Yin, Xingmin; Jin, Yun; Li, Zhou; Liu, Yulin
A icle
How does open inno a ion p omo e ci cula economy
p ac ices? E idence om Chinese lis ed companies
Jou nal o Inno a ion & Knowledge (JIK)
P o ided in Coope a ion wi h:
Else ie
Sugges ed Ci a ion: Yin, Xingmin; Jin, Yun; Li, Zhou; Liu, Yulin (2025) : How does open inno a ion
p omo e ci cula economy p ac ices? E idence om Chinese lis ed companies, Jou nal o Inno a ion
& Knowledge (JIK), ISSN 2444-569X, Else ie , Ams e dam, Vol. 10, Iss. 3, pp. 1-13,
h ps://doi.o g/10.1016/j.jik.2025.100702
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/327603
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/
How does open inno a ion p omo e ci cula economy p ac ices? E idence
om Chinese lis ed companies
Xingmin Yin
a
, Yun Jin
a
, Zhou Li
b
, Yulin Liu
c,*
a
School o Economics, Ocean Uni e si y o China, Qingdao 266100, China
b
School o In e na ional Business and Managemen , Sichuan In e na ional S udies Uni e si y, Chongqing 400031, China
c
Cen e o Public Economy & Public Policy Resea ch, Chongqing Uni e si y, Shazheng S ee , Shapingba Dis ic , Chongqing 400044, China
ARTICLE INFO
JEL classi ica ion codes:
C30
E60
E70
Q5
O36
Keywo ds:
Open inno a ion
Ci cula economy p ac ices
T ade c edi
G een pa en s
Ci cula economy policies
ABSTRACT
Al hough i ms con end wi h mul iple challenges in implemen ing ci cula economy (CE) p ac ices, open inno-
a ion (OI) can be an e ec i e collabo a i e s a egy o i ms o o e come echnological and in o ma ional
ba ie s and achie e CE objec i es. Using da a om A-sha e and Science and Technology Inno a ion Boa d
(STAR Ma ke ) lis ed companies in China om 2003 o 2022, we employ a wo-way ixed e ec s model o
examine he impac o OI on i ms’ CE p ac ices. The empi ical esul s indica e ha OI signi ican ly p omo es CE
p ac ices, in luencing hese p ac ices h ough mechanisms such as ade c edi p o ision, g een pa en accu-
mula ion, and enhanced in o ma ion exchange. Mo eo e , go e nmen CE policies can s eng hen he e ec o OI
on CE p ac ices. Addi ionally, in e company OI has a s onge posi i e impac on CE p ac ices compa ed wi h
in acompany and coope a i e OI. The e ec s o OI and he mode a ing in luence o CE policies a e mo e p o-
nounced in he eas e n egion, which is likely d i en by supe io indus ializa ion in he egion. These indings
cla i y he ela ionship be ween OI and CE p ac ices o ad ance global CE p ac ices. Fu u e esea ch could
explo e b oade da ase s, adop al e na i e OI indica o s, and conduc c oss-coun y compa isons o u he
expand his esea ch.
In oduc ion
As ex eme clima e impac s economic, social, and en i onmen al
de elopmen , coun ies a e inc easingly explo ing new g een economic
models o a oid he high emissions and pollu ion challenges o adi-
ional c ude economic g ow h models (Yuan & Pan, 2023; Yuana e al.,
2024). The ci cula economy (CE) concep is based on he p inciple o
“ educe, euse, and ecycle” (Geissdoe e e al., 2017), meaning ha
goods a e no disca ded a e single use bu ecycled and eused as much
as possible (S ucki e al., 2023). I is widely acknowledged ha majo
economies wo ldwide gene ally conside CE de elopmen as a possible
app oach o p omo ing pollu ion p e en ion and comp ehensi e was e
u iliza ion o achie e mu ually bene icial ou comes o he en i onmen
and he economy. Conside ing ha China is one o he mos populous
coun ies in he wo ld, he na ion aces eno mous esou ce p essu e and
en i onmen al challenges. The Chinese go e nmen has been ac i ely
p omo ing CE de elopmen , conside ing i an impo an pa h o
achie ing sus ainable de elopmen . In ecen yea s, China has de el-
oped a se ies o policies o encou age indi iduals and i ms o ake
ac ion, educe esou ce was e, and imp o e esou ce u iliza ion e i-
ciency. Fo ins ance, he Ci cula Economy P omo ion Law, imple-
men ed in 2009, is a key egula o y amewo k in ended o p omo e
esou ce ecycling and sus ainable economic p ac ices ac oss a ious
sec o s in China (Chen & Dages ani, 2023).
As he main body and key o ce o ad ancing CE de elopmen , i ms
ace many challenges in in eg a ing CE p ac ices (Puglia e al., 2024;
S ucki e al., 2023; Yang e al., 2023; Zhang e al., 2023). P e ious
s udies ha e p ima ily conduc ed ex ensi e esea ch on he ac o s ha
a ec CE p ac ices om mac ope spec i e and mic ope spec i e. A he
mac ole el, social and cul u al ac o s and go e nmen CE policies a e
closely ela ed o CE p ac ices (E e sbe g & F i z, 2022;
Gonz´
alez-Mo eno e al., 2024; Shang e al., 2022). Social and cul u al
ac o s such as indi idualism, an inno a ion-d i en mindse (E e sbe g
& F i z, 2022), and he public’s ecogni ion o he CE (Gonz´
alez-Mo eno
e al., 2024), ha e a c ucial in luence on shaping CE p ac ices. Fo
ins ance, highe socie al accep ance o CE p inciples can d i e consume
demand o sus ainable p oduc s and encou age i ms o adop ci cula
business models (Gonz´
alez-Mo eno e al., 2024). A he mic ole el,
* Co esponding au ho .
E-mail add ess: [email p o ec ed] (Y. Liu).
Con en s lis s a ailable a ScienceDi ec
Jou nal o Inno a ion & Knowledge
jou nal homepage: www.else ie .com/loca e/jik
h ps://doi.o g/10.1016/j.jik.2025.100702
Recei ed 27 No embe 2024; Accep ed 24 Ma ch 2025
Jou nal o Inno a ion & Knowledge 10 (2025) 100702
A ailable online 3 Ap il 2025
2444-569X/© 2025 The Au ho s. Published by Else ie España, S.L.U. on behal o Jou nal o Inno a ion & Knowledge. 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/ ).
i ms’ inancial condi ions (de Oli ei a Ne o e al., 2022; Meili & S ucki,
2023), he s eng h o ecei ed go e nmen subsidies (Lahane & Kan ,
2021; La ain e al., 2024), p i a e in es men (Al-Omoush e al., 2022),
and he echnology and in o ma ion exchange le els (S ucki e al., 2023)
ha e ecei ed conside able a en ion. Cu en ly, he main ba ie s o
i ms’ engagemen in CE p ac ices ha e been demons a ed a he
mic ole el, pa icula ly in e ms o inancial cons ain s and knowledge
ese es (Meili & S ucki, 2023; Wang e al., 2023). Collabo a ion be-
ween companies (i.e., open inno a ion; OI) is a po en ial app oach o
o e coming he signi ican echnological and in o ma ional ba ie s
ela ed o he equi emen o new echnologies, knowledge, and
p oduc de elopmen when p ac icing he CE.
OI has been de ined as an impo an ype o echnological and
in o ma ional exchange collabo a ion be ween i ms (Zhang e al.,
2023). The concep e e s o i ms’ le e aging ex e nal and in e nal
ideas and mobilizing in e nal and ex e nal esou ces o enhance ech-
nological capabili ies (Wang e al., 2023). The bounda ies o OI ha e
been expanded o a pu pose ul dis ibu ed inno a ion p ocess ha elies
on in o ma ion low ac oss company bounda ies (Chesb ough & Boge s,
2014). OI can enhance i ms’ compe i i e ad an ages. S udies ha e
demons a ed ha OI b ings signi ican bene i s o i ms by p omo ing
new p oduc de elopmen , sho ening he p oduc de elopmen cycle,
and accele a ing i ms’ g ow h (Hu on e al., 2024; To oslu e al.,
2023). OI also has a signi ican in luence on imp o ing i ms’ pe o -
mance. Resea ch has shown ha i can enhance i ms’ expo , inno a-
ion, business, and inancial pe o mance (Ogiemwonyi e al., 2023;
Sch¨
ape e al., 2023). As en i onmen al challenges become inc easingly
u gen , schola s ha e also begun o explo e he impac o OI on com-
panies’ g een ans o ma ion. S udies ha e indica ed ha open
eco-inno a ion p omo es enewable ene gy and g een echnology
de elopmen and applica ion h ough pa icipa ion and collabo a ion in
inno a ion ne wo ks (Csed˝
o e al., 2023; Roh e al., 2021). OI p omo es
echnological ad ancemen and signi ican ly imp o es g een o al ac o
p oduc i i y by in eg a ing ex e nal knowledge and op imizing in e nal
esou ces o p omo e e icien ene gy use (Wang e al., 2023).
Se e al s udies ha e demons a ed he co ela ion be ween OI and
CE p ac ices, indica ing ha OI can p omo e co po a e p ac ices in he
CE. Companies mus equen ly communica e wi h s akeholde g oups
(cus ome s, supplie s, academia, and he communi y) o o e come in-
e nal capabili y limi a ions and in eg a e CE p ac ices (Fang e al.,
2024; Jesus & Jugend, 2023; Pe o i e al., 2024). Fu he mo e, i ms
can disco e and apply eady-made ci cula solu ions h ough collabo-
a ions wi h lead g een ans o ma ion companies, educing he isks
associa ed wi h adop ing CE p ac ices (Calab ese e al., 2024; Kennedy
e al., 2017; Leonidou e al., 2024). Fu he mo e, some schola s ha e
a gued ha CE p ac ices p omo e OI. En i onmen al economic p ac ices
ha e a posi i e impac on en e p ises’ inno a ion ac i i ies; howe e ,
due o limi ed esea ch and de elopmen (R&D) capabili ies, small and
medium-sized en e p ises (SMEs) o en adop OI o acqui e exis ing
knowledge and capabili ies o educe inno a ion isks
(Sa ango-Lalangui e al., 2023). The esea ch on he ela ionship be-
ween OI and CE p ac ices has been inc easing. Howe e , he unde lying
in luencing mechanisms emain unclea , and exis ing s udies ha e p e-
dominan ly employed case s udy me hods, o en ocusing on limi ed
sample sizes (Calab ese e al., 2024; Pe o i e al., 2024). Fu he
empi ical esea ch is essen ial o es ablishing a mo e comp ehensi e
unde s anding o how OI d i es CE adop ion ac oss di e en con ex s.
Building on he exis ing li e a u e, his s udy o e s a comp ehensi e
unde s anding o how i ms’ OI in luences CE p ac ices and he ole o
go e nmen policies in his ela ionship. Speci ically, we i s de elop a
heo e ical amewo k o analyze he ansmission mechanisms h ough
which OI p omo es CE p ac ices, explo ing he h ee key pa hways o
comme cial c edi acquisi ion, g een knowledge ese es, and in o -
ma ion exchange. Second, we conduc an empi ical analysis using un-
balanced panel da a om lis ed companies on he A-sha es and Science
and Technology Inno a ion Boa d (STAR Ma ke ) in China om 2003 o
2022. Using obus econome ic me hods, we p esen empi ical e i-
dence on he in luencing mechanisms and examine he mode a ing e -
ec o CE policies. Finally, we u he deepen ou analysis by
conside ing i m and egional he e ogenei y o in es iga e how di e en
ypes o OI ca ego ized based on i m bounda ies exe di e se impac s
on CE p ac ices. We also explo e he condi ions unde which go e n-
men CE policies ampli y o cons ain he e ec i eness o OI in p o-
mo ing sus ainable business p ac ices.
This s udy makes h ee no able con ibu ions o he li e a u e. Fi s ,
by in eg a ing heo e ical analysis wi h empi ical e i ica ion, we o e
no el insigh s in o he mechanisms h ough which OI acili a es CE
p ac ices; a opic ha has been ela i ely unde explo ed. Second, we
ex end he unde s anding o policy in e en ions by sys ema ically
examining he mode a ing ole o go e nmen CE policies, p o iding
aluable insigh s ega ding he condi ions ha can enhance o hinde
hei e ec i eness. Thi d, unlike p e ious s udies ha ha e p ima ily
used case s udies o ocused on speci ic indus ies, ou esea ch co e s a
b oad sample o lis ed i ms ac oss a ious sec o s (excluding inancial
i ms), which enhances he gene alizabili y o ou indings. These con-
ibu ions ad ance academic in es iga ions on OI and CE p ac ices and
ha e aluable policy implica ions o os e ing sus ainable indus ial
ans o ma ion.
The emainde o his pape is o ganized as ollows. Sec ion 2 p e-
sen s ou heo e ical analysis and hypo heses. Sec ion 3 de ails he
s udy’s me hodology. Sec ion 4 p esen s he analysis and esul s, Sec ion
5 discusses he esul s, and Sec ion 6 p o ides he conclusion, co e ing
key indings, con ibu ions and implica ions, limi a ions, and di ec ions
o u u e esea ch.
Theo e ical amewo k and hypo heses
Concep de ini ions
OI
OI e e s o he use o companies’ in e nal and ex e nal inno a ions
o imp o e he inno a ion p ocess while also making in e nal knowl-
edge and echnology a ailable o ex e nal ma ke en i onmen s. As his
inno a ion ac i i y con inues o de elop, he concep o OI encompasses
in e nal and ex e nal collabo a ion and ecosys ems o d i e business
model and se ice inno a ion de elopmen (Chesb ough, 2017). OI has
e ol ed o include new models such as inno a ion ecosys ems (Zhang
e al., 2025) and c owd unding (Dabbous e al., 2024). T adi ional OI
e e s o knowledge low ha p ima ily occu s ac oss i m bounda ies,
encompassing inbound, ou bound, and coupled p ocesses (Chesb ough,
2017). The inbound p ocess emphasizes iden i ying and abso bing
ex e nal knowledge o enhance in e nal inno a ion capabili ies, which is
also known as inbound OI (Bianchi e al., 2016). Co espondingly, he
ou bound p ocess e e s o pushing in e nal inno a ion esul s o he
ex e nal en i onmen (ou bound OI), which can be achie ed h ough
echnology licensing ag eemen s, s a ups, and o he p ocesses (Enkel
e al., 2009). Coupled p ocesses e e o combined inbound and
ou bound OI o de elop inno a ions and achie e comme cializa ion
h ough collabo a ions and pa ne ships be ween companies and
s akeholde s (Enkel e al., 2009).
CE
CE is a concep o p omo ing sus ainable de elopmen ha o igi-
na ed om indus ial ecology ha is in ended o ad ance mo e e icien
esou ce u iliza ion, was e educ ion, and cleane echnology adop ion
(Ande sen, 2007). CE emphasizes educe, euse, and ecycle p inciples
o achie e closed-loop cycles o p oduc s, echnologies, and biology by
main aining p oduc s, componen s, and ma e ials a high- alue le els
(Geissdoe e e al., 2017). CE p ac ices can be implemen ed a he
mic ole el, mesole el, and mac ole el (K is ensen & Mosgaa d, 2020).
A he mic ole el, i ms ocus on sel -imp o emen and business de el-
opmen , in luencing p o i abili y and epu a ions h ough engaged and
X. Yin e al.
Jou nal o Inno a ion & Knowledge 10 (2025) 100702
2
publicized en i onmen al managemen p ac ices. Fi ms can use ecycled
ma e ials, design p oduc s o disassembly, and implemen ake-back
schemes o eco e used p oduc s o emanu ac u ing (Fang e al.,
2024). A he mesole el, i ms use in o ma ion sys ems o acili a e
esou ce exchange while p omo ing egional economic and na u al
en i onmen de elopmen . A no able example is ag i- ood indus ial
pa ks ha implemen CE p inciples by using ood was e and byp oduc s
o gene a e new alue (A anaso ska e al., 2022). A he mac ole el, he
CE in ol es a b oade geog aphical scope, wi h go e nmen s imple-
men ing public policies and sys ems o p omo e he de elopmen o
en i onmen al s a egies such as he Ex ended P oduce Responsibili y
(EPR) policy (Kunz e al., 2018).
Hypo heses de elopmen
As a coope a i e pa adigm, OI can p o ide possible solu ions o
o e come he ba ie s i ms ace in adop ing CE p ac ices by p omo ing
in e nal and ex e nal coope a ion and knowledge low. Fi s , OI can
acili a e close collabo a ion among s akeholde s in go e nmen , social
ins i u ions, and communi ies. By es ablishing an OI ecosys em, all
pa ies can join ly explo e he challenges and oppo uni ies o CE
p ac ices o o mula e and implemen inno a i e solu ions (Lu &
Chesb ough, 2022; Sch¨
ape e al., 2023). Ac i e pa icipa ion o go -
e nmen depa men s and communi ies in CE p ac ices has a key in lu-
ence on p o iding policy suppo and ma ke guidance o i ms,
p omo ing he p ac ical implemen a ion o he CE concep . Addi ionally,
OI p omo es c oss-bounda y knowledge sha ing and low, p o iding
i ms wi h mo e con enien access o ex e nal knowledge and echnol-
ogy (Ogiemwonyi e al., 2023; S isa han e al., 2023). Fi ms can ob ain
he la es esea ch esul s and cu ing-edge echnologies in he indus y
mo e easily h ough OI channels, accele a ing he implemen a ion o CE
p ac ices and imp o ing he e iciency o such p ac ices. OI also p o ides
i ms wi h he abili y o espond mo e lexibly and apidly o ma ke
demands and changes (Csed˝
o e al., 2023; Jang & on Zed wi z, 2023).
Fi ms can ob ain ma ke eedback mo e quickly, imp o e and op imize
exis ing CE p ac ices, and de elop such p ac ices o be e mee ma ke
demands. Howe e , despi e hese ad an ages, in he con ex o he CE,
OI also in oduces po en ial challenges. Fi ms engaging in OI may isk
exposing p op ie a y g een echnologies o compe i o s, which could
weaken hei long- e m ma ke posi ion (Fang e al., 2024). Mo eo e ,
supply chain pa ne s may ha e di e ing sus ainabili y goals, which can
make i di icul o align s a egies o CE implemen a ion (Leonidou
e al., 2024; Pa anayak e al., 2024). Despi e hese challenges, he
o e all bene i s o OI in d i ing CE p ac ices end o ou weigh he isks
(Csed˝
o e al., 2023; Ogiemwonyi e al., 2023). Based on he abo e
heo e ical analysis, we p opose Hypo hesis 1.
H1. OI p omo es co po a e adop ion o CE p ac ices.
Companies ha adop OI s a egies ypically engage in close
communica ion and collabo a ion wi h a ious s akeholde s. This
in e ac ion educes in o ma ion asymme y, pa icula ly wi hin supply
chain ne wo ks (Cao e al., 2022). Th ough ongoing communica ion and
coope a ion, companies and s akeholde s can sha e in o ma ion, e-
sou ces, and needs mo e easily, minimizing mo al haza ds in pa ne -
ships and os e ing mu ual us (Chiu & Lin, 2022; Han e al., 2021).
Es ablishing and main aining s ong us ela ionships when imple-
men ing a CE s a egy is c ucial. Fo example, close collabo a ion be-
ween ups eam and downs eam companies is necessa y in he
p oduc ion o ecyclable p oduc s o e ec i ely mi iga e a ious isks in
he p oduc ion p ocess (Liu e al., 2023). Ups eam supplie s mus us
downs eam manu ac u e s o handle and use ecyclable ma e ials
app op ia ely. Building us enables ups eam supplie s o ac i ely
coope a e wi h downs eam pa ne s o explo e me hods o inc easing
ma e ial ecycling a es, educing was e, and op imizing CE p oduc ion
p ocesses. Fu he mo e, such us is e lec ed in inc eased access o
comme cial c edi o businesses. This indica es ha supplie s a e mo e
willing o assume he isks associa ed wi h delayed paymen s om
us ed companies (Xiu e al., 2023). The e o e, OI enables companies o
secu e mo e ade c edi . In he esou ce-based iew (RBV), i ms’
compe i i e ad an ages s em om unique esou ces and capabili ies
(Ba ney, 1991). In e es - ee unds can educe inancing cos s, enhance
cash low, and imp o e ope a ional lexibili y, enabling companies o
main ain a s ong inancial posi ion o ad ance CE ini ia i es (Lu &
Chesb ough, 2022; Sch¨
ape e al., 2023). Based on his heo e ical
analysis, we p opose Hypo hesis 2a.
H2a. OI p omo es co po a e CE p ac ices by enhancing access o
comme cial c edi .
As a dynamic and lexible model, OI p o ides companies wi h
ex ensi e oppo uni ies o acqui e g een pa en s, pa icula ly in a
con ex in which en i onmen al challenges a e escala ing and g een
echnologies a e becoming inc easingly complex (Ogiemwonyi e al.,
2023; Roh e al., 2021). Companies can pu chase g een pa en s, ob ain
licensing om uni e si ies o o he esea ch ins i u ions, o collabo a e
wi h complemen a y i ms in collabo a i e inno a ion, c ea ing alu-
able oppo uni ies o expand hei g een capabili ies and knowledge
ese es (Roh e al., 2021). In his p ocess, companies do mo e han
simply acqui e g een pa en s; hey gain deep insigh s in o he c i ical
skills and knowledge equi ed o ob ain pa en s h ough collabo a ion
wi h pa en pa ne s (Zhao e al., 2022). This collabo a ion also helps
i ms de elop capabili ies o de elop new pa en ed echnologies,
enhancing R&D capabili ies (Figuei edo e al., 2020). Companies can
gain a deepe unde s anding o he g een echnology sec o and s ay
ab eas o he la es ends and de elopmen s in g een inno a ion,
expanding hei expe ise and echnological s eng h in he ield (Lei ˜
ao
e al., 2024; Zeng e al., 2022). The e o e, OI can signi ican ly enhance a
company’s g een knowledge ese es. Acco ding o he RBV (Ba ney,
1991), g een knowledge ese es a e a c i ical s a egic esou ce ha
es ablishes a ounda ion o sus ainable compe i i e ad an age. A
well-de eloped g een knowledge ese e p o ides a solid base o CE
implemen a ion. Fi ms wi h ex ensi e g een knowledge can op imize
esou ce e iciency, de elop ecyclable and eco- iendly p oduc s, and
adop closed-loop p oduc ion sys ems (Ai e al., 2024). We p opose
Hypo hesis 2b based on he abo e heo e ical analysis.
H2b. OI p omo es co po a e CE p ac ices by inc easing g een knowl-
edge ese es.
Wi hin he CE amewo k, a company’s co e goals include esou ce
conse a ion and euse, which can be achie ed h ough con inuously
op imizing p oduc ion p ocesses and p oduc li e cycles o ensu e e i-
cien esou ce ci cula ion. The OI model encou ages companies o
ac i ely sha e in o ma ion and knowledge wi h supply chain pa ne s,
cus ome s, and o he s akeholde s, p o iding hem wi h oppo uni ies o
gain deepe insigh s in o ma ke demands and ends. Companies can
access comp ehensi e ma ke in o ma ion and gain a deepe unde -
s anding o consume s’ needs and p e e ences h ough engagemen and
collabo a ion wi h ex e nal pa ne s (Fishe e al., 2024). S akeholde
heo y highligh s he c ucial ole o a ious s akeholde s such as sup-
plie s, cus ome s, and egula o y bodies in shaping i ms’ s a egic ini-
ia i es (Ozdemi e al., 2023). The p ocess o in o ma ion sha ing and
knowledge exchange enables companies o design and p oduce p oduc s
ha mee ma ke demands mo e accu a ely, enhancing hei adap -
abili y and compe i i eness (Jang & on Zed wi z, 2023). OI also equips
companies wi h he abili y o iden i y and swi ly adap o en i on-
men al changes, which is c ucial in CE p ac ices. In he OI amewo k,
companies can lexibly acqui e ex e nal in o ma ion and espond
p omp ly o mee e ol ing ma ke demands and en i onmen al ends
(Al-Omoush e al., 2024). This agili y in ecogni ion and adap a ion
enables companies o seize oppo uni ies, e ec i ely o e come chal-
lenges, and main ain compe i i eness. Companies should es ablish
lexible inno a ion mechanisms and con inuously adjus CE models o
adap o he apidly changing ma ke en i onmen (Guo e al., 2023;
X. Yin e al.
Jou nal o Inno a ion & Knowledge 10 (2025) 100702
3
Vasi e al., 2024). Based on he abo e heo e ical analysis, we p opose
Hypo hesis 2c.
H2c. OI acili a es co po a e CE p ac ices by enhancing in o ma ion
exchange.
CE p ac ices a e cha ac e ized by long payback pe iods, high ma ke
demand unce ain y, and subs an ial in es men cos s. As businesses
o en ind i challenging o ansi ion om a linea app oach o a CE
model independen ly, go e nmen CE policies ha e a signi ican in lu-
ence on encou aging companies’ ac i e engagemen in CE p ac ices
(Chena az & Dimi o , 2024; Puglia e al., 2024). Fo example, go -
e nmen s can implemen incen i e p og ams o suppo companies in
designing and p oducing ecyclable p oduc s. These incen i es may
include subsidies o ax educ ions o businesses in es ing in CE p o-
cesses. Ano he impo an policy ool is he EPR policy, which encou -
ages companies o conside hei p oduc s’ end-o -li e s age du ing he
design phase o make hem easie o euse, epai , o ecycle, os e ing
inno a ion in p oduc design and was e managemen (Kunz e al., 2018).
Mo eo e , CE policies also in luence consume beha io (Ha ley e al.,
2023; Yuana e al., 2024). Fo example, go e nmen s may o e sub-
sidies o consume s who pu chase ecyclable p oduc s, encou aging
hem o buy ecycled goods and ac i ely pa icipa e in epai and euse
ac i i ies. The e o e, go e nmen CE policies can guide i ms o engage
in OI wi hin he CE domain by p o iding incen i es and egula o y
amewo ks and s imula ing demand o ecyclable p oduc s (Puglia
e al., 2024). Based on he abo e analysis, we p opose Hypo hesis 3.
H3. CE policies posi i ely mode a e he ela ionship be ween OI and
co po a e CE p ac ices.
Based on Hypo heses 1, 2a, 2b, 2c, and 3, he heo e ical amewo k
o his pape is illus a ed in Fig. 1.
Me hods
Va iable de ini ions and desc ip ions
Dependen a iable: CE p ac ices
P e ious esea ch has employed a a ie y o me hods o measu e
co po a e CE p ac ices such as su eys (Meili & S ucki, 2023; S ucki
e al., 2023), con en analysis (Yuan & Pan, 2023), and ex analysis
(Alka aan e al., 2023; Chen & Dages ani, 2023). Conside ing he
a ailabili y and objec i i y o he da a, we employ ex analysis o
e alua e co po a e p ac ices in he CE.
Independen a iable: OI
Fi ms’ inno a ion is o en e lec ed in pa en acquisi ion (Belde bos
e al., 2014). Mo eo e , pa en acquisi ion is an op imal a iable o
comp ehensi ely s udy OI ac oss di e se indus ies due o high da a
a ailabili y and s anda diza ion (Walsh e al., 2016). Acco dingly, based
on he OI concep , we conside lis ed companies’ join pa en publica-
ion as ep esen a i e o OI (A o a e al., 2016). Based on he a ying
scopes o co po a e bounda ies and he cha ac e is ics o pa en da a, we
classi y co po a e OI in o h ee ypes. Fi s , in acompany OI, which
in ol es mul iple i ms wi hin he same publicly lis ed company join ly
disclosing pa en s and is deno ed as Open_inno _in_comp. Second,
in e company OI, which in ol es i ms om di e en publicly lis ed
companies join ly disclosing pa en s and is deno ed as Open_-
inno _be _comp. Thi d, coope a i e inno a ion, which in ol es a sub-
sidia y o one publicly lis ed company join ly disclosing pa en s wi h an
a ilia ed company o ano he publicly lis ed company, which is deno ed
as Open_inno _wi h_comp.
Media ing a iables
•T ade c edi acquisi ion. T ade c edi enables a pa y o delay pay-
men solely based on i s exis ing c edi wo hiness and e lec s he
us be ween ups eam and downs eam companies in a supply
chain. The amoun ha a company delays paying i s supplie s is
e lec ed in i s inancial s a emen s as accoun s payable. Re e encing
p e ious s udies (Ding e al., 2023; Gan e al., 2024), we use he a io
o accoun s payable o o al liabili ies o measu e i ms’ ade c edi
access.
•G een knowledge ese es. G een pa en s e e o pa en s ela ed o
en i onmen al p o ec ion, sus ainable de elopmen , and simila
conce ns, including echnologies in en i onmen al p o ec ion, clean
ene gy, and CE. We use he numbe o g een pa en s a company holds
o e lec i s ac ual pe o mance and capabili y in g een echnology
inno a ion (Ai e al., 2024). Since some companies do no ha e g een
pa en s, we use he na u al loga i hm o he numbe o g een pa en s
plus one o measu e i ms’ g een knowledge acquisi ion.
•In o ma ion exchange. In o ma ion exchange be ween a company
and i s ex e nal en i onmen encompasses knowledge and ma ke
in o ma ion exchange. Such exchange e lec s a i m’s inno a ion
capaci y, compe i i e ad an ages, and abili y o adap o ma ke
changes. We measu e knowledge exchange by he numbe o ci a-
ions o a lis ed company’s pa en s, and quan i y ma ke in o ma ion
exchange by he coe icien o a ia ion in he company’s annual
R&D in ensi y (R&D expendi u e/ e enue), capi al in ensi y (capi al
expendi u e/ e enue), and ad e ising in ensi y (ad e ising
expendi u e/ e enue) (Yang e al., 2023).
Mode a ing a iable: CE policy
CE policies include egula o y, olun a y, and economic incen i e
policies (Chena az & Dimi o , 2024). China has p oduced nume ous
policy documen s ela ed o de eloping a CE, and speci ic policies mus
be aligned wi h China’s p ac ical ci cums ances. The Ci cula Economy
Fig. 1. F amewo k.
X. Yin e al.
Jou nal o Inno a ion & Knowledge 10 (2025) 100702
4
Pilo Ci y Ini ia i e is in ended o in eg a e CE p inciples in o indus y,
ag icul u e, se ices, and u ban in as uc u e de elopmen by p o-
mo ing CE p oduc ion me hods and g een li es yles ac oss all s ages o
p oduc ion, dis ibu ion, and consump ion (Luo & Leipold, 2022; Ma
e al., 2022). This policy is expec ed o ha e a signi ican impac in he
egions whe e i is implemen ed. The e o e, we cons uc a dummy
a iable o measu e he impac o CE policies on i ms, indica ing
whe he he p o ince whe e a company is loca ed is pa icipa ing in he
Ci cula Economy Pilo Ci y Ini ia i e.
Con ol a iables
Re e encing p e ious esea ch (Fang e al., 2024; Yuan & Pan, 2023),
we in oduce he ollowing con ol a iables ha may in luence co po-
a e p ac ices in he CE in o ou model.
•Fi m size (SIZE), measu ed by he na u al loga i hm o o al asse s;
•Fi m age (AGE), measu ed by he numbe o yea s since i ms’
es ablishmen o he obse a ion yea ;
•Deg ee o Combined Le e age(DCL), ep esen ed by he a io o he
change in ne p o i o ha o main business e enue;
•Cash low s a us (CFO), exp essed as he a io o ne cash low om
ope a ing ac i i ies o o al asse s;
•To al asse u no e , measu ed by he a io o ope a ing e enue o
o al asse s;
•Boa d size (BOARD), measu ed by he na u al loga i hm o he
numbe o boa d membe s;
•CEO duali y (DUAL), indica ing whe he he CEO and boa d chai a e
he same pe son (1 i yes, 0 i no);
•Big Fou audi (AUDIT), indica ing whe he he annual epo is
audi ed by a Big Fou accoun ing i m (Deloi e, E ns & Young,
KPMG, and P icewa e houseCoope s) (1 i yes, 0 i no).
Me hods o gene a ing key a iables
Dependen a iable: CE p ac ices
Fi s , since all ex ual ma e ials used in his s udy a e in Chinese, we
p ep ocess he ex ual ma e ials using Jieba, which is a widely adop ed
Chinese wo d segmen a ion lib a y. This s ep is essen ial because, unlike
English, Chinese ex s do no ha e na u al wo d bounda ies. Following
segmen a ion, we employ La en Di ichle Alloca ion (LDA) opic
modeling o ex ac key hemes ela ed o CE p ac ices om policy
documen s. LDA is a p obabilis ic model ha assumes each documen
consis s o mul iple opics, wi h each opic ep esen ed by a se o key-
wo ds. Analyzing a collec ion o 11 au ho i a i e policy documen s om
ins i u ions such as he S a e Council, he Na ional De elopmen and
Re o m Commission, and he S a e En i onmen al P o ec ion Adminis-
a ion, we iden i y 13 co e opic wo ds ha encapsula e c i ical aspec s
o CE p ac ices. The esul ing opic wo ds include CE, g eening,
enewable esou ces, clean p oduc ion, emanu ac u ing, comp ehen-
si e u iliza ion, bulk solid was e, ci cula ans o ma ion, ecycling,
comp ehensi e esou ce u iliza ion, eco e y, disca ded ma e ials, and
esou ce eco e y. To add ess he conce n ha companies in di e en
p o inces may use a ying e minologies o desc ibe simila CE p ac-
ices, we employ Wo d2Vec; a machine lea ning model ha iden i ies
seman ic ela ionships be ween wo ds. We ain he Wo d2Vec model on
cen al and p o incial go e nmen wo k epo s o expand ou ini ial se
o 13 keywo ds in o a mo e comp ehensi e dic iona y o 55 e ms
closely associa ed wi h CE p ac ices. Finally, we coun he equency o
he 55 CE- ela ed e ms in each company’s epo , using i as an indi-
ca o o i ms’ CE engagemen .
Independen a iable: OI
Fi s , we es ablish a da abase o map each lis ed company’s s ock
code o i s o icial name, as well as he names o i ms’ subsidia ies and
join en u es. Nex , we indi idually ma ch he applican names o each
pa en agains his da abase. I a pa en has only one applican bu
ma ches wo di e en lis ed company codes, his indica es ha he
applica ion ep esen s a join en u e be ween wo lis ed companies,
and we classi y he pa en as coope a i e OI. Fo pa en s wi h wo o
mo e applican s, i all applican s ma ch only one lis ed company code,
he pa en is ca ego ized as in e company OI. Howe e , i di e en ap-
plican s ma ch di e en lis ed company codes, he pa en is classi ied as
in acompany OI. Finally, we agg ega e he annual OI da a o each
lis ed company. As many companies did no engage in join pa en ap-
plica ions and ze o canno be loga i hmically ans o med, we use he
loga i hm o he numbe o join pa en s plus one as an indica o o OI
o all sample companies.
Da a collec ion and sampling
The da a used in his s udy p ima ily include pa en da a and i m-
le el da a. The pa en da a a e sou ced om he China Na ional
Knowledge In as uc u e ull- ex pa en da abase and include ields
such as pa en applica ion numbe s, publica ion da es, and applican s
o 5.54 million in en ion pa en s. A e il e ing he applican ield o
lis ed companies’ names, subsidia ies, and join en u es, we iden i ied
1,139,691 pa en s ha could be a ibu ed o ou sample o lis ed
companies. Fi m-le el da a a e ob ained om he China S ock Ma ke &
Accoun ing Resea ch da abase, including basic company in o ma ion,
inancial s a emen s, and equi y s uc u e in o ma ion. We ob ain
companies’ annual epo ex s om he Cnin o websi e and p ocess
hem using Jieba o ex segmen a ion. A e o ganizing he da a, we
pe o med he ollowing da a cleaning s eps. (1) Excluding lis ed com-
panies in he inancial and insu ance sec o s; (2) excluding samples wi h
ST, *ST, and PT ma ke s; (3) ma ching by s ock code and yea , we
in eg a e CE p ac ices, OI, and con ol a iables, cons uc ing an un-
balanced panel da ase wi h 40,219 sample obse a ions om 2003 o
2022.Table 1 p esen s he co ela ion ma ix and desc ip i e s a is ics o
he da a.
Model speci ica ion
Benchma k eg ession model
Based on he heo e ical analysis and hypo heses in Sec ion 2, we
cons uc a wo-way ixed e ec s (TWFE) model o examine he impac
o OI on co po a e CE p ac ices. The speci ic model is as ollows:
CEWFi =β0+β1Open inno i +∑βnCon Va si +λi+
η
+
ε
i
(1)
whe e he dependen a iable (CEWFi ) ep esen s he i m’s CE p ac-
ices. The key independen a iable (Open inno i ) deno es he i m’s OI.
Con Va si ep esen s he p e iously in oduced se o con ol a iables,
λi deno es indi idual ixed e ec s (FEs),
η
deno es yea FEs, and
ε
i is he
e o e m.
Media ing e ec model
We use he ollowing model o examine he media ing e ec s o ade
c edi acquisi ion, g een knowledge ese es, and in o ma ion exchange.
CEWFi =β0+β1Open inno i +∑βnCon Va si +λi+
η
+
ε
i
(2)
Mi =β0+β1Open inno i +∑βnCon Va si +λi+
η
+
ε
i
(3)
whe e Mi ep esen s he media ing a iables o ade c edi acquisi ion
(CTA), g een knowledge ese es (GKRs), and in o ma ion exchange
le el as media o a iables.
X. Yin e al.
Jou nal o Inno a ion & Knowledge 10 (2025) 100702
5
Mode a ing e ec model
CEWFi =β0+β1Open inno i +β2Policyi +β3Policyi
×Open inno i +∑βnCon Va si +λi+
η
+
ε
i (4)
whe e Policyi deno es he mode a ing a iable o OI policy.
Analysis and esul s
Benchma k eg ession
We used a TWFE model o assess he impac o OI on co po a e CE
p ac ices. The eg ession esul s a e p esen ed in Table 2 Column (1)
p esen s he eg ession esul s including only con ol a iables, wi hou
conside ing FEs. Columns (2) and (3) show he eg ession esul s wi h
indi idual and yea FEs, espec i ely. The esul s indica e ha he co-
e icien o OI is posi i e and signi ican a he 1 % le el, indica ing ha
OI enhances co po a e CE p ac ices.
Endogenei y es ing
Ins umen al a iable me hod
Re e encing p e ious esea ch, we use he a e age OI le el o o he
i ms o he same size in he same yea (Han e al., 2024) and he numbe
o pa en owne ship li iga ion cases esol ed in he p o ince whe e he
i m is loca ed in he same yea as ins umen al a iables (IVs) o i ms’
OI. This app oach add esses endogenei y issues as hese wo a iables
sa is y he ele ance and exogenei y assump ions o IVs. The speci ic
jus i ica ion is desc ibed below.
We ca ego ize all lis ed companies in o deciles by size and calcula e
he a e age OI le el (A gopen_inno ) o i ms in each g oup, excluding
he i m i sel , using his a e age as an IV o OI le el. Fi s , he OI le el
among o he i ms o he same scale does no di ec ly in luence i ms’
engagemen in CE p ac ices, sa is ying he exogenei y assump ion.
Addi ionally, a i m’s OI le el is in luenced by hose o simila -sized
i ms, sa is ying he ele ance assump ion, mee ing IV ele ance and
exogenei y equi emen s.
We ob ain he numbe o pa en owne ship dispu es esol ed in each
p o ince om he Peking Uni e si y Law Da abase and use he loga-
i hm o his numbe plus one (Sui ) as an IV o OI a he p o incial
le el. Acco ding o he pa adox o openness (Lau sen & Sal e , 2014),
manage s engage i ms in OI by collabo a ing wi h a wide ange o
ex e nal ac o s while simul aneously s i ing o p o ec hei p op ie a y
knowledge om being copied by compe i o s. The numbe o pa en
owne ship dispu es in a i m’s p o ince is an indica o o he s eng h o
local pa en p o ec ion (Rudy & Black, 2015), sa is ying he ele ance
assump ion. Exogenei y is e lec ed in he ac ha p o incial-le el
pa en li iga ion cases do no di ec ly in luence a i m’s engagemen in
CE p ac ices.
This s udy employs wo-s age leas squa es (2SLS) and gene alized
me hod o momen s (GMM) es ima ion o IV eg ession. The esul s in
he i s and second s ages o he 2SLS eg ession a e espec i ely p e-
sen ed in Columns (1) and (2) o Table 3, and he GMM es ima ion e-
sul s a e p esen ed in Column (3).
The esul s o he IV es show ha he p- alue o he LM s a is ic is
0.0000, ejec ing he null hypo hesis o IV unde iden i ica ion. The Wald
F s a is ic is 100.669, a exceeding he 10 % c i ical h eshold o 19.93,
indica ing no weak IV p oblem. The p- alue o he Sa gan s a is ic is
0.1526, suppo ing he null hypo hesis ha all IVs a e exogenous and
con i ming no o e iden i ica ion issues wi h he IVs.
In Column (1) o Table 3, he IVs A gopen_inno and Sui a e sig-
ni ican a he 0.01 le el, wi h posi i e coe icien s, sa is ying he ele-
ance assump ion o he IVs. In Column (2) o Table 3, he explana o y
a iable is posi i ely signi ican a he 0.01 le el, indica ing ha a e
add essing endogenei y issues, he e ec o i ms’ OI on CE p ac ices
emains signi ican ly posi i e, wi h he coe icien no ably inc eased.
The GMM es ima ion esul s in Column (3) also suppo his conclusion.
Table 1
Desc ip i e s a is ics and co ela ion ma ix.
Va iables 1 2 3 4 5 6 7 8 9 10
1 CEP 1.000
2 Open_inno 0.186*1.000
3 SIZE 0.252*0.292*1.000
4 AGE 0.188*0.132*0.263*1.000
5 DCL 0.030*-0.018*0.058*0.008*1.000
6 CFO 0.018*0.029*0.052*0.0000 -0.068*1.000
7 TAT 0.008 0.011*0.034*-0.024*-0.022*0.114*1.000
8 BOARD -0.010*0.033*0.195*-0.076*0.069*0.047*0.036*1.000
9 DUAL -0.010*0.014*-0.138*-0.021*-0.062*-0.012*-0.054*-0.191*1.000
10 AUDIT 0.023*0.138*0.320*0.015*-0.007 0.071*0.024*0.095*-0.057*1.000
Mean 10.397 0.371 22.044 16.637 2.415 0.052 0.6345 2.143 0.268 0.061
SD 16.377 0.972 1.320 6.422 2.921 0.071 0.4294 0.202 0.443 0.240
Min 0.0000 0.000 15.417 0.000 0.877 -0.163 0.0741 1.609 0.000 0.000
Max 97.000 9.095 28.606 62.000 21.347 0.258 2.5182 2.708 1.000 1.000
*
No e. indica e ha he bila e al es is signi ican a 5 %.
Table 2
Benchma k eg ession esul s.
(1) (2) (3)
Va iables CEP CEP CEP
Open_inno 1.3450*** 1.1023*** 1.0302***
(0.0761) (0.0748) (0.1734)
CFO 1.9544** 1.7742** 2.2639**
(0.8300) (0.8222) (1.0596)
BOARD -0.6888 -0.2799 -0.5375
(0.4237) (0.4178) (0.8373)
TAT 0.9470*** 0.6509*** 0.6863
(0.2118) (0.2094) (0.5127)
SIZE 2.3034*** 2.2110*** 2.1722***
(0.0879) (0.0875) (0.2243)
AGE 0.6510*** 0.0585 0.8258***
(0.0150) (0.0366) (0.0344)
DUAL -0.3504** -0.4771*** -0.3965
(0.1673) (0.1640) (0.2947)
AUDIT -1.7398*** -1.8598*** -1.7072**
(0.3766) (0.3685) (0.8079)
DCL 0.0286 0.0376** 0.0268
(0.0192) (0.0189) (0.0259)
yea e NO YES YES
i m e NO NO YES
N 40219 40219 40219
2_w 0.217 0.249 0.249
No e. The alue o -s a is ics epo ed in pa en heses; ***, **, * indica e ha he
bila e al es is signi ican a 1 %, 5 %, and 10 %, espec i ely.
X. Yin e al.
Jou nal o Inno a ion & Knowledge 10 (2025) 100702
6
Di e ence-in-di e ences
A i m’s OI ac i i ies may p omo e CE p ac ices, whe eas i s CE
p ac ices migh encou age OI, indica ing a po en ial bidi ec ional cau-
sali y issue. The e o e, his s udy nex in es iga es he causal ela ion-
ship be ween he wo by ea ing i ms’ OI ac i i ies as a quasi-na u al
expe imen . Since he iming o OI a ies ac oss i ms, we apply a
s agge ed di e ence-in-di e ences (DID) model. Re e encing p e ious
esea ch (Al Ab i e al., 2017), his s udy iden i ies he i s yea in which
a i m con inuously engaged in OI o a leas h ee yea s as he poin o
occu ence o he quasi-na u al expe imen .
The DID model es ablished in his s udy is as ollows:
CEWFi =β0+β1T ea i×Pos i +∑βnCon Va si +λi+
η
+
ε
i
(5)
As shown in Column (4) o Table 3, i ms’ OI adop ion signi ican ly
enhances hei CE p ac ices. We also conduc pa allel ends and dy-
namic e ec s es s using he DID model.
The esul s in Fig. 2 demons a e ha he in e ac ion e ms a e no
s a is ically signi ican p io o OI implemen a ion, indica ing no
no able di e ences be ween he ea men and con ol g oups, ali-
da ing he pa allel ends assump ion. Howe e , a e he ini ia ion o
OI, he in e ac ion coe icien s become signi ican ly posi i e and exhibi
an upwa d end. This sugges s ha OI p omo es co po a e CE p ac ices
and he e ec pe sis s o e ime.
P opensi y sco e ma ching
To ensu e ha he cha ac e is ics o i ms in he ea men and
con ol g oups a e simila , we employ p opensi y sco e ma ching (PSM)
o pai ea men g oup samples wi h he mos compa able con ol g oup
samples. This app oach can mi iga e he p oblem o sample selec ion
bias.
Fig. 3 illus a es he esul s o he PSM, e ealing ha signi ican
di e ences be ween ce ain a iables we e e iden in he con ol and
ea men g oups p io o ma ching. Howe e , a e ma ching, no sig-
ni ican di e ences in any a iables a e appa en be ween he ma ched
con ol and ea men g oups.
We e-es ima e he DID model using he ma ched samples, p esen ing
he esul s in Column (5) o Table 3. The indings om he PSM-DID
analysis suppo he baseline conclusions o his s udy.
Gi en ha PSM ma ches samples ac oss di e en yea s, i is possible
ha con ol and ea men g oup samples om signi ican ly di e en
yea s could be ma ched. To add ess his, we applied a pe iod-by-pe iod
ma ching app oach, pai ing ea men and con ol g oup i ms om he
same yea , using he yea be o e he ea men i m’s policy in e en ion
as he baseline. This app oach imp o es compa abili y by con olling o
mac oeconomic luc ua ions ha could o he wise con ound he esul s.
The esul s shown in Column (6) o Table 3 also suppo ou baseline
conclusions.
Robus ness es s
We nex eplace he co e explana o y a iable and exclude samples
om ce ain indus ies o es he obus ness o he baseline eg ession.
Speci ically, we change he measu e o OI om he loga i hm o one plus
he numbe o OIs (Open_inno ) o he p opo ion o OI pa en s o o al
Table 3
Tes ing esul s o endogenei y.
(1) (2) (3) (4) (5) (6)
Va iables Open_inno CEP CEP CEP CEP CEP
A gopen_inno 0.1187***
(0.0085)
Sui 0.0106***
(0.0039)
Open_inno 14.1575*** 14.1575***
(1.3913) (1.3908)
T ea _pos 2.3807*** 2.7724*** 2.5233***
(0.4450) (0.2016) (0.2294)
con ol_ a YES YES YES YES YES YES
yea e YES YES YES YES YES YES
i m e YES YES YES YES YES YES
N 40219 40219 40219 40219 34219 23156
2_w . . . 0.249 0.244 0.258
No e. The alue o -s a is ics epo ed in pa en heses; ***, **, * indica e ha he bila e al es is signi ican a 1 %, 5 %, and 10 %, espec i ely.
Fig. 2. Pa allel end es . Fig. 3. Ma ching e ec .
X. Yin e al.
Jou nal o Inno a ion & Knowledge 10 (2025) 100702
7
pa en s held by he lis ed company (Open_inno _ a io). The esul s o
his es ima ion a e shown in Column (1) o Table 4. Fu he mo e, since
collabo a i e inno a ion ypically occu s wi hin a single i m, we
exclude his ype o OI and ocus only on in acompany and in e com-
pany OIs (Open_inno 1). The esul s o his es ima ion a e p esen ed in
Column (2) o Table 4.
Addi ionally, conside ing ha he impac o OI on co po a e CE
p ac ices may ha e a lag e ec , we use lagged e ms as he co e
explana o y a iable. The es ima ion esul s a e p esen ed in Column (3)
o Table 4. Mo eo e , since some indus ies a e highly ela ed o CE and
could a ec he obus ness o he esul s, we exclude da a om
ecological p o ec ion and en i onmen al managemen and was e
esou ce ecycling sec o s, p esen ing he esul s in Column (4) o
Table 4. All obus ness es esul s show ha he coe icien o OI e-
mains posi i e and signi ican a he 1 % le el, demons a ing he al-
idi y o he benchma k indings and con i ming Hypo hesis 1.
Mechanism analysis
Based on Hypo heses 2a–2c, his s udy employs a wo-s ep me hod o
examine whe he comme cial c edi a ailabili y, GKRs, and in o ma ion
exchange media e he impac o OI on co po a e CE p ac ices (Wei e al.,
2024). The eg ession esul s a e p esen ed in Table 5.
Column (1) o Table 5 shows ha OI signi ican ly enhances com-
me cial c edi access, suppo ing Hypo hesis 2a. Column (2) o Table 5
indica es ha OI s eng hens g een knowledge ese e, con i ming Hy-
po hesis 2b. Columns (3) and (4) o Table 5 demons a e ha OI
signi ican ly imp o es in o ma ion exchange, p o iding e idence o
Hypo hesis 2c. Addi ionally, Column (5) o Table 5 e eals ha CE
policies signi ican ly in luence he ela ionship be ween OI and co po-
a e CE p ac ices, suppo ing Hypo hesis 3. All e ec s a e s a is ically
signi ican a he 1 % le el. In Fig. 4, bo h lines ha e posi i e slopes,
alida ing Hypo hesis 1, and he slope o he line o companies imple-
men ing CE policies is g ea e han ha o hose no implemen ing such
policies, suppo ing Hypo hesis 3.
He e ogenei y es s
Collabo a ion he e ogenei y es
OI in companies in ol es c ossing bounda ies be ween di e en
en i ies (Caccamo e al., 2023). Based on ou p e ious classi ica ion o
OI ypes among lis ed companies, we examine he impac o di e en
ypes o OI on co po a e p ac ices in he CE.
The esul s in Table 6 show ha he coe icien s o di e en ypes o
OI a e all posi i e and s a is ically signi ican a he 1 % le el. No ably,
in e company OI has he la ges coe icien among hem.
Region he e ogenei y es
China is a as coun y, wi h signi ican de elopmen dispa i ies
ac oss di e en egions, pa icula ly be ween eas e n and midwes e n
egions. To assess he impac o OI on he CE p ac ices mo e accu a ely,
we di ide he sample in o companies om eas e n and midwes e n e-
gions, in oducing a a iable o indus ializa ion le el (IL) di e ences,
measu ed by he a io o indus ial added alue o employmen .
Fig. 5 illus a es he indus ializa ion di e ences be ween eas e n
and midwes e n egions. The eas e n egion, ma ked by he ed line,
clea ly shows a highe IL compa ed wi h he midwes e n egion.
Table 7 p esen s he nume ical esul s o he egional he e ogenei y
analysis. Columns (1) and (2) o Table 7 epo he baseline eg ession
model esul s o i ms in eas e n and midwes e n egions, espec i ely.
The indings indica e ha he coe icien s o OI a e posi i e and s a-
is ically signi ican a he 1 % le el, wi h a la ge e ec obse ed o
i ms in he eas e n egion.
Columns (3) and (4) o Table 7 inco po a e CE policy (policy) and he
in e ac ion e m Open_inno ×policy in o he baseline eg ession
model, ollowed by subg oup eg essions o i ms in eas e n and mid-
wes e n egions. The esul s e eal ha while he coe icien o OI e-
mains posi i e and signi ican a he 1 % le el, he in e ac ion e m
be ween policy and OI is only posi i e and signi ican a he 1 % le el in
he eas e n egion.
Columns (5) and (6) o Table 7 u he ex end he eg ession model
by adding he iple in e ac ion e m Open_inno ×policy ×IL, along
wi h he in e ac ion e ms Open_inno ×IL and policy ×IL. The esul s
demons a e ha while he coe icien s o OI emain signi ican ly
posi i e, he iple in e ac ion e m is only signi ican ly posi i e in he
eas e n egion, indica ing ha he mode a ing e ec o CE policy is mo e
p onounced in a eas wi h highe indus ializa ion le els.
Fig. 6 isually illus a es his conclusion. The line ep esen ing
highe indus ializa ion le els and CE policy implemen a ion has he
s eepes slope, demons a ing ha he posi i e impac o OI on CE
p ac ices is s onges unde hese condi ions.
Discussion
Ou indings highligh he s ong in luence o OI on ad ancing CE
p ac ices. This aligns wi h he conclusions o Leonidou e al. (2024), who
emphasized he signi icance o mul ina ional co po a e inno a ion
pa ne ships and Jesus and Jugend (2023), who ound ha s akeholde
in e ac ions and collabo a ions enhance knowledge dissemina ion and
p omo e coope a i e inno a ions ha a e bene icial o CE.
Ou esul s also demons a e ha OI s eng hens us ac oss he
supply chain, imp o ing access o ade c edi and suppo ing CE p ac-
ices. This is consis en wi h he DSM case s udy by Kennedy e al.
(2017). The RBV pe spec i e indica es ha access o comme cial c edi
ep esen s a aluable and a e inancial esou ce ha enhances a i m’s
compe i i e ad an ages in implemen ing CE p ac ices. Addi ionally, ou
indings pa allel he s udy’s emphasis on “ echnology scou ing” as a
c i ical s ep be o e launching ci cula p ojec s, ein o cing ou esul s
con i ming ha OI expands GKRs by enabling i ms o collabo a e wi h
uni e si ies, esea ch ins i u ions, and indus y pa ne s. This enables
companies o acqui e g een pa en s o ob ain pa en licenses, he eby
s eng hening GKRs (Asna Asha i e al., 2023; Kwon e al., 2023; Zhang
e al., 2023).
We also ind ha CE policies posi i ely mode a e he ela ionship
be ween OI and co po a e engagemen in CE p ac ices, pa icula ly in
egions wi h highe indus ializa ion le els. This aligns wi h Casse a
e al. (2023), who a gued ha go e nmen incen i es and egula o y
amewo ks encou age p oduc and p ocess inno a ions h ough
ex e nal collabo a ions, u he alida ing he mode a ing e ec s
obse ed in ou s udy. CE policies es ablish a mo e a o able en i on-
men o CE adop ion h ough mechanisms such as subsidies, ax
Table 4
Robus ness checks esul s.
(1) (2) (3) (4)
Va iables CEP CEP CEP CEP
Open_inno _ a io 2.1755***
(0.2583)
Open_inno 1 0.9307***
(0.0698)
Open_inno _lag1 0.9863***
(0.0919)
Open_inno 1.0325***
(0.0758)
con ol_ a YES YES YES YES
yea e YES YES YES YES
i m e YES YES YES YES
N 40219 40219 31788 39917
2_w 0.247 0.249 0.248 0.247
No e. The alue o -s a is ics epo ed in pa en heses; ***, **, * indica e ha he
bila e al es is signi ican a 1 %, 5 %, and 10 %, espec i ely.
X. Yin e al.
Jou nal o Inno a ion & Knowledge 10 (2025) 100702
8
