed o he speed
o s ock p ice adjus men o news. Li and Zhang (2015) belie e ha ea nings o ecas s issued by fi ms no sub-
jec o he sho selling es ic ion we e mo e ambiguous and ha he ma ke eac ion o hei p ofi o ecas s
was g ea e , so he quali y o in o ma ion disclosu e may affec he speed o s ock p ice adjus men o news.
On ha basis, we e e o Li and Xiao (2015) and, based on a lis ed company’s ea nings o ecas , di ide i s
con en in o ou ypes o in o ma ion o use as good news: con inuing p ofi abili y, p ofi abili y in ad ance,
sligh p ofi abili y and e u n o p ofi abili y. We coun he numbe o good news i ems disclosed in qua e ly
ea nings o ecas s each yea and name his a iable News, which we use o con ol managemen ea nings o e-
cas in o ma ion o he effec s o p ice disco e y. In addi ion, we use ea nings managemen o con ol o
in o ma ion anspa ency and sha eholding a io o he la ges sha eholde and p opo ion o independen
di ec o s o con ol o he influence o owne ship s uc u e and go e nance le el.
To measu e he influence o sho selling on he speed o s ock p ice adjus men o news, we use fi m fixed
and yea fixed effec s models o con ol o company and ime a ian s in he DID analysis (Be and and
Mullaina han, 2003). A ms ong e al. (2012) use his model when s udying co po a e go e nance and in o -
ma ion en i onmen s. Specific o he field o sho selling, Jin e al. (2015) also use his me hod in hei
esea ch on sho selling and in es men efficiency.
The basic model is as ollows:
qdi i; ¼aiþa þa1Shocki; þCon oli; þei; ð2Þ
in which qdiff
i,
is he p ice adjus men speed o s ock iin pe iod ,a
i
is he fi m fixed effec , a
is he yea fixed
effec . The sho -sale efinancing a iable (Shock) measu es whe he a s ock is allowed o be aded in sho -
sale efinancing. When a s ock is allowed o be aded in sho -sale efinancing, Shock is 1; be o e a secu i y is
allowed o be aded in sho -sale efinancing and o all o he con ol fi ms, Shock is 0. a
1
measu es he speed
o s ock p ice adjus men o nega i e news. When he coefficien o a
1
is significan ly nega i e, i indica es ha
sho selling accele a es he speed o s ock p ice adjus men o nega i e news. O he wise, he opposi e is ue.
386 K. Gao, M. Ding / China Jou nal o Accoun ing Resea ch 12 (2019) 379–394
Chu and Fang (2016) use sho selling olume o measu e sho selling and show ha sho selling can
educe s ock p ice c ash isk o a ce ain ex en . Howe e , sho selle s choose companies wi h mo e concen-
a ed bad news o ade, so he e may be some sample selec ion bias in his conclusion. B is e al. (2007) find
ha he ma ke ’s nega i e bias a e inc eases a e he de egula ion o sho selling, which indica es ha s ock
p ice c ash isk may inc ease a e de egula ion o sho selling. Callen and Fang (2015) each simila conclu-
sions. We also find ha he elaxa ion o sho selling es ic ions in China leads o c ash isk. China allowed
sho selling la e and as a esul , in es o s a e mainly indi iduals, and among ins i u ional in es o s he d
beha io is significan (Xu e al., 2013). Imma u e in es o s may engage in excessi e sho selling wi h o e -
confidence in p i a e in o ma ion o engage in sho selling due o an excessi ely pessimis ic mood, e en ually
leading o sha e p ices alling and inc easing c ashes (Scheinkman and Xiong, 2003).
6. Conclusion
Sho selling allows in es o s holding nega i e news o ade in a imely manne , hus accele a ing he speed
o s ock p ice adjus men o nega i e news (B is e al., 2007). Howe e , inconsis en conclusions ha e been
d awn ega ding his effec in China. Sho -sale efinancing, launched in Feb ua y 2013, expanded he sou ces
o secu i ies a ailable o bo owing and acili a ed sho selling. In he ea ly pilo pe iod, secu i ies financial
companies selec ed only some s ocks as pilo s. The pilo and non-pilo sho selling fi ms a e ou ea men
and con ol g oups, espec i ely. To a ce ain ex en , his minimizes he diffe ences be ween he ea men
and con ol g oups in e ms o liquidi y, ola ili y and u no e , and p o ides an oppo uni y o in es iga e
he economic consequences o sho selling.
Based on his, we use he DID me hod o explo e he influence o Chinese sho selling on he speed o p ice
adjus men o nega i e in o ma ion. We find ha sho selling is conduci e o he imely and effec i e eflec-
ion o nega i e news in s ock p ices. Th ough sho selling, s ock p ices can quickly adjus and eac o neg-
a i e news in he ma ke o accele a e he speed o p ice adjus men o nega i e in o ma ion, educe s ock p ice
delay and imp o e p icing efficiency. In addi ion, in es o s a e mo e likely o conduc sho selling by mining
bad news in an en i onmen wi h less in o ma ion anspa ency. Sho -sale efinancing has diffe en effec s
unde diffe en fi m owne ship s uc u es. In addi ion, sho selling is mo e likely o encou age sho selle s
o dig o nega i e news, ul ima ely aising c ash isk. The esul s will help sho selle s play hei ma ke ole
and imp o e p icing efficiency.
Acknowledgemen
Kaijuan Gao acknowledges he financial suppo om he Na ional Na u al Science Founda ion o China
(No. 71802078) and he Fundamen al Funds o he Cen al Uni e si ies (No. 531107051035).
Re e ences
Ai ken, M.J., F ino, A., Mcco y, M.S., Swan, P.L., 1998. Sho sales a e almos ins an aneously bad news: e idence om he Aus alian
s ock exchange. J. Finan. 53 (6), 2205–2223.
A ms ong, C.S., Balak ishnan, K., Cohen, D., 2012. Co po a e go e nance and he in o ma ion en i onmen : e idence om s a e
an i akeo e laws. J. Accoun . Econ. 52 (1), 185–204.
Be and, M., Mullaina han, S., 2003. Enjoying he quie Li e? Co po a e go e nance and manage ial p e e ences. J. Poli . Econ. 111 (5),
1043–1075.
Boehme , E., Wu, J., 2013. Sho selling and he p ice disco e y p ocess. Re . Finan. S ud. 26 (2), 287–322.
B is, A., Goe zmann, W.N., Zhu, N., 2007. Efficiency and he bea : sho sales and ma ke s a ound he wo ld. J. Finan. 62 (3), 1029–1079.
Callen, J.L., Fang, X., 2015. Sho in e es and s ock p ice c ash isk. J. Bank. Finan. 60, 181–194.
Chang, E.C., Cheng, J.W., Yu, Y., 2007. Sho -sales cons ain s and p ice disco e y: e idence om he Hong Kong ma ke . J. Finan. 62
(5), 2097–2121.
Chang, E.C., Luo, Y., Ren, J., 2014. Sho -selling, ma gin- ading, and p ice efficiency: e idence om he Chinese ma ke . J. Bank. Finan.
48 (C), 411–424.
Chen, J., Hong, H., S ein, J.C., 2001. Fo ecas ing c ashes: ading olume, pas e u ns, and condi ional skewness in s ock p ices. J. Finan.
Econ. 61 (3), 345–381.
K. Gao, M. Ding / China Jou nal o Accoun ing Resea ch 12 (2019) 379–394 393

Chen, S.L., Ma, H., 2017. Sho -selling p essu e and co po a e me ge s and acquisi ions: e idence om quasi-na u al expe imen on
sho -selling de egula ion. Manage. Wo ld 07, 142–156 (in Chinese).
Chu, J., Fang, J.X., 2016. Ma gin- ading, sho selling and he de e io a ion o c ash isk. Econ. Res. J. 51 (05), 143–158 (in Chinese).
Desai, H., Ramesh, K., Thiaga ajan, S.R., Balachand an, B.V., 2002. An in es iga ion o he in o ma ional ole o sho in e es in he
Nasdaq ma ke . J. Finan. 57 (5), 2263–2287.
Diamond, D.W., Ve ecchia, R.E., 1987. Cons ain s on sho -selling and asse p ice adjus men o p i a e in o ma ion. J. Finan. Econ. 18
(2), 277–311.
Die he , K.B., Lee, K.H., We ne , I.M., 2009. Sho -sale s a egies and e u n p edic abili y. Re . Finan. S ud. 22 (2), 575–607.
Du, X., Xiu, Z., 2009. Ins i u ional en i onmen , blockholde cha ac e is ics and owne ship concen a ion in China. Chin. J. Accoun .
Res. 2 (2), 27–57.
Du ne , A., Li, K., Mo ck, R., Yeung, B.Y., 2004. Capi al ma ke s and capi al alloca ion: implica ions o economies in ansi ion. Econ.
T ansi . 12 (4), 593–634.
Fama, E.F., 1965. The beha io o s ock-ma ke p ices. J. Bus. 38 (1), 34–105.
Figlewski, S., 1981. The in o ma ional effec s o es ic ions on sho sales: some empi ical e idence. J. Finan. Quan . Anal. 16 (4), 463–
476.
Gao, K., Lin, W., 2018. Ma gin ading, sho selling, and bond yield sp ead. Chin. J. Accoun . Res. 11 (01), 51–70.
Gu, Z.H., Hao, X.C., 2011. Sho -sale cons ain , in es o s’ beha io and he bubbles o A-s ock ma ke . J. Finan. Res. 02, 129–148 (in
Chinese).
Hong, H., S ein, J.C., 2003. Diffe ences o opinion, sho -sales cons ain s, and ma ke c ashes. Re . Finan. S ud. 16 (2), 487–525.
Hou, K., Moskowi z, T.J., 2005. Ma ke ic ions, p ice delay, and he c oss-sec ion o expec ed e u ns. Re . Finan. S ud. 18 (3), 981–
1020.
Jiang, X.Y., Xu, N.H., 2015. Co po a e o e -in es men and s ock p ice c ash isk. J. Finan. Res. 8, 145–162 (in Chinese).
Jin, Q.L., Hou, Q.C., Li, G., 2015. De egula ion on sho sale cons ain , co po a e in es men and alua ion. Econ. Res. J. 10, 76–88 (in
Chinese).
Ko ha i, S.P., Leone, A.J., Wasley, C.E., 2005. Pe o mance ma ched disc e iona y acc ual measu es. J. Accoun . Econ. 39 (1), 163–197.
Li, K., Xu, L.B., Zhu, W.H., 2014. Sho sale cons ains and s ock misp icing: he e idence om he ma gin ansac ions ins i u ion. Econ.
Res. J. 10, 165–178 (in Chinese).
Li, Y., Zhang, L., 2015. Sho selling p essu e, s ock p ice beha io , and managemen o ecas p ecision: e idence om a na u al
expe imen . J. Accoun . Res. 53 (1), 79–117.
Li, X.Z., Xiao, T.S., 2015. A e managemen ea nings o ecas s use ul o analys o ecas e isions. Nankai Bus. Re . 18 (2), 30–38 (in
Chinese).
Li, Z.S., Chen, C., Lin, B.X., 2015. In o ma ion go e nance effec s o sho selling and ma gin ading. Econ. Res. J. 04, 165–177 (in
Chinese).
Mille , E.M., 1977. Risk, unce ain y, and di e gence o opinion. J. Finan. 32 (4), 1151–1168.
Ni, X.R., Zhu, Y.J., 2017. Does sho -sale p essu e affec fi m’s isk beha io ? Empi ical e idence om he A-sha e ma ke . China Econ.
Qua . 02, 1173–1198 (in Chinese).
Saffi, P.A.C., Sigu dsson, K., 2011. P ice efficiency and sho selling. Re . Finan. S ud. 24 (3), 821–852.
Scheinkman, J.A., Xiong, W., 2003. O e confidence and specula i e bubbles. J. Poli. Econ. 111 (6), 1183–1220.
S ulz, R.M., 2005. The limi s o financial globaliza ion. J. Finan. 60 (4), 1595–1638.
Su, D.W., Ni, B., 2018. Qualified secu i ies o sho -sale efinancing, sho sale cons ains and s ock p ice. Econ. Res. J. 03, 110–125 (in
Chinese).
Tang, S., Wu, Q.J., Wen, D.E., 2016. Sho selling, s ock p ice in o ma i eness and s ock p ice c ash: e idence om ma gin ading in
China. J. Finan. Econ. 42 (8), 74–84 (in Chinese).
Wang, Y., Chen, Y., Wang, J., 2015. Managemen ea nings o ecas s and analys o ecas s: e idence om manda o y disclosu e sys em.
Chin. J. Accoun . Res. 8 (2), 133–146.
Xiao, H., Kong, A.G., 2014. A s udy on he mechanism o he impac o he secu i ies ma gin ading on he fluc ua ions o he special
na u e o he sock p ice: a es based on he diffe ence-in-diffe ence model. Manage. Wo ld 08, 30–43 (in Chinese).
Xin, Q.Q., Kong, D.M., Hao, Y., 2014. T anspa ency and s ock e u n ola ili y. J. Finan. Res. 10, 193–206 (in Chinese).
Xu, H.W., Chen, X., 2012. Does he in oduc ion o ma gin ading in China p omo e he p icing efficiency o he unde lying s ock?
E idence om empi ical esea ch based on he double diffe ence model. Manage. Wo ld 05, 52–61 (in Chinese).
Xu, N.X., Yu, S.Y., Yin, H.Z., 2013. He ding beha io o ins i u ional in es o s and he isk o s ock p ice c ash. Manage. Wo ld 07, 31–
43 (in Chinese).
394 K. Gao, M. Ding / China Jou nal o Accoun ing Resea ch 12 (2019) 379–394