Examining export trade and corporate innovation: A multiphase difference-in-differences method
Abstract
EconStor is a publication server for scholarly economic literature, provided as a non-commercial public service by the ZBW.
Full text
Zhou, Donghua; Peng, Jian ei; Gao, Xue
A icle
Examining expo ade and co po a e inno a ion: A
mul iphase di e ence-in-di e ences me hod
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: Zhou, Donghua; Peng, Jian ei; Gao, Xue (2021) : Examining expo ade and
co po a e inno a ion: A mul iphase di e ence-in-di e ences me hod, China Jou nal o Accoun ing
Resea ch, ISSN 1755-3091, Else ie , Ams e dam, Vol. 14, Iss. 2, pp. 207-230,
h ps://doi.o g/10.1016/j.cja .2021.01.002
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/241837
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/
Examining expo ade and co po a e inno a ion:
A mul iphase diffe ence-in-diffe ences me hod
Donghua Zhou
⇑
, Jian ei Peng, Xue Gao
School o Accoun ancy, Jiangxi Uni e si y o Finance and Economics, China
ARTICLE INFO
A icle his o y:
Recei ed 15 June 2020
Accep ed 26 Janua y 2021
A ailable online 23 Ma ch 2021
Keywo ds:
Expo ade
Technological inno a ion
Economies o scale
Risk- aking
ABSTRACT
Using a mul iphase diffe ence-in-diffe ences model, his s udy in es iga es he
ela ionship be ween expo ade and he co po a e echnological inno a ion
o lis ed companies. I e eals ha engaging in expo ade inc eases co po a e
inno a ion inpu and ou pu . In e ms o pa en ou pu , expo ade g ea ly
p omo es he ou pu o in en ion pa en s and u ili y model pa en s wi h a high
echnological con en . These conclusions emain alid a e a se ies o obus -
ness and endogenei y es s. Rega ding he mechanisms o he obse ed ela-
ionships, expo ade s imula es co po a e echnological inno a ion mainly
by ealizing economies o scale and inc easing isk- aking. The posi i e co e-
la ion be ween expo ade and co po a e echnological inno a ion is s on-
ges among s a e-owned en e p ises, non-high- ech en e p ises, en e p ises
based in cen al and eas e n China, en e p ises engaged in gene al ade, and
en e p ises expo ing o de eloped economies. Gi en he g owing ade ic-
ions ongoing a he ime o w i ing, he conclusions o his s udy p o ide i al
p ac ical guidance and empi ical e idence o a na ional s a egy o
inno a ion-d i en de elopmen .
Ó2021 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
Acco ding o he 2017 S a is ics Repo
1
eleased by IP5, a coali ion o he wo ld’s fi e la ges pa en
offices, he numbe o pa en applica ions filed by China in 2017 eached 1.381 million, which is 3% highe
han in 2016. As shown in Fig. 1, o he fi e coun ies and egions wi h he mos pa en filings (which also
h ps://doi.o g/10.1016/j.cja .2021.01.002
1755-3091/Ó2021 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 .
q
The au ho s acknowledge financial suppo om he Na ional Na u al Science Founda ion o China (No. 71972091). We a e also
g a e ul o he help ul commen s made by Minggui Yu and pa icipan s in he CJAR Summe Resea ch Wo kshop in 2020.
1
www.fi eipoffices.o g/s a is ics.h ml.
China Jou nal o Accoun ing Resea ch 14 (2021) 207–230
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
include he Eu opean Union, he U.S., Japan, and Sou h Ko ea), China accoun s o he la ges p opo ion o
he o al numbe o pa en applica ions wo ldwide, and i s sha e is inc easing.
Howe e , lacking co e echnologies and p op ie a y b ands, mos locally ope a ing Chinese en e p ises a e
s ill me ely p ocesso s o medium- and low-end p oduc s in he global alue chain (Tao e al., 2020). Facing
g owing Sino–U.S. ade ic ions and he inc easingly s ingen es ic ions on echnology ans e imposed
by he U.S. go e nmen , Chinese en e p ises, which gene ally lack high-end co e echnologies, a e suffe ing
unde cons ain s imposed by Eu ope, he U.S., and o he coun ies and egions p oducing echnology spil-
lo e effec s (Zhang e al., 2018). To o e come i s echnological dependence on U.S.-led de eloped coun ies
in he Wes , China mus imp o e i s echnological inno a ion. A he New Economy Fo um held in Beijing on
No embe 22, 2019, China’s Gene al Sec e a y Xi Jinping s a ed: ‘‘Inno a ion is a g ea challenge o ou ime.
The wo ld is unde going majo and unp eceden ed changes his cen u y, and a new ound o echnological
change and indus ial e olu ion is in he ascendan .”
2
Acco ding o Kelle (2010), he echnological inno a-
ion o mos coun ies is a ibu able o echnology spillo e s p oduced by ade. The e o e, whe he expo
ade can effec i ely p omo e co po a e inno a ion is an issue o g ea p ac ical significance.
Mos s udies o co po a e inno a ion ha e ocused on he oles o na ional policies, co po a e go e nance,
and o he non- ade ac o s. Rele an na ional policies include ju idical p o ec ion (Pan e al., 2015), indus-
ial policies (Li and Zheng, 2016; Yu e al., 2016a; 2016b; Howell, 2017), go e nmen subsidies (Yang e al.,
2015; Mao and Xu, 2015; Guo, 2018), and ax e o ms (Lin e al., 2013; Li and Wang, 2017; Sun, 2017). Rel-
e an co po a e go e nance ac o s comp ise mainly owne ship s uc u e (La Po a e al., 1999; Li and Yu,
2
h p://www.xinhuane .com/poli ics/leade s/2019–11/22/c_1125264342.h m.
Fig. 1. Pa en applica ions o he fi e coun ies and egions wi h he mos pa en filings om 2013 o 2017, The ela ed da a a e a ailable
a h ps://www.fi eipoffices.o g/wcm/connec /fi eipoffices/8c519416-173d-4b32-99ed-5387045c46a2/IP5+S a is ics+Repo +2018_2012
2019_ ull.pd ?MOD=AJPERES&CVID= (p. 40 Fig. 3.7).
208 D. Zhou e al. / China Jou nal o Accoun ing Resea ch 14 (2021) 207–230
2015; Chen e al., 2019), employee mo i a ion (Meng e al., 2019; Zhou e al., 2019), business assessmen (Yu
e al., 2016a; 2016b; Li e al., 2018), and execu i e cha ac e is ics (Yu e al., 2018a; 2018b; He e al., 2019).
Since he s a o i s economic e o ms and libe aliza ion, China has expe ienced apid economic g ow h
and s eng hened i s ade connec ions wi h o he coun ies, becoming a ading powe wi h global influence
(He e al., 2020). The apid g ow h o expo ade g ea ly inc eases inno a ion ac i i ies (Li e al., 2016).
When en e p ises s a expo ing, hey encoun e mo e in ense compe i ion in a b oade ma ke , which
encou ages hem o inno a e. Fo example, echnology spillo e s om he ad anced p oduc s o de eloped
coun ies ha e enabled Chinese en e p ises o lea n om and abso b new echnologies (Li and Zhu, 2006;
Tao e al., 2020). Inc eased ade demands also gene a e mo e p ofi s o en e p ises, hus enhancing hei cul-
u e o and capaci y o inno a ion. In his sense, expo ade is an impo an ac o influencing co po a e
inno a ion. Un o una ely, howe e , he ela ionship be ween expo ade and co po a e inno a ion has
a ely been explo ed.
In e ms o inpu , en e p ises engage in expo ade na u ally inc ease hei inpu o p oduc ion ac o s
because he demand o hei p oduc s is inc easing o e seas. In addi ion, he inc eased ou pu scale in a
b oade ma ke lowe s a e age p oduc ion cos s and ealizes economies o scale. F om he pe spec i e o ech-
nology spillo e s, expo ing en e p ises g ea ly inc ease hei isk- aking by lea ning om ad anced p oduc-
ion echnologies, echnological p ocesses, and o ganiza ional managemen (Xu e al., 2015). Economies o
scale and isk- aking and can os e co po a e inno a ion (Lilee a and T efle , 2010; Zhou e al., 2019). This
s udy seeks o answe he ollowing wo ques ions. How do economies o scale influence he ela ionship
be ween expo ade and co po a e inno a ion? How does he le el o isk- aking influence he ela ionship
be ween expo ade and co po a e inno a ion?
To answe hese ques ions, we empi ically examine he ela ionship be ween expo ade and he co po a e
inno a ion o lis ed companies. We find ha engaging in expo ade p omo es co po a e inno a ion in e ms
o bo h inpu and ou pu . Ou conclusions emain alid ollowing a se ies o es s o obus ness and endo-
genei y. Expo ade s imula es co po a e echnological inno a ion mainly h ough economies o scale and
inc eased le els o isk- aking. The s imula ing effec s o expo ade a e obse ed mainly in s a e-owned
en e p ises, non-high- ech en e p ises, en e p ises based in cen al and eas e n China, and en e p ises ha
engage in gene al ade and expo o de eloped coun ies.
This s udy makes wo con ibu ions o he field. Fi s , ela ed s udies ha e a ely examined lis ed compa-
nies; mos ha e used da a on small and medium-sized en e p ises (SMEs) om he China Indus y Business
Pe o mance Da abase o explo e he ela ionship be ween expo ade and co po a e echnological inno a-
ion (Li e al., 2016). Acco ding o he Schumpe e ian hypo hesis, la ge en e p ises usually ha e a s onge
and mo e independen capaci y o inno a ion (Schumpe e , 1942). SMEs gene ally engage in supe ficial,
applica ion-o ien ed inno a ion ac i i ies, such as he pu e imi a ion o efinemen o exis ing echnologies.
In con as , lis ed companies ha e a undamen al and adical po en ial o echnological inno a ion, such
as e olu ionizing exis ing echnologies o pionee ing explo a ions o unknown fields (Chi, 2002). They gen-
e ally engage in high-end inno a ion ac i i ies ha e ec long-las ing echnological ba ie s and accu a ely
eflec hei coun y’s le el o echnological inno a ion. Thus, by examining he ela ionship be ween lis ed
companies’ expo ade and co po a e inno a ion, his s udy se es as an impo an supplemen o he li e -
a u e in his field. The second con ibu ion made by his s udy is o cla i y he mechanism by which expo
ade influences co po a e inno a ion om he pe spec i es o isk- aking and economies o scale. In ligh
o he Sino–U.S. ade ic ions ongoing a he ime o w i ing, his s udy offe s i al p ac ical guidance o
Chinese en e p ises on main aining hei expo ade, imp o ing hei inno a ion abili y, abandoning hei
hea y dependence on U.S. science and echnology, and con ibu ing o China’s ans o ma ion in o an
inno a ion-d i en na ion.
Pa II o his pape p esen s he li e a u e e iew and hypo heses. Pa III desc ibes he design o he
s udy. Pa IV discusses he empi ical esul s. Pa V p esen s he conclusions and offe s ecommenda ions
o Chinese en e p ises o imp o e hei inno a ion capaci y unde he backg ound o Sino–U.S. ade
ic ions.
D. Zhou e al. / China Jou nal o Accoun ing Resea ch 14 (2021) 207–230 209
2. Li e a u e e iew and hypo heses
2.1. Li e a u e e iew
In he field o in e na ional ade, wo key hypo heses ha e been p oposed ega ding he ela ionship
be ween expo ade and co po a e echnological inno a ion. The fi s ela es o ‘‘sel -selec ion”effec s and
he second conce ns ‘‘lea ning by expo ing”effec s. The fi s hypo hesis s a es ha expo ba ie s (specifically
he cos s o anspo a ion, dis ibu ion and ma ke ing, and adap ing p oduc s o mee o eign s anda ds)
allow only en e p ises wi h high p oduc i i y o o e come he impac o sunk cos s and en e he expo ma -
ke . Hence, acco ding o his hypo hesis, co po a e echnological inno a ion affec s expo s. In con as , he
hypo hesis ega ding lea ning by expo ing effec s s a es ha expo ing en e p ises can acqui e new echnolog-
ical knowledge om compe i o s, pa ne s, and cus ome s. Hence, he hypo hesis sugges s ha expo ade
affec s co po a e echnological inno a ion (De Loecke , 2007; Zhao and Li, 2007; Qian e al., 2011; Li e al.,
2016; Zhang e al., 2016).
2.1.1. Sel -selec ion effec s
The concep o sel -selec ion effec s is fi s p oposed by Cle ides e al. (1998), who s a e ha inno a ion
inc eases en e p ises’ p oduc i i y and affec s hei expo decisions. In a s udy o U.S. en e p ises, Be na d
and Jensen (1999; 2004b) find ha e en new en an s in o he expo ma ke a e mo e p oduc i e han
non-expo ing en e p ises a e. Examining manu ac u ing da a om Sou h Ko ea, Aw e al. (2000) disco e
sel -selec ion effec s among newly expo ing en e p ises. Using a ade model wi h he e ogeneous en e p ises,
Meli z (2003) explains he ela ionship be ween en e p ise diffe ences and expo beha io s and shows ha
ade openness inc eases he p oduc i i y o an en i e indus y h ough he sel -selec ion effec s p oduced
by en e p ises’ expo ing beha io s. S udies in his field ha e also disco e ed sel -selec ion effec s among B i-
ish en e p ises (Gi ma e al., 2004), Ge man en e p ises (Wagne , 2002; A nold and Hussinge , 2005), and
Chinese en e p ises (Qian e al., 2011; Yi and Fu, 2011; Qiu e al., 2012). Howe e , o he schola s no e ha
he p oduc i i y o expo ing Chinese en e p ises is lowe han ha o non-expo ing en e p ises, i.e., he ‘‘p o-
duc i i y pa adox.”They a gue ha he la ge numbe o p ocessing ade en e p ises in China has d agged
down he a e age p oduc i i y o expo ing en e p ises, he eby nega ing sel -selec ion effec s (Li, 2010; Lu
e al., 2010).
2.1.2. Lea ning by expo ing effec s
Acco ding o he lea ning by expo ing hypo hesis, expo ing en e p ises in a b oad and compe i i e ma ke
lea n ad anced p oduc ion echnologies ha u he s eng hen hei abili y o lea n (Xiang and Ma, 2013).
Long a ho spo issue in he field o in e na ional ade, lea ning by expo ing effec s ha e been ex ensi ely
s udied by schola s bo h in China and ab oad, bu no unanimous conclusions ha e been eached. As indica ed
by he le els o economic de elopmen in expo ing coun ies, lea ning by expo ing effec s do no a ise in he
majo i y o de eloped coun ies (Be na d and Jensen, 2004a; Wagne , 2002; G eenaway and Yu, 2004). How-
e e , such effec s ha e been obse ed in newly indus ializing economies and de eloping coun ies (Cle ides
e al., 1998; Aw e al., 2000; Al a ez and Lopez, 2005). By aking pa in in e na ional compe i ion, newly
indus ializing economies and de eloping coun ies a e able o lea n om and abso b ad anced echnologies
om de eloped coun ies, hus imp o ing hei independen inno a ion abili y (Aw e al., 2000; De Loecke ,
2007; Al a ez and Lopez, 2005). Howe e , om he pe spec i e o indus y cha ac e is ics, lea ning by expo -
ing effec s a e diminished o nega ed in indus ies wi h la ge numbe s o o eign- unded en e p ises
(G eenaway and Knelle , 2007).
In ecen yea s, ex ensi e effo s ha e been made o explo e lea ning by expo ing effec s in samples o Chi-
nese en e p ises. Focusing on modes o ade, Zhang e al. (2008), Li and Zhao (2010), Fan and Feng (2013),
and Bao e al. (2014) find ew o no long- e m lea ning by expo ing effec s among Chinese en e p ises. They
conclude ha significan lea ning by expo ing effec s exis among gene al ade en e p ises bu no p ocessing
ade en e p ises, which accoun o he majo i y o Chinese en e p ises. S udies show ha p ocessing ade
en e p ises a e locked in o labo -in ensi e links wi h low echnological con en (Bao e al., 2014). Such en e -
p ises a e elian on low-end manu ac u ing, exogenous expo ade in e media ies, and sufficien ly cheap
210 D. Zhou e al. / China Jou nal o Accoun ing Resea ch 14 (2021) 207–230
labo , and lack he p essu e and mo i a ion o inc ease hei efficiency (Jing e al., 2012). Unlike gene al ade
en e p ises, p ocessing ade en e p ises en e he in e na ional ma ke by joining he global alue chains o
mul ina ional co po a ions. They nei he di ec ly compe e wi h o he en e p ises no en e in o di ec con ac
wi h he ul ima e pu chase s in he in e na ional ma ke , so hey do no expe ience he lea ning by expo ing
effec s obse ed among gene al ade en e p ises (L e al., 2016). Howe e , some schola s ake he opposi e
iew. Using mic oda a om China, Dai and Yu (2012) find ha en e p ises expo ing o he fi s ime see a
p oduc i i y inc ease o 2% in hei fi s yea o expo ing, bu no significan inc eases in he ollowing yea s.
Examining su ey da a eleased by he Wo ld Bank, Wang e al. (2011) obse e lea ning by expo ing effec s
among Chinese en e p ises. Using he p opensi y sco e ma ching (PSM) diffe ence-in-diffe ences (DID) (PSM-
DID) me hod, Qiu e al. (2012) pe o m a wo-sided es o he causali y be ween he expo beha io s and
p oduc i i y o Chinese manu ac u ing en e p ises. They conclude ha significan lea ning by expo ing effec s
exis among Chinese manu ac u ing en e p ises and ha hese effec s g adually inc ease o e ime. Explo ing
he ela ionship be ween en e p ises’ engagemen in expo ade and independen inno a ion, Li e al. (2016)
epo ha he effec s o expo a ion on co po a e echnological inno a ion show an in e ed ‘‘U”shape, wi h
an ini ial inc ease ollowed by a dec ease.
2.2. Theo e ical analysis and hypo heses
Expo ade can p omo e he echnological inno a ion o a coun y (Wagne , 2007). When an en e p ise
begins o expo , i mus conduc echnological inno a ion o s ay compe i i e in he in e na ional a ena by
mee ing he demands o new cus ome s and ma ke s. Compa ed wi h non-expo ing en e p ises, expo ing
en e p ises a e mo e p oduc i e (Meli z, 2003; Wagne , 2007) and mo e compe i i e (Mao and Sheng,
2014). In pu sui o de elopmen , expo ing en e p ises end o seek business expansion in he in e na ional
ma ke a e mee ing domes ic demands (Mao and Wang, 2006). Howe e , an expanded ma ke b ings mo e
in ense compe i ion, and o ea n a place in he highly compe i i e in e na ional ma ke , expo ing en e p ises
mus ei he p o ide be e p oduc s o cha ge lowe p ices. Bo h op ions equi e echnological inno a ion,
ei he o upg ade p oduc quali y (Fa uq, 2010) o o educe cos s, leading o lowe p ices (Chen, 2002).
Acco ding o he lea ning by expo ing hypo hesis, expo ade offe s de eloping coun ies an impo an
means o acqui ing echnology spillo e s om de eloped coun ies and accessing ad anced echnological
knowledge. Wi hin an indus y, echnology spillo e s om new o eign cus ome s and compe i o s help en e -
p ises o s eng hen hei echnological inno a ion. Ac oss indus ies, expo ade p omo es mu ual in eg a-
ion, exchange, and coope a ion h ough he global di ision o labo . Technology spillo e s also acili a e he
sha ing o inno a ion- ela ed esou ces, hus p omo ing echnological inno a ion among expo ing en e -
p ises. Mo eo e , expo ade leads o ‘‘induced inno a ion”, as expo ing en e p ises can imp o e hei ech-
nological inno a ion by ‘‘lea ning du ing expo ing”
3
(Li e al., 2016). By compe ing in he b oade
in e na ional ma ke , en e p ises can s eng hen hei abili y o sol e new p oblems using hei exis ing
esou ces, enhance hei echnological inno a ion h ough he knowledge hey ha e gained om ab oad, d aw
inspi a ion om diffe en ma ke s and cul u es, and imp o e hei abili y o inno a e (Salomon and Sha e ,
2005).
Beyond encou aging en e p ises o engage in echnological inno a ion and inno a i e lea ning, compe i ion
and echnology spillo e s offe c ucial ma u e condi ions and esou ces ha enable en e p ises o aise hei
le el o inno a ion. Fo expo ing en e p ises, he domes ic ma ke is no longe he sole conce n; a b oade
in e na ional expo ma ke is bound o c ea e mo e demand. By mee ing bo h in e na ional and domes ic
demand, expo ing en e p ises a e able o gain mo e ade p ofi s and engage in mo e inno a i e ac i i ies
(Li, 2002). Conside ing he effec s o expo demand, we p opose ou fi s hypo hesis, as ollows.
H1: O he condi ions being equal, expo ade p omo es co po a e echnological inno a ion.
Domes ic ma ke agmen a ion, high c edi cos s, he absence o in ellec ual p ope y p o ec ion, and
o he dis o ions in he ins i u ional en i onmen p e en local Chinese en e p ises om using hei domes ic
3
Li e al. (2016) sugges ha expo ing en e p ises can imp o e hei echnological inno a ion by ‘‘lea ning du ing expo ing”, i.e.,
‘‘induced inno a ion”(p. 76).
D. Zhou e al. / China Jou nal o Accoun ing Resea ch 14 (2021) 207–230 211
ma ke capaci y o ealize economies o scale and achie e apid g ow h. Ins ead, expo ade becomes an
impo an means by which Chinese en e p ises can ealize economies o scale (Zhang e al., 2009). As men-
ioned abo e, a b oade in e na ional ma ke c ea es mo e demand, which d i es expo ing en e p ises o
expand he scale o hei p oduc ion, inc ease hei inpu o he ac o s o p oduc ion, launch mass p oduc ion,
and ealize economies o scale. These ou comes in u n play i al oles in s imula ing co po a e inno a ion
(Lilee a and T efle , 2010). The heo y o Ma shallian ex e nali ies s esses he impo ance o economies o
scale and a ious ex e nali ies in p omo ing inno a ion (Han and Ke, 2012). When manu ac u ing a p oduc
ea ns a company inc easing e u ns o scale, he company ob ains a cos ad an age due o he expanded scale
o p oduc ion and educed cos pe uni p oduc (K ugman, 1980). In addi ion, economies o scale p omo e
indus ial expansion and s eng hen indus ial compe i i eness, which u he mo i a e con inuous inno a ion
(Flo ida, 1994). The e o e, ealizing economies o scale helps expo ing en e p ises o educe hei p oduc ion
cos s, inc ease hei p ofi s, and lowe hei inno a ion isks while mo i a ing hem o engage in echnological
inno a ion. Conside ing he effec s o economies o scale, we p opose ou second hypo hesis, as ollows.
H2: O he condi ions being equal, engaging in expo ade mo i a es en e p ises o engage in echnological
inno a ion by inc easing hei economies o scale.
En e p ise inno a ion is key o gaining a compe i i e ad an age (Teece e al., 1997; Bae , 2012), imp o ing
co po a e pe o mance, and inc easing sha eholde weal h. Howe e , inno a ion also equi es in es men s o
ime, ene gy, and esou ces (such as expendi u e on R&D and managemen o inno a ion) ha ha e e y
unce ain e u ns (Wu e al., 2005; Ling e al., 2008). En e p ises ha ake mo e isks usually ha e a highe
ole ance o inno a ion- ela ed isks and unce ain ies, as well as g ea e confidence in inno a ion p ojec s
wi h high le els o isk and unce ain y (Faccio e al., 2011). In u n, hese fi ms’ highe le els o ole ance
and confidence inc ease hei inno a ion pe o mance.
Expo ing en e p ises ha seek o compe e in he in e na ional ma ke mus gua an ee p oduc quali y. To
mee consume s’ demands, newly expo ing en e p ises mus imp o e hei echnological p ocesses and s an-
da ds, upg ade hei machine y and equipmen , and o ganize aining o hei employees o lay a solid oun-
da ion o inc eased isk- aking (Ge effi e al., 2005). Due o lea ning by expo ing effec s, en e p ises wi h
be e p oduc ion me hods, capi al equipmen , p oduc design, and o ganiza ional managemen a e be e able
o capi alize on isky in es men oppo uni ies (G ossman and Helpman, 1990). Mo eo e , en e p ises ha
enjoy lowe a e age p oduc ion cos s and highe p oduc i i y om inc eased ou pu s a e mo e mo i a ed
and be e able o engage in p omising and p ofi able in es men ac i i ies. This in u n p omo es hei
isk- aking (Xu e al., 2015) and inno a ion. Conside ing he impo ance o echnological inno a ion o
isk- aking, we p opose ou hi d hypo hesis, as ollows.
H3: O he condi ions being equal, expo ade mo i a es en e p ises o conduc echnological inno a ion
by inc easing hei isk- aking.
3. Resea ch design
3.1. Sample selec ion and da a sou ce
As he China cus oms da abase cu en ly only con ains expo ade da a on Chinese en e p ises o he
pe iod 2000–2015, ou sample pe iod is 2000–2015. Mos ela ed s udies ha e used indus ial da a on SMEs
o s udy he effec o expo ade on inno a ion. Howe e , unlike lis ed companies, SMEs engage mainly in
imi a i e and low-end inno a ion ac i i ies (Chi, 2002), which do no ully eflec China’s s eng hs in inno-
a ion. Hence, we ake companies lis ed on he Shanghai and Shenzhen A-sha e s ock ma ke s as ou sample.
We ob ain da a on he financial and o he cha ac e is ics o he en e p ises om he China S ock Ma ke
and Accoun ing Resea ch Da abase, which collec s all o he ele an financial indica o s o lis ed companies.
We elimina e companies ha me any o he ollowing c i e ia. (1) Con ained ‘‘ST”(indica ing special ea -
men fi ms) o ‘‘PT”(indica ing pa icula ans e fi ms) in hei abb e ia ions. The financial da a o such
lis ed companies a e p ocessed be o e elease and canno eflec hei financial s a uses u h ully. (2)
Asse –liabili y a ios abo e 1. The business condi ions o hese lis ed companies a e abno mal and ha e no
e e en ial alue. (3) Classified as financial o insu ance companies. Such fi ms adop diffe en ial financial
accoun ing sys ems, which may comp omise da a consis ency. (4) Appea ed no mo e han h ee imes du ing
212 D. Zhou e al. / China Jou nal o Accoun ing Resea ch 14 (2021) 207–230
he sample pe iod. Each sampled company is equi ed o ha e a leas wo yea s o obse ed alues be o e and
a e s a ing o expo , so ha he con inuous effec s o expo ade on co po a e inno a ion could be seen
mo e clea ly. (5) Lis ed o delis ed in he same yea as beginning o expo . (6) Missing subs an ial amoun s o
da a. To be e s udy he effec s o expo ade on co po a e inno a ion, we classified he sampled companies
in o h ee ca ego ies: non-expo ing (non-expo ) en e p ises, exis ing expo ing en e p ises, and newly expo -
ing (new-expo ) en e p ises (Dai and Yu, 2012; Qiu e al., 2012; Li e al., 2016).
4
A e elimina ing exis ing
expo ing en e p ises, we ob ain da a on 1507 lis ed companies and 13,666 company-yea obse ed alues
(4960 o new-expo en e p ises and 8706 o non-expo en e p ises). To a oid he effec s o ex eme alues,
we winso ize he main a iables a he uppe and lowe 1%.
3.2. Model design and a iable defini ions
To explo e he effec s o engaging in expo ade on he echnological inno a ion o ou sampled lis ed
companies, we in oduce a mul i-s age diffe ence-in-diffe ences (mul i-DID) model o es ing. Following s ud-
ies o expo ade (De Loecke , 2007; Qiu e al., 2012; Li e al., 2016; Zhang e al., 2016) and co po a e ech-
nological inno a ion (Li e al., 2016; Zhou e al., 2019), we con ol o co po a e financial a iables and
co po a e go e nance a iables, as well as he co po a e fixed effec (Fi m) and he annual fixed effec (Yea ):
Inno i; ¼a0þa1T ea i Pos þXCon ols þXFi m þXYea þei; ð1Þ
whe e co po a e inno a ion (Inno
i,
) is he explained a iable, measu ed in e ms o inno a ion inpu and
inno a ion ou pu . Inno a ion inpu is measu ed by co po a e R&D in ensi y, which is defined as he a io
o o al R&D inpu s o o al asse s. Lis ed companies may ans e hei inno a ion ac i i ies o hei sub-
sidia ies o affilia ed companies (Yu e al., 2016a; 2016b). In he abo e model, inno a ion ou pu is measu ed
by he na u al loga i hms o o al numbe o pa en applica ions (To al), numbe o in en ion pa en applica-
ions (In en ), numbe o u ili y model pa en applica ions (U ili y), and numbe o appea ance design pa en
applica ions (Design) a e adding 1 o hem. These a iables co e he inno a ion ac i i ies o a lis ed com-
pany and i s subsidia ies, associa ed companies, and join en u e pa ne s. T ea
i
is a dummy a iable in o-
duced o dis inguish he ea men g oup om he con ol g oup, se o 1 o new-expo en e p ises and 0 o
non-expo en e p ises. Pos
is a dummy a iable o en e p ise expo ime, se o 1 o he fi s and subse-
quen yea s o expo ing bu o 0 o he p e ious yea s. T ea
i
Pos
is a co e explana o y a iable se o 1
o he fi s and subsequen yea s o expo ing bu o 0 o he p e ious yea s. Calcula ed by he DID me hod,
he coefficien o his a iable measu es he effec s o expo ade on he echnological inno a ions o he lis ed
companies, i.e., he a e age change in he le el o co po a e inno a ion o new-expo en e p ises ela i e o
non-expo en e p ises.
To es he he e ogenei y o he lis ed companies, we also in oduce a se ies o fi m cha ac e is ic a iables,
such as owne ship na u e (Soe), echnological le el (Tech), dis ic (Dis ic ), expo des ina ion coun y ype
(Coun y), and mode o expo ade (T ade ype). I a lis ed company is s a e-owned, Soe is se o 1; o he wise,
i is se o 0. Following Gu e al. (2018), we classi y he lis ed companies as high- ech en e p ises o non-high-
ech en e p ises. I a lis ed company is a high- ech en e p ise, Tech is se o 1; o he wise, i is se o 0. In e ms
o egional dis ibu ion, i a company is loca ed in wes e n China, Dis ic is se o 0, bu o cen al and eas -
e n China, i is se o 1. Following Nielsen (2013) and he Human De elopmen Index
5
, we classi y he expo
des ina ion coun y ypes o he lis ed companies as de eloped o de eloping. I a lis ed company p edomi-
nan ly expo s o de eloped coun ies, Coun y is se as 1; o he wise, i is se o 0. Finally, ollowing Li
e al. (2016), we classi y he modes o expo ade o he lis ed companies as ei he gene al ade o p ocessing
and mixed ade. I he olume o he p ocessing ade o a lis ed company accoun o less han 25% o i s
o al expo ade olume, he lis ed company is deemed as mainly engaged in gene al ade, so T ade ype
4
A non-expo en e p ise is defined as an en e p ise ha do no expo du ing he sample pe iod. An exis ing expo ing en e p ise is
defined as an en e p ise ha expo h oughou he sample pe iod. A new-expo en e p ise is defined as an en e p ise ha begin expo ing
a yea a e ha ing been lis ed du ing he sample pe iod.
5
Human De elopmen Index is a ailable a h p://hd .undp.o g/si es/de aul /files/hd 2019.pd . (p. 300–311).
D. Zhou e al. / China Jou nal o Accoun ing Resea ch 14 (2021) 207–230 213
is assigned a alue o 1. O he wise, he company is deemed as mainly engaged in p ocessing and mixed ade,
so T ade ype is assigned a alue o 0.
The con ol a iables (Con ols) a e as ollows: o al ac o p oduc i i y (T p), company size (Size), finan-
cial le e age a io (Le ), e u n on asse s (Roa), company g ow h (G ow h), capi al s ock pe capi a (Ak), an-
gible asse a io (Tangibili y), cash le el (Cash), company lis ing age (Age), financial cons ain s (KZ), boa d
size (Boa d), independen di ec o a io (Indep), managemen sha eholding a io (Sha e), and chai pe son–
CEO duali y (Dual). Fu he , o con ol o he owne ship, indus y, and egional cha ac e is ics o he en e -
p ise, we add na u e o owne ship (Soe), echnology le el (Tech), and loca ion (Dis ic ) o he model.
6
Specific
a iable defini ions can be ound in Table 1.
4. Empi ical esul s
4.1. Summa y s a is ics
Table 2 epo s he summa y s a is ics o he main a iables. Column (1) p esen s he s a is ical cha ac e -
is ics o he ull sample; columns (2) and (3) p esen he mean pe o mance o he main a iables o
Table 1
Defini ions o a iables.
Va iables Name Sign Defini ion
Dependen
Va iables
R&D in ensi y RD R&D expendi u e/ o al asse s
To al numbe o pa en
applica ions
To al ln( o al numbe o pa en applica ions + 1)
Numbe o in en ion pa en
applica ions
In en ln(numbe o in en ion pa en applica ions + 1)
Numbe o u ili y model
pa en applica ions
U ili y ln(numbe o u ili y model pa en applica ions + 1)
Numbe o appea ance design
pa en applica ions
Design ln(numbe o appea ance design pa en applica ions + 1)
Independen
Va iables
Expo ype dummy a iable T ea
i
1 o new-expo en e p ises, 0 o non-expo en e p ises
Expo ime dummy a iable Pos
1 o expo yea and a e , o he wise 0
Expo dummy a iable T ea
i
Pos
In e ac ion e m o T ea
i
and Pos
Mode a ing
Va iables
Na u e o owne ship Soe 1 o s a e-owned en e p ises, o he wise 0
Technology le el Tech 1 o high- ech en e p ises, 0 o non-high- ech en e p ises
Loca ion Dis ic 1 o en e p ises in cen al and eas e n China, 0 o en e p ises in
wes e n China
Expo des ina ion coun y
ype
Coun y 1 o en e p ises wi h de eloped expo des ina ion coun ies, 0 o
en e p ises wi h de eloping expo des ina ion coun ies
Expo ade ype T ade ype 1 o gene al ade, o he wise 0
Con ol
Va iables
To al ac o p oduc i i y T p P oduc i i y, measu ed by Le insohn- Pe in (2003) me hod
Company size Size ln( o al asse s)
Financial le e age a io Le To al liabili ies/ o al asse s
Re u n on asse s Roa Ne p ofi s/ o al asse s
Company g ow h G ow h (cu en ope a ing income - las ope a ing income)/las ope a ing
income
Capi al s ock pe capi a Ak Fixed asse s/numbe o employees a he end o he pe iod
Tangible asse a io Tangibili y Tangible asse s/ o al asse s
Cash le el Cash Cash and cash equi alen s/ o al asse s
Company lis ed age Age ln(company lis ed age + 1)
Financial cons ain s KZ KZ index. The la ge he KZ index, he g ea e he financial
cons ain s o a lis ed company
Boa d size Boa d ln(numbe o boa d membe s)
Independen di ec o a io Indep Numbe o independen di ec o s/numbe o boa d membe s
Managemen sha eholding
a io
Sha e Numbe o sha es held by managemen / o al numbe o sha es in
he company
Chai pe son–CEO duali y Dual 1 o chai pe son–CEO duali y, o he wise 0
6
We hank an anonymous e e ee o se e al aluable commen s on he con ol a iables ha helped o g ea ly imp o e he pape .
214 D. Zhou e al. / China Jou nal o Accoun ing Resea ch 14 (2021) 207–230
en e p ises, he issue o inhe en selec ion bias canno be igno ed. To inc ease he obus ness o he esul s
abo e, we u he use he PSM me hod o explo e he ela ionship be ween expo ade and he echnological
inno a ion o lis ed companies. Specifically, we fi s es ima e a logi eg ession o p edic he expo p obabil-
i y o he lis ed companies wi h all o he con ol a iables men ioned abo e. Nex , we selec samples o non-
expo en e p ises ha a e simila o hose o new-expo en e p ises by one- o-fi e nea es neighbo PSM o
elimina e significan diffe ences be ween he ea men g oup and he con ol g oup be o e he expo shock.
This educes he bias caused by he sel -selec ion effec o lis ed companies be o e and a e expo ing. Finally,
we e-es ima e he mul i-DID model using he ma ched sample o explo e he ne a e age impac o expo
ade on he echnological inno a ion o lis ed companies. The es ima ion esul s a e shown in Table 8.
The coefficien es ima es o T ea
i
Pos
a e all significan ly posi i e a he 1% le el o R&D in ensi y
(RD), he o al numbe o pa en applica ions (To al), and he numbe o in en ion pa en applica ions
(In en ), indica ing ha expo ade p omo es he echnological inno a ion o lis ed companies. Ou es ima-
ion esul s a e consis en wi h he baseline esul s.
(2) Endogenei y p oblems caused by wo-way causali y and non- andom policies
Al hough he DID me hod is an app op ia e ool o e alua e he effec o expo ade policy, i s effec i e-
ness depends on he andomness (exogenei y) o he policy a iables. I an en e p ise’s expo beha io is
ela ed o i s inno a ion capabili ies, he explana o y a iable o expo ing is endogenous. To sol e he p ob-
lems o wo-way causali y and policy endogenei y in he model, we implemen wo main measu es. One is o
add he lag e m o he explained a iable o he eg ession model o con ol ac o s ha a y be ween en e -
p ises and ac oss ime o elimina e he impac o en e p ise inno a ion on expo ade. The second is o find
an exogenous ins umen al a iable (IV).
(1) Adding he lag e m o he explained a iable o he model
As indica ed abo e, we add ess po en ial endogenei y p oblems in he model by adding he lag e m o he
explained a iables o he model. Howe e , his b ings new endogenei y p oblems due o dynamic panel de i-
a ion. To sol e he la e p oblems, we ollow Zhang and Zheng (2013) and Feng and Liu (2017) in using he
wo-s age sys em gene alized me hod o momen s (GMM) me hod o he eg ession and adding yea dummy
a iables o he model o con ol he ime end.
The eg ession esul s a e shown in Table 9. Columns (1)-(3) epo he es ima ion esul s o ou dynamic
panel models o he R&D in ensi y (RD), o al numbe o pa en applica ions (To al), and numbe o
Table 8
T ea men o endogenei y p oblems: p opensi y sco e ma ching es .
Va iables (1) (2) (3)
RD
i,
To al
i,
In en
i,
T ea
i
Pos
0.0042*** 0.4033*** 0.4318***
(3.303) (3.456) (3.956)
Cons an 0.0340** 6.3203*** 5.2824***
(2.407) (4.139) (3.849)
Fi m fixed effec s Yes Yes Yes
Yea fixed effec s Yes Yes Yes
N7541 7541 7541
Adj R
2
0.143 0.290 0.284
No e: (1) As expo ade and inno a ion may ha e an indus y clus e effec , ou ma ching p ocess is ea ed by indus y.[We hank an
anonymous e e ee o se e al aluable commen s on conduc ing PSM by indus y ha helped o g ea ly imp o e he pape .] In addi ion,
ou indus y di ision me hod ollows he classifica ion me hod used in he mains eam li e a u e; ha is, he manu ac u ing indus y is
di ided acco ding o he CSRC 2012 Indus y Classifica ion and all o he indus ies a e di ided acco ding o he CSRC 2012 Indus y
Classifica ion. (2) Con ol a iables a e included in all o he abo e models; (3) T-s a is ics based on obus s anda d e o s clus e ed a he
fi m le el a e epo ed in pa en heses. ***, **, * deno e s a is ical significance a he 1%, 5%, and 10% le els, espec i ely.
D. Zhou e al. / China Jou nal o Accoun ing Resea ch 14 (2021) 207–230 221
in en ion pa en applica ions (In en ) o he lis ed companies. The esul s o he Hansen es and he esidual
sequence co ela ion es suppo he alidi y o he ins umen al a iable in he model, he esidual fi s -o de
sequence co ela ion (AR (1)), and he second-o de sequence co ela ion (AR (2)).
In addi ion, Table 9 shows ha a e con olling he explained a iables o he lagging pe iod and using
wo-s age sys em GMM o sol e p oblems o endogenei y, he coefficien es ima es o T ea
i
Pos
a e s ill
significan ly posi i e a he 1% o 5% le el. This again indica es ha expo ade p omo es he inno a ion
ac i i ies o lis ed companies in e ms o bo h inno a ion inpu and inno a ion ou pu .
(2) Ins umen al a iable me hod
Nex , we add ess he endogenei y p oblem associa ed wi h ade policy o imp o e he eliabili y o he con-
clusions o his pape . We use he IV me hod as he final obus ness es ima ion, aking he lag e m o en e -
p ise expo ade olume as he IV.
Specifically, we fi s ake he na u al loga i hm o expo ade olume and hen ake he lag e m o he
loga i hmic expo ade olume as he IV (IV). Gene ally, he expo beha io o Chinese en e p ises is cha -
ac e ized by pe sis ence (Chen e al., 2012; Dai and Zheng, 2015). The la ge an en e p ise’s expo ade ol-
ume in he p e ious yea , he highe he p obabili y o he en e p ise’s engaging in expo beha io in he
cu en yea . The e o e, he lag e m o en e p ise expo ade olume, selec ed as he IV o expo ade, sa -
isfies he co ela ion hypo hesis. In addi ion, o sa is y he exclusi i y hypo hesis o he IV app oach, he e
mus be no di ec co ela ion be ween he expo ade olume o he p e ious pe iod and en e p ise inno a-
ion in he cu en pe iod. Tha is, he echnological inno a ion le el o he en e p ise in he cu en pe iod
mus no affec i s expo ade olume in he p e ious pe iod. In sum, ou IV is alid.
The eg ession esul s a e shown in Table 10, in which column (1) p esen s he esul s o he fi s s age o he
IV eg ession and columns (2)-(4) epo he esul s o he second s age o he ins umen al a iable eg ession.
Fi s , he es ima ion esul s o he fi s s age show ha he g ea e he expo ade olume o lis ed companies
in he p e ious pe iod, he mo e willing en e p ises a e o expo in he cu en pe iod. The es ima ion esul s
o he second s age show ha a e he use o he IV, expo ade s ill significan ly imp o es he inno a ion
inpu and ou pu o lis ed companies. The e o e, he esul s a e add essing endogenei y p oblems again sup-
po he conclusions o his pape .
Table 9
T ea men o endogenei y p oblems: wo-s age sys em gene alized me hod o momen s.
Va iables (1) (2) (3)
RD
i,
To al
i,
In en
i,
RD
i, -1
0.2350*
(1.811)
To al
i, -1
0.2447***
(4.496)
In en
i, -1
0.2526***
(4.461)
T ea
i
Pos
0.0193* 3.4028*** 2.7499***
(1.875) (5.094) (6.125)
Cons an 0.0344 23.5889*** 22.9229***
(0.254) (3.221) (2.813)
AR(1) 0.022 0.000 0.000
AR(2) 0.426 0.284 0.606
Hansen 0.352 0.258 0.147
Yea fixed effec s Yes Yes Yes
N9,005 9,005 9,005
No e: (1) Con ol a iables a e included in all o he abo e models; (2) T-s a is ics based on obus s anda d e o s clus e ed a fi m le el
a e epo ed in pa en heses. ***, **, * deno e s a is ical significance a he 1%, 5%, and 10% le els, espec i ely.
222 D. Zhou e al. / China Jou nal o Accoun ing Resea ch 14 (2021) 207–230
4.5. Addi ional analysis
4.5.1. Mechanism analysis
The esul s abo e indica e ha engaging in expo ade p omo es he echnological inno a ion o lis ed
companies. I is also impo an o explo e he pa hs and mechanisms o his effec . Acco ding o he heo e ical
analysis o economies o scale and o isk- aking, fi s , an inc ease in ma ke demand leads en e p ises o
expand hei p oduc ion scale and inc ease hei inpu o p oduc ion ac o s o achie e economies o scale,
hus p omo ing en e p ise inno a ion. Second, compe i ion and lea ning effec s by expo ing significan ly
imp o e en e p ise managemen , p oduc ion echnology, and p oduc p ocess and inc ease en e p ises’ isk-
aking, he eby enhancing hei echnological inno a ion. Based on his, we es ima e he ollowing models:
Yi; ¼c0þc1T ea i Pos þXCon ols þXFi m þXYea þei; ð3Þ
RiskTi; ¼g0þg1T ea i Pos þXCon ols þXFi m þXYea þei; ð4Þ
In model (3), he explained a iable Y
i,
ep esen s economies o scale, measu ed by he loga i hm o he
numbe o employees, in e media e p oduc inpu , ope a ing income, and p ofi s (He e al., 2020). In model
(4), RiskT
i,
ep esen s he isk- aking le el o an en e p ise. Re e ing o John e al. (2008), Boubak i e al.
(2013), and Yu e al. (2013), we choose as ou p ima y measu e o co po a e isk- aking (RiskT
i,
) he ola ili y
o a fi m’s indus y-adjus ed e u n on asse s (ADJ_ROA
i,
) o e h ee-yea o e lapping pe iods (i.e., 1, ,
+ 1). The calcula ion p ocess is as ollows:
RiskTi; ¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1
NX
N
¼1
ðADJ ROAi; 1
NX
N
¼1
ADJ ROAi; Þ2
u
u
jN¼3ð5Þ
ADJ ROAi; ¼EBITDAi;
ASSETSi; 1
XX
X
¼1
EBITDAi;
ASSETSi; ð6Þ
whe e EBITDA
i,
is he ea nings be o e in e es , axes, dep ecia ion, and amo iza ion o fi m iin yea , and
ASSETS
i,
deno es he o al asse s o he en e p ise a he end o he pe iod. Table 11 epo s he es ima ion
esul s o he mechanism es . Fi s , expo ade significan ly inc eases he numbe o employees, inpu o
in e media e p oduc s, ope a ing income and p ofi s o he lis ed companies, expanding he scale o hei p o-
duc ion and sales in e ms o inpu and ou pu . This ealizes economies o scale, educes uni p oduc cos s,
ob ains cos ad an ages, and imp o es en e p ise compe i i eness, he eby p omo ing en e p ise inno a ion.
Second, expo ade significan ly inc eases en e p ises’ isk- aking, educing he isk associa ed wi h R&D
in es men and inc easing echnological inno a ion.
Table 10
T ea men o endogenei y p oblems: ins umen al a iable eg ession.
Fi s s age Second s age
Va iables (1) (2) (3) (4)
T ea
i
Pos
RD
i,
To al
i,
In en
i,
T ea
i
Pos
0.0079** 0.9245*** 1.0550***
(2.411) (2.886) (3.726)
IV 0.0142***
(11.023)
Cons an 0.3797** 0.0240* 7.2671*** 5.1453***
(2.154) (1.951) (5.182) (4.135)
Fi m fixed effec s Yes Yes Yes Yes
Yea fixed effec s Yes Yes Yes Yes
N9005 9005 9005 9005
No e: (1) Con ol a iables a e included in all o he abo e models; (2) T-s a is ics based on obus s anda d e o s clus e ed a he fi m
le el a e epo ed in pa en heses. ***, **, * deno e s a is ical significance a he 1%, 5%, and 10% le els, espec i ely.
D. Zhou e al. / China Jou nal o Accoun ing Resea ch 14 (2021) 207–230 223
4.5.2. He e ogenei y analysis
In ecen yea s, nume ous schola s ha e s udied he e ogeneous ade heo y, acco ding o which a com-
pany’s ype affec s i s expo ade beha io (Jing e al., 2013; Li e al., 2016; Liu and Tong, 2017). Acco d-
ingly, we u he examine he ela ionship be ween expo ade and he echnological inno a ion o he
sampled lis ed companies wi h e e ence o he na u e o owne ship, echnology le el, loca ion, expo des i-
na ion coun y ype, and expo ade ype.
The eg ession esul s a e shown in Table 12. Panel A shows ha expo ade plays a key ole in p omo ing
he inno a ion o s a e-owned en e p ises bu does no significan ly p omo e he inno a ion o p i a e en e -
p ises, and e en has a significan inhibi o y effec on he appea ance design pa en applica ions o p i a e
en e p ises. This indica es ha compa ed wi h p i a e en e p ises, s a e-owned en e p ises ha e lowe p oduc-
ion efficiency and inno a ion efficiency (Wu, 2012; Dong e al., 2014), lea ing mo e oom o imp o e hei
expo lea ning (Jing e al., 2013). Second, s a e-owned en e p ises show a significan ly lowe le el o isk-
aking (Li and Yu, 2012) han p i a e en e p ises do. The e o e, a e s a e-owned en e p ises s a expo ing,
hey ake significan ly mo e isks. Finally, because o hei la ge scale and p o ec ion by he go e nmen (Yu
e al., 2019), s a e-owned en e p ises a e mo e likely han p i a e en e p ises a e o ob ain economies o scale,
making hem be e able o compe e in he in e na ional ma ke . The e o e, engaging in expo ade has a
g ea e posi i e effec on inno a ion abili y o s a e-owned lis ed companies han hei p i a e coun e pa s.
The inno a ion beha io o en e p ises may be affec ed by indus y diffe ences. P e ious s udies ha e shown
ha expo ade p omo es echnological inno a ion mainly o en e p ises in high- ech indus ies and
medium–high- ech indus ies. I has no significan posi i e effec —and may e en ha e a nega i e effec —on
he echnological inno a ion beha io o en e p ises in low- ech indus ies (Li e al., 2016). Following p e ious
s udies, based on he High- ech Indus y (Manu ac u ing) Classifica ion (2017) and High- ech Indus y (Se -
ice Indus y) Classifica ion (2018)
7,8,9
issued by China’s Na ional Bu eau o S a is ics, we classi y he sam-
pled en e p ises as high- ech o non-high- ech o explo e he impac o expo ade on he echnological
inno a ion o lis ed companies om he pe spec i e o echnological de elopmen le el. The eg ession esul s
a e shown in Table 12, Panel B. The es ima ion esul s show ha expo ade has a significan posi i e impac
on he echnological inno a ion o non-high- ech lis ed companies. The possible easons a e as ollows. Fi s ,
non-high- ech en e p ises ha e a lowe echnological s a ing poin han high- ech en e p ises do and hus
mo e oom o imp o emen . Second, compa ed wi h non-high- ech en e p ises, high- ech en e p ises engag-
ing in expo ade a e mo e ulne able o echnological blockades by Wes e n de eloped coun ies. Thi d,
high- ech en e p ises need o in es hea ily in R&D o suppo hei high- ech ac i i ies. P o iding p o ec ion
o high- isk R&D ac i i ies i sel incu s conside able isk. The e o e, he effec o expo ade on high- ech
en e p ises’ le el o isk- aking is no p onounced. Fou h, because o hei high echnical le el, high- ech
7
The ele an documen s a e a ailable om he Na ional Bu eau o S a is ics.
8
The High- ech Indus y (Manu ac u ing) Classifica ion (2017) is a ailable a h p://www.s a s.go .cn/ jsj/ jbz/201812/ 20181218_
1640081.h ml.
9
The High- ech Indus y (Se ice Indus y) Classifica ion (2018) is a ailable a h p://www.s a s.go .cn/ jsj/ jbz/201805/ 20180509_
1598315.h ml.
Table 11
Expo ade and co po a e inno a ion: mechanism es .
Va iables (1) (2)
Employee MI Sales P ofi RiskT
T ea
i
Pos
0.0737** 0.0631** 0.0430* 0.0865* 0.0058***
(2.109) (2.297) (1.762) (1.952) (2.704)
Fi m fixed effec s Yes Yes Yes Yes Yes
Yea fixed effec s Yes Yes Yes Yes Yes
N9799 9557 9827 8836 8798
Adj R
2
0.504 0.890 0.924 0.725 0.715
No e: (1) Con ol a iables a e included in all o he abo e models; (2) T-s a is ics based on obus s anda d e o s clus e ed a he fi m
le els a e epo ed in pa en heses. ***, **, * deno e s a is ical significance a he 1%, 5%, and 10% le els, espec i ely.
224 D. Zhou e al. / China Jou nal o Accoun ing Resea ch 14 (2021) 207–230
en e p ises can es ablish ce ain echnical ba ie s in he domes ic ma ke o ob ain economies o scale and
educe he ma ginal cos o hei p oduc s. The e o e, as high- ech en e p ises al eady ha e a ce ain deg ee
o economies o scale, engaging in expo ade has li le subs an ial impac on hei economies o scale. Based
on hese poin s, expo ade is pa icula ly beneficial o non-high- ech en e p ises in enhancing hei inno a-
ion abili y.
In addi ion, egions a e o en he e ogeneous in hei le el o expo ade and/o le el o en e p ise inno-
a ion. The e o e, acco ding o he egions in which hey a e loca ed, we classi y he lis ed companies as ei he
en e p ises in eas e n China o en e p ises in cen al o wes e n China. The eg ession esul s a e shown in
Table 12
Expo ade and co po a e inno a ion: he e ogenei y es .
Panel A:F om he pe spec i e o na u e o owne ship
Va iables S a e-owned Non-s a e-owned
(1)(2)(3)(4)(5)(6)
RD
i,
To al
i,
In en
i,
RD
i,
To al
i,
In en
i,
T ea
i
Pos
0.0075** 0.8660*** 0.8365*** 0.0049 0.0783 0.3187
(2.217) (3.293) (3.497) (1.272) (0.267) (1.203)
Cons an 0.0331* 8.7731*** 6.3988*** 0.0108 4.5509* 4.9028**
(1.789) (3.133) (2.692) (0.275) (1.729) (2.082)
N3240 3240 3240 2039 2039 2039
Adj R
2
0.124 0.322 0.299 0.133 0.190 0.216
Panel B:F om he pe spec i e o echnology le el
Va iables High- ech Non-high- ech
(1)(2)(3)(4)(5)(6)
RD
i,
To al
i,
In en
i,
RD
i,
To al
i,
In en
i,
T ea
i
Pos
0.0059 0.2443 0.0801 0.0065*** 0.8495*** 0.8649***
(0.919) (0.907) (0.289) (3.277) (3.292) (3.840)
Cons an 0.0459 10.8588*** 13.2561*** 0.0192 4.8308** 3.6171**
(0.668) (2.714) (3.505) (1.557) (2.561) (2.406)
N1295 1295 1295 4106 4106 4106
Adj R
2
0.254 0.374 0.363 0.082 0.251 0.240
Panel C:F om he pe spec i e o loca ion
Va iables Cen al/eas e n egion Wes e n egion
(1) (2) (3) (4) (5) (6)
RD
i,
To al
i,
In en
i,
RD
i,
To al
i,
In en
i,
T ea
i
Pos
0.0083*** 0.6483*** 0.7295*** 0.0006 0.1591 0.4126
(3.091) (2.768) (3.572) (0.186) (0.334) (0.897)
Cons an 0.0083 6.6887*** 5.3947*** 0.0325* 6.3458* 5.1467*
(0.435) (2.932) (2.658) (1.654) (1.830) (1.695)
N4297 4297 4297 1104 1104 1104
Adj R
2
0.111 0.268 0.259 0.229 0.288 0.263
Panel D:F om he pe spec i e o expo des ina ion coun y ype
Va iables De eloped coun ies De eloping coun ies
(1)(2)(3)(4)(5)(6)
RD
i,
To al
i,
In en
i,
RD
i,
To al
i,
In en
i,
T ea
i
Pos
0.0141** 0.7506 0.8168* 0.0065*** 0.6153*** 0.7120***
(2.053) (1.326) (1.753) (2.968) (3.053) (4.025)
Cons an 0.0229 5.2135*** 4.1470** 0.0156 5.7330*** 4.7335***
(1.452) (2.585) (2.356) (0.990) (2.997) (2.810)
N4887 4887 4887 5493 5493 5493
Adj R
2
0.086 0.232 0.218 0.111 0.259 0.248
D. Zhou e al. / China Jou nal o Accoun ing Resea ch 14 (2021) 207–230 225
Table 12, Panel C. The es ima ion esul s show ha expo ade mainly p omo es he echnological inno a-
ion o en e p ises in cen al and eas e n China; i does no significan ly p omo e he echnological inno a ion
o en e p ises in wes e n China. This indica es, fi s , ha he de elopmen o mode n indus y in he wes e n
egion is slow, lacking p ecision p oduc ion equipmen , an app op ia e financial en i onmen , and scien ific
and echnological human esou ces. The e o e, a e engaging in expo ade, e en i wes e n Chinese en e -
p ises ha e lea ned om in e na ional ad anced echnology, he ans o ma ion efficiency o hei inno a ion
achie emen s is ela i ely low because hey lack he en i onmen and human esou ces conduci e o inno a-
ion. Second, because wes e n China is geog aphically a om he main expo ma ke s o Eu ope and he U.
S., he cos o en e ing hese expo ma ke s is highes o wes e n Chinese en e p ises (Tong and Liu, 2014).
The e o e, he expo cos s incu ed by hese en e p ises a e ela i ely high and hei esul ing expo p ofi s
a e ela i ely low, esul ing in ela i ely li le in es men in inno a ion.
Fu he mo e, p e ious s udies ha e shown ha newly indus ialized economies and de eloping coun ies
can imp o e hei independen inno a ion capabili y (Aw e al., 2000; De Loecke , 2007; Al a ez and
Lopez, 2005) by pa icipa ing in in e na ional compe i ion o lea n om and abso b ad anced echnology
om de eloped coun ies. Howe e , his effec has no been confi med in he as majo i y o de eloped indus-
ialized coun ies (Be na d and Jensen, 1999; 2004a; Wagne , 2002; G eenaway and Yu, 2004). The e o e, he
ole o expo ade in p omo ing co po a e inno a ion may a y be ween coun ies based on hei expo des-
ina ion ype.
10
The e o e, acco ding o he di ision o he Human De elopmen Index
11
, we classi y he
expo des ina ion coun ies o he sampled lis ed companies as de eloped o de eloping economies o conduc
compa a i e analysis. As shown in Table 12, Panel D, expo ade significan ly p omo es he inno a ion le el
o en e p ises in de eloped coun ies bu only sligh ly p omo es he inno a ion ou pu o en e p ises in de el-
oping coun ies. This indica es ha expo ade has a significan posi i e effec on he inno a ion le el o
en e p ises ha expo o de eloped coun ies, bu only sligh ly p omo es he inno a ion ou pu o en e p ises
ha expo o de eloping coun ies. The mo e de eloped an en e p ise’s expo des ina ion coun y, he highe
he en e p ise’s deg ee o inno a ion knowledge, esul ing in highe en e p ise inno a ion.
Finally, acco ding o Dai e al. (2014), p ocessing ade wi h low echnological con en is he main eason
o he p oduc i i y pa adox obse ed among Chinese en e p ises. P ocessing ade en e p ises may no expe-
ience lea ning by expo ing effec s (Xiang and Ma, 2013). The e o e, he p omo ing effec o expo ade on
en e p ise inno a ion may a y wi h he mode o expo ade. The e o e, ollowing Li e al. (2016), we dis-
inguish be ween modes o ade o u he analysis.
12
The eg ession esul s a e shown in Table 12, Panel E.
Fi s , expo ade significan ly p omo es he inno a ion o en e p ises ha a e mainly engaged in gene al
10
We hank an anonymous e e ee o se e al aluable commen s on he e ogenei y analysis ha helped o g ea ly imp o e he pape .
11
Human De elopmen Index is a ailable a h p://hd .undp.o g/si es/de aul /files/hd 2019.pd . (p. 300–311).
12
Following Li e al. (2016), i he p opo ion o p ocessing ade o hei o al expo ade is less han 25%, we classi y expo ing
en e p ises as mainly engaged in gene al ade. I he p opo ion o p ocessing ade o hei o al expo ade is mo e han 75%, we
classi y expo ing en e p ises as mainly engaged in p ocessing ade. We classi y he o he s as mainly engaged in mixed ade.
Panel E: F om he pe spec i e o expo ade ype
Va iables Gene al ade P ocessing and mixed ade
(1)(2)(3)(4)(5)(6)
RD
i,
To al
i,
In en
i,
RD
i,
To al
i,
In en
i,
T ea
i
Pos
0.0087*** 0.4736*** 0.6013*** 0.0135 0.5030 0.5881
(2.769) (2.909) (3.984) (1.386) (1.278) (1.640)
Cons an 0.0078 6.4110*** 4.6439*** 0.0252 5.4628** 4.8340**
(0.572) (3.839) (3.172) (1.217) (2.254) (2.315)
N6318 6318 6318 3895 3895 3895
Adj R
2
0.103 0.231 0.222 0.064 0.170 0.161
(1) To imp o e he obus ness o he esul s, he samples used in he abo e models we e PSM pai ed a e excluding lis ed companies ha
expo bu hen exi he expo ma ke . (2) Con ol a iables a e included in all o he abo e models. (3) T-s a is ics based on obus
s anda d e o s clus e ed a he fi m le el a e epo ed in pa en heses. ***, **, * deno e s a is ical significance a he 1%, 5%, and 10%
le els, espec i ely.
226 D. Zhou e al. / China Jou nal o Accoun ing Resea ch 14 (2021) 207–230
ade, bu has no significan impac on he inno a ion le el o en e p ises mainly engaged in p ocessing and
mixed ade. This indica es ha , unlike en e p ises mainly engaged in gene al ade, en e p ises mainly
engaged in p ocessing and mixed ade do no expe ience he expo lea ning effec , consis en wi h p e ious
findings.
5. Conclusion
The pe o mance o expo ade a he mic o le el o en e p ises has long been a ho opic o esea ch on
in e na ional ade. Based on da a on Shanghai and Shenzhen A-sha e lis ed companies om 2000 o 2015, we
use a mul i-DID model o s udy he ela ionship be ween expo ade and echnological inno a ion. The find-
ings indica e ha expo ade p omo es he echnological inno a ion o lis ed companies in e ms o bo h
inno a ion inpu and inno a ion ou pu . F om he pe spec i e o pa en ou pu , expo ade significan ly p o-
mo es lis ed companies’ applica ions o in en ion pa en s and u ili y model pa en s wi h a high echnical con-
en . The conclusions epo in his pape emain obus a e a se ies o obus ness es s. Expo ade
beha io gene ally leads en e p ises o expand hei p oduc ion scale, inc ease hei inpu o p oduc ion ac-
o s o achie e economies o scale, and inc ease hei isk- aking, he eby aising hei le el o echnological
inno a ion. In e ms o en e p ise he e ogenei y, expo ade mainly p omo es he echnological inno a ion
o s a e-owned en e p ises, non-high- ech en e p ises, en e p ises in cen al and eas e n China, en e p ises ha
expo o de eloped economies, and en e p ises ha a e mainly engaged in gene al ade.
A he ime o w i ing, he ade nego ia ions esul ing om Sino–U.S. ade ic ions ha e eached an
impasse. The U.S. launches a ade wa agains China p ima ily o es ain he de elopmen o China’s
high- ech en e p ises by imposing puni i e a iffs o cu b China’s scien ific and echnological inno a ion
(Huang e al., 2018a; 2018b). Gi en he U.S.’s ze o-sum hinking, i.e., seeking o main ain i s s a us as a g ea
powe while cu bing China’s de elopmen , he ade conflic is complex and likely o be long-las ing (Yu e al.,
2018a; 2018b). Since he U.S.’s announcemen o a iffs agains China Sino-US ade ic ion on Ma ch 22,
2018
13
, Chinese en e p ises’ olume o expo ade o he U.S. has dec eased significan ly, and he alue-
inno a ion chain o Chinese expo en e p ises has been damaged (Young, 1991; Gibbon e al., 2008). Today,
wi h apid economic globaliza ion, na ional s eng h is ul ima ely measu ed by he capaci y o inno a ion. A
p esen , China is on he b ink o en e ing he anks o he wo ld’s mos inno a i e coun ies. The conclusions
o his pape a e o p ac ical significance o Chinese en e p ises seeking o s eng hen hei expo ade and
imp o e hei scien ific and echnological inno a ion agains he backg ound o Sino–U.S. ade ic ions.
This may in u n help China o eme ge as an inno a i e coun y by b eaking down he scien ific and echno-
logical blockade e ec ed by he U.S. and elimina ing i s high- ech dependence on he U.S.
Decla a ion o Compe ing In e es
The au ho decla e ha he e is no conflic o in e es .
Re e ences
Al a ez, R., Lopez, R.A., 2005. Expo ing and pe o mance: e idence om Chilean plan s. J. Can. J. Econ. 38 (4), 1384–1400.
A nold, J.M., Hussinge , K., 2005. Expo beha io and fi m p oduc i i y in Ge man manu ac u ing: a fi m-le el analysis. J. Re . Wo ld
Econ. 141 (2), 219–243.
Aw, B.Y., Chung, S., Robe s, M.J., 2000. P oduc i i y and u no e in he expo ma ke : mic o-le el e idence om he Republic o
Ko ea and Taiwan (China). J. The Wo ld Bank Economic Re iew 14 (1), 65–90.
Bae , M., 2012. Pu ing c ea i i y o wo k: he implemen a ion o c ea i e ideas in o ganiza ions. J. Acad. Manage. J. 55 (5), 1102–1119.
Be na d, A.B., Jensen, J.B., 1999. Excep ional expo e pe o mance: cause, effec , o bo h? J. In . Econ. 47, 1–25.
Be na d, A.B., Jensen, J.B., 2004a. En y, expansion, and in ensi y in he US expo boom, 1987–1992. J. Re . In . Econ. 12 (4), 662–675.
Be na d, A.B., Jensen, J.B., 2004b. Why some fi ms expo . J. Re . Econ. S a is . 86 (2), 561–569.
13
On Ma ch 22, 2018, he U.S. go e nmen plans o impose a iffs o up o 25% on China’s impo s o US $50 billion, which is ega ded
as he beginning o Sino-US ade ic ion (Huang e al., 2018).
D. Zhou e al. / China Jou nal o Accoun ing Resea ch 14 (2021) 207–230 227
Boubak i, N., Cosse , J.C., Saffa , W., 2013. The ole o s a e and o eign owne s in co po a e isk- aking: e idence om p i a iza ion. J.
Financ. Econ. 108 (3), 641–658.
Cle ides, S.K., Lach, S., Tybou , J.R., 1998. Is lea ning by expo ing impo an ? Mic o-dynamic e idence om Colombia, Mexico, and
Mo occo. J. Qua . J. Econ. 113 (3), 903–948.
De Loecke , J., 2007. Do expo s gene a e highe p oduc i i y? e idence om Slo enia. J. In . Econ. 73 (1), 69–98.
Faccio, M., Ma chica, M.T., Mu a, R., 2011. La ge sha eholde di e sifica ion and co po a e isk- aking. J. Re . Financ. S ud. 24 (11),
3601–3641.
Fa uq, H., 2010. Impac o echnology and physical capi al on expo quali y. J. De . A eas 44 (1), 167–185.
Flo ida, R., 1994. Agglome a ion and indus ial loca ion: an econome ic analysis o Japanese-affilia ed manu ac u ing es ablishmen s in
au omo i e- ela ed indus ies. J. U ban Econ. 36 (1), 23–41.
Ge effi, G., Hunph ey, J., S u geon, T., 2005. The go e nance o global alue chains. J. Re . In . Poli . Econ. 12 (1), 78–104.
Gibbon, P., Bai , J., Pon e, S., 2008. Go e ning global alue chains: an in oduc ion. J. Econ. Soc. 37 (3), 315–338.
Gi ma, S., Go
¨ g, H., S obl, E., 2004. Expo s, in e na ional in es men , and plan pe o mance: e idence om a non-pa ame ic es . J.
Econ. Le . 83 (3), 317–324.
G eenaway, D., Yu, Z., 2004. Fi m-le el in e ac ions be ween expo ing and p oduc i i y: Indus y-specific e idence. J. Re . Wo ld Econ.
140 (3), 376.
G eenaway, D., Knelle , R., 2007. Fi m he e ogenei y, expo ing and o eign di ec in es men . J. Econ. J. 117 (517), 134–161.
G ossman, G.M., Helpman, E., 1990. T ade, inno a ion, and g ow h. J. Am. Econ. Re . 80 (2), 86–91.
Holms om, B., 1989. Agency cos s and inno a ion. J. Econ. Beha . O g. 12 (3), 305–327.
Howell, S.T., 2017. Financing inno a ion: e idence om R&D g an s. J. Am. Econ. Re . 107 (4), 1136–1164.
Huang, Y., Lin, C., Liu, S., Tang, H., 2018. T ade linkages and fi m alue: E idence om he 2018 US-China ade wa . R. G adua e
Ins i u e o In e na ional and De elopmen S udies Wo king Pape No. 11-2018.
John, K., Li o , L., Yeung, B., 2008. Co po a e go e nance and isk- aking. J. Finance 63 (4), 1679–1728.
Kelle , W., 2010. In e na ional ade, o eign di ec in es men , and echnology spillo e s. M. Handbook o he Economics o Inno a ion
2, 793-829. No h-Holland.
K ugman, P.R., 1980. Scale economies, p oduc diffe en ia ion, and he pa e n o ade. J. Am. Econ. Re . 70 (5), 950–959.
La Po a, R., Lopez-de-Silanes, F., Shlei e , A., 1999. Co po a e owne ship a ound he wo ld. J. J. Financ. 54 (2), 471–517.
Lilee a, A., T efle , D., 2010. Imp o ed access o o eign ma ke s aises plan -le el p oduc i i y o some plan s. J. Qua . J. Econ. 125 (3),
1051–1099.
Ling, Y., Simsek, Z., Luba kin, M.H., Veiga, J.F., 2008. T ans o ma ional leade ship’s ole in p omo ing co po a e en ep eneu ship:
examining he CEO-TMT in e ace. J. Acad. Manage. J. 51 (3), 557–576.
Lu, D., 2010. Excep ional expo e pe o mance? E idence om Chinese manu ac u ing fi ms. J. Manusc ip , Uni e si y o Chicago.
Meli z, M.J., 2003. The impac o ade on in a-indus y ealloca ions and agg ega e indus y p oduc i i y. J. Econome ica 71 (6), 1695–
1725.
Nielsen, L., 2013. How o classi y coun ies based on hei le el o de elopmen . J. Social Indica . Res. 114 (3), 1087–1107.
Salomon, R.M., Sha e , J.M., 2005. Lea ning by expo ing: new insigh s om examining fi m inno a ion. J. Econ. Manage. S a . 14 (2),
431–460.
Schumpe e , J.A., 1942. Capi alism, socialism, and democ acy. J. Am. Econ. Re . 3 (4), 594–602.
Teece, D., Pisano, G., Shuen, A., 1997. Dynamic capabili ies and s a egic managemen . J. S a egic Manage. J. 18 (7), 509–533.
Wagne , J., 2002. The causal effec s o expo s on fi m size and labo p oduc i i y: fi s e idence om a ma ching app oach. J. Econ. Le .
77 (2), 287–292.
Wagne , J., 2007. Expo s and p oduc i i y: a su ey o he e idence om fi m-le el da a. J. Wo ld Econ. 30 (1), 60–82.
Wu, S., Le i as, E., P iem, R.L., 2005. CEO enu e and company in en ion unde diffe ing le els o echnological dynamism. J. Acad.
Manage. J. 48 (5), 859–873.
Young, A., 1991. Lea ning by doing and he dynamic effec s o in e na ional ade. J. Qua . J. Econ. 106 (2), 369–405.
Bao, Q., Ye, N., Shao, M., 2014. Expo lea ning, he e ogeneous ma ching and dynamic changes o fi m p oduc i i y. J. Wo ld Econ. 37
(4), 26–48 (in Chinese).
Chen, L., Wan, P., Xu, Y., 2019. The owne ship s uc u e and inno a ion beha io o mixed owne ship en e p ises – based on he
empi ical es o na u al expe imen and eg ession discon inui y. J. Managemen Wo ld 35 (10), 186–205 (in Chinese).
Chen, Y., Li, Y., Zhou, S., 2012. Expo du a ion o Chinese en e p ises and i s de e minan s. J. Econ. Res. J. 47 (7), 48–61 (in Chinese).
Chen, Z., 2002. The effec i e pa h o mode n en e p ise cos con ol – A case s udy o cos con ol o Jinling G oup Plas ic Fac o y. J.
Managemen Wo ld 12, 128–133 (in Chinese).
Chi, R., 2002. Compa a i e analysis o echnological inno a ion o diffe en scale en e p ises. J. So Science 4, 84–88 (in Chinese).
Dai, M., Yu, M., 2012. R&D inpu , expo and p oduc i i y imp o emen be o e expo : e idence om Chinese manu ac u ing
en e p ises. J. China Econ. Qua . 11 (1), 211–230 (in Chinese).
Dai, M., Yu, M., Mai a, M., 2014. The mys e y o p oduc i i y o Chinese expo companies: he ole o p ocessing ade. J. China Econ.
Qua . 13 (2), 675–698 (in Chinese).
Dai, X., Zheng, L., 2015. How does ins i u ional quali y affec China’s climbing global alue chain? J. In . T ade 12, 51–63 (in Chinese).
Dong, X., Zheng, J., Yuan, P., 2014. Resea ch on he loss o inno a ion efficiency o s a e owned en e p ises. J. China Indus ial
Economics 2, 97–108 (in Chinese).
Fan, J., Feng, M., 2013. The puzzle o he p oduc i i y pa adox o China’s manu ac u ing expo en e p ises: a es based on he diffe ence
o expo densi y. J. Manage. Wo ld 8, 16–29 (in Chinese).
228 D. Zhou e al. / China Jou nal o Accoun ing Resea ch 14 (2021) 207–230
Feng, H., Liu, S., 2017. Income ax, he e ogenei y o isk sha ing and inno a ion. J. China Indus ial Econ. 8, 138–155 (in Chinese).
Gu, X., Chen, Y., Pan, S., 2018. Economic policy unce ain y and inno a ion: An empi ical analysis based on Chinese lis ed companies. J.
Econ. Res. J. 53 (2), 109–123 (in Chinese).
Guo, Y., 2018. Signal ansmission mechanism o go e nmen inno a ion subsidy and en e p ise inno a ion. J. China Indus ial
Economics 9, 98–116 (in Chinese).
Han, F., Ke, S., 2012. T acing he spa ial sou ces o China’s manu ac u ing agglome a ion: based on he comp ehensi e pe spec i e o
Ma shall ex e nali y and new economic geog aphy. J. Manage. Wo ld 10, 55–70 (in Chinese).
He, H., Zhang, J., Zhang, T., 2020. China’s o eign an i-dumping and co po a e inno a ion: empi ical esea ch om co po a e pa en
da a. J. Finance Econ. 46 (2), 4–20 (in Chinese).
He, Y., Yu, W., Dai, Y., 2019. Senio manage s’ p o essional expe ience and en e p ise inno a ion. J. Managemen Wo ld 35 (11), 174–
192 (in Chinese).
Huang, L., Wang, J., Meng, X., 2018b. Rebalancing o economic globaliza ion and Sino US ade ic ion. J. China Indus ial Economics
10, 156–174 (in Chinese).
Jing, X., Liu, Z., Yu, W., 2012. Resea ch on he dynamic effec s o en e p ise expo – expe ience om Chinese manu ac u ing en e p ises:
2001–2007. J. China Economic Qua e ly 11 (3), 1097–1112 (in Chinese).
Jing, F., Tao, P., Gao, Y., 2013. Do Chinese companies ha e an expo lea ning effec ? – an empi ical s udy based on he pe spec i e o
owne ship s uc u e. J. Wo ld Economy S udies 3, 41–47 (in Chinese).
Li, W., Zheng, M., 2016. Subs an ial inno a ion o s a egic inno a ion? – The impac o mac o-indus ial policies on mic o-en e p ise
inno a ion. J. Econ. Res. J. 51 (4), 60–73 (in Chinese).
Li, B., Yue, Y., Chen, T., 2016. Expo and independen echnological inno a ion o en e p ises: empi ical esea ch om en e p ise pa en
da a. J. The Jou nal o Wo ld Economy 39 (12), 72–94 (in Chinese).
Li, C., Zhao, M., 2010. Does expo ade inc ease he p oduc i i y o Chinese en e p ises? – A es based on he da a o Chinese
manu ac u ing en e p ises in 2007. J. Financ. Econ. 36 (4), 14–24 (in Chinese).
Li, C., 2010. Is he e a ‘‘p oduc i i y pa adox”in Chinese expo companies?:A es based on da a om Chinese manu ac u ing
companies. J. Wo ld Econ. 33 (7), 64–81 (in Chinese).
Li, L., Wang, C., 2017. Tax bu den, inno a ion capabili y, and en e p ise upg ading: empi ical e idence om companies lis ed on he New
Thi d Boa d. J. Econ. Res. J. 52 (11), 119–134 (in Chinese).
Li, P., 2002. The ela ionship be ween in e na ional ade and echnological inno a ion. J. Wo d Economy S udies 5, 79–83 (in Chinese).
Li, W., Yu, M., 2015. The owne ship s uc u e o p i a ized en e p ises and en e p ise inno a ion. J. Manage. Wo ld 4, 112–125 (in
Chinese).
Li, W., Yu, M., 2012. Na u e o owne ship, ma ke iza ion p ocess and en e p ise isk- aking. J. China Ind. Econ. 12, 115–127 (in Chinese).
Li, X., Zheng, X., Zhang, J., 2018. Will he pe o mance dilemma o manu ac u ing companies p omo e inno a ion – an analysis based on
he expansion o he expec ed gap dimension. J. China Ind. Econ. 8, 174–192 (in Chinese).
Li, X., Zhu, Z., 2006. In e na ional ade, R&D spillo e s and p oduc i i y g ow h. J. Econ. Res. J. 2, 31–43 (in Chinese).
Lin, Z., Lin, H., Deng, X., 2013. Income ax e o m and echnological inno a ion o Chinese en e p ises. J. China Ind. Econ. 3, 111–123
(in Chinese).
Liu, Z., Tong, J., 2017. Fac o ma ke dis o ion, he e ogenei y ac o s and he expo -p oduc i i y ela ionship o Chinese en e p ises. J.
Wo ld Econ. 40 (12), 76–97 (in Chinese).
L , D., Shen, K., Jian, Z., 2016. Re es o ‘‘expo lea ning effec ”: Empi ical analysis based on ade ype. J. Econ. Re . 02, 124–136 (in
Chinese).
Mao, Q., Sheng, B., 2014. T ade libe aliza ion and Chinese fi ms’expo beha io : Dose WTO en y acili a e expo pa icipa ion? J.
China Econ. Qua . 13 (02), 647–674 (in Chinese).
Mao, Q., Xu, J., 2015. The impac o go e nmen subsidies on en e p ises’ new p oduc inno a ion – based on he pe spec i e o
‘‘mode a e ange”o subsidy in ensi y. J. China Indus ial Econ. 6, 94–107 (in Chinese).
Mao, Y., Wang, J., 2006. Resea ch on independen inno a ion pa h based on p oduc upg ading. J. Manage. Wo ld 5, 114–120 (in
Chinese).
Meng, Q., Li, X., Zhang, P., 2019. Can ESOP p omo e en e p ise inno a ion? – empi ical E idence om he pe spec i e o employees. J.
Manage. Wo ld 35 (11), 209–228 (in Chinese).
Pan, Y., Pan, J., Dai, Y., 2015. Co po a e li iga ion isk, judicial local p o ec ionism and en e p ise inno a ion. J. Econ. Res. J. 50 (3),
131–145 (in Chinese).
Qian, X., Wang, J., Huang, Y., Wang, S., 2011. Expo and p oduc i i y o Chinese indus ial en e p ises – Sel selec ion effec o expo
lea ning effec ? J. J. Quan i . Tech. Econ. 28 (2), 37–51 (in Chinese).
Qiu, B., Liu, X., Zhao, W., 2012. Expo lea ning o sel selec ion: E idence om diffe ence-in-diffe ence app oaches o China’s
manu ac u ing mic o en e p ises based on p opensi y sco e ma ching. J. J. Wo ld Econ. 35 (4), 23–40 (in Chinese).
Sun, J., 2017. Replacing business ax wi h VAT, en e p ise p ofi a e and en e p ise inno a ion. J. Manage. Wo ld 11, 180–181 (in
Chinese).
Tao, A., Wu, W., Kuai, P., 2020. Does impo and expo ade inhibi en e p ise inno a ion – based on he adjus men o income gap. J.
J. In . T ade 3, 116–130 (in Chinese).
Tong, J., Liu, Z., 2014. Geog aphical agglome a ion and expo choice o en e p ises: A s udy based on he dependence o o eign capi al
financing. J. Wo ld Economy 37 (7), 67–85 (in Chinese).
Wang, H., Xu, H., Yang, J., 2011. Expo , he e ogenei y and fi m p oduc i i y: e idence om Chinese fi ms. J. Finance Econ. 37 (6), 36–
46 (in Chinese).
D. Zhou e al. / China Jou nal o Accoun ing Resea ch 14 (2021) 207–230 229
Wu, Y., 2012. The dual efficiency losses in Chinese s a e-owned en e p ises. J. Econ. Res. J. 47 (3), 15–27 (in Chinese).
Xiang, S., Ma, W., 2013. Whe he he expo en e p ises ha e lea ning-by-expo ing effec : an empi ical analysis based on China’s
mic ofi ms da a. J. Wo ld Econ. S ud. 10, 37–42 (in Chinese).
Xu, J., Mao, Q., Tong, J., 2015. How does expo affec he isk- aking abili y o en e p ises? J. Indus ial Econ. Res. 2, 1–14 (in Chinese).
Yang, Y., Wei, J., Luo, L., 2015. Who inno a e om he go e nmen subsidies? – he join mode a e effec s o he owne ship and he
ac o ma ke dis o ions. J. Manage. Wo ld 01, 75–86 (in Chinese).
Yi, J., Fu, J., 2011. En e p ise p oduc i i y and expo : E idence om en e p ise le el in Zhejiang P o ince. J. The Jou nal o Wo ld
Economy 34 (5), 74–92 (in Chinese).
Yu, M., Fan, R., Zhong, H., 2016a. Indus ial policy and echnological inno a ion o Chinese en e p ises. J. China Ind. Econ. 12, 5–22 (in
Chinese).
Yu, M., Li, W., Pan, H., 2013. Manage ial o e confidence and en e p ise isk aking. J. Financ. Res. 1, 149–163 (in Chinese).
Yu, M., Zhong, H., Fan, R., 2019. P i a iza ion, financial cons ain s, and co po a e inno a ion – e idence om China’s indus ial
en e p ises. J. Financ. Res. 4, 75–91 (in Chinese).
Yu, M., Zhong, H., Fan, R., 2016b. Does he new pe o mance app aisals (EVA) p omo e inno a ion o cen al go e nmen -owned
en e p ises (CGOEs)? J. Econ. Res. J. 51 (12), 104–117 (in Chinese).
Yu, Z., Zhou, B., Xie, X., Wang, Z., 2018a. Pa icipa ion in he econs uc ion o he GVCs and Sino-U.S. ade ic ions. J. China
Indus ial Economics 7, 24–42 (in Chinese).
Yu, Y., Zhao, Q., Ju, X., 2018b. In en o execu i es and inno a ion. J. China Indus ial Economics 3, 136–154 (in Chinese).
Zhang, D., Zheng, X., 2013. Is house pu chase limi an effec i e policy o con ol housing p ice. J. The Jou nal o Quan i a i e and
Technical. Economics 30 (11), 56–72 (in Chinese).
Zhang, J., Li, Y., Liu, Z., 2009. Does expo p omo e he p oduc i i y o Chinese companies? – Empi ical e idence om Chinese local
manu ac u ing companies: 1999–2003. J. Manage. Wo ld 12, 11–26 (in Chinese).
Zhang, J., Li, Y., Liu, Z., 2008. Expo s and he p oduc i i y o Chinese local en e p ises – based on he empi ical analysis o Jiangsu
manu ac u ing en e p ises. J. Manage. Wo ld 11, 50–64 (in Chinese).
Zhang, J., Zhang, F., Chen, Z., 2016. A e-examina ion o he ela ionship be ween expo and p oduc i i y: new mic o e idence om
China. J. Wo ld Econ. 39 (6), 54–76 (in Chinese).
Zhang, X., Zhang, W., Cui, K., 2018. Imbalance o ade benefi dis ibu ion and ade ic ion – also on he solu ion o Sino-US ade
ic ion. J. In e ade 10, 52–57 (in Chinese).
Zhao, W., Li, S., 2007. Expo and en e p ise p oduc i i y: he la es de elopmen o empi ical heo y. J. In . T ade 7, 24–30 (in Chinese).
Zhou, D., Huang, J., Zhao, Y., 2019. Employee s ock owne ship plans and co po a e inno a ion. J. Accoun . Res. 3, 63–70 (in Chinese).
230 D. Zhou e al. / China Jou nal o Accoun ing Resea ch 14 (2021) 207–230