Liu, Yang; Cheng, Jing; Dai, Jingjing
A icle
Biased inno a ion and ne wo k e olu ion: Digi al d i e o
g een inno a ion o manu ac u ing in China
Jou nal o Applied Economics
P o ided in Coope a ion wi h:
Uni e si y o CEMA, Buenos Ai es
Sugges ed Ci a ion: Liu, Yang; Cheng, Jing; Dai, Jingjing (2024) : Biased inno a ion and ne wo k
e olu ion: Digi al d i e o g een inno a ion o manu ac u ing in China, Jou nal o Applied
Economics, ISSN 1667-6726, Taylo & F ancis, Abingdon, Vol. 27, Iss. 1, pp. 1-31,
h ps://doi.o g/10.1080/15140326.2024.2308951
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/314255
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/4.0/
Jou nal o Applied Economics
ISSN: (P in ) (Online) Jou nal homepage: www. and online.com/jou nals/ ecs20
Biased inno a ion and ne wo k e olu ion: digi al d i e
o g een inno a ion o manu ac u ing in China
Yang Liu, Jing Cheng & Jingjing Dai
To ci e his a icle: Yang Liu, Jing Cheng & Jingjing Dai (2024) Biased inno a ion and ne wo k
e olu ion: digi al d i e o g een inno a ion o manu ac u ing in China, Jou nal o Applied
Economics, 27:1, 2308951, DOI: 10.1080/15140326.2024.2308951
To link o his a icle: h ps://doi.o g/10.1080/15140326.2024.2308951
© 2024 The Au ho (s). Published by In o ma
UK Limi ed, ading as Taylo & F ancis
G oup.
Published online: 29 Jan 2024.
Submi you a icle o his jou nal
A icle iews: 731
View ela ed a icles
View C ossma k da a
Ci ing a icles: 2 View ci ing a icles
Full Te ms & Condi ions o access and use can be ound a
h ps://www. and online.com/ac ion/jou nalIn o ma ion?jou nalCode= ecs20
RESEARCH ARTICLE
Biased inno a ion and ne wo k e olu ion: digi al d i e o
g een inno a ion o manu ac u ing in China
Yang Liu
a
, Jing Cheng
b
and Jingjing Dai
a
a
School o Economics and Managemen , Kunming uni e si y, Kunming, China;
b
School o Cons uc ion
Enginee ing, Yunnan Ag icul u al Uni e si y, Kunming, China
ABSTRACT
The s udy aims o explo e he spa ial associa ion ne wo k cha ac-
e is ics o biased g een inno a ion in he manu ac u ing sec o and
i s co e d i e s. This s udy cons uc s a Malmquis -Luenbe ge
decomposi ion index model o iden i y he inpu and ou pu biases
o g een echnological inno a ion (GIIM and GIOM) in he manu-
ac u ing indus y. This s udy uses a modi ied g a i y model and
social ne wo k analysis me hod o conduc a obus assessmen o
GIIM spa ial associa ion ne wo k o 30 p o inces in China om 2012
o 2021. The esul s show: (1) The GIIM associa ion ne wo k s uc-
u e is s able and has good accessibili y, wi h close connec ions
be ween p o inces and blocks, and signi ican spillo e e ec s
be ween p o inces. (2) The egional ne wo k shows a “co e-
pe iphe y” spa ial a ia ion, wi h he co e a ea expanding and he
pe iphe al a ea sh inking. (3) The digi al ans o ma ion cha ac e -
is ics o he ne wo k componen s and he in ensi y o en i onmen-
al egula ion ha e a signi ican impac on GIIM.
ARTICLE HISTORY
Recei ed 3 Sep embe 2023
Accep ed 10 Janua y 2024
KEYWORDS
social ne wo k analysis;
spa ial and e olu iona y
analysis; biased g een
inno a ion; digi al
ans o ma ion
1. In oduc ion
The manu ac u ing sec o has been ins umen al o China’s economic expansion, con-
ibu ing nea ly 30% o he na ion’s GDP in 2021, a no able p opo ion compa ed o
o he majo global economies, as epo ed by he Na ional Bu eau o S a is ics. Despi e
his success, he sec o cas s a long shadow o en i onmen al conce ns. I is a majo
consume o ene gy and a majo emi e o ca bon, accoun ing o o e 30% o China’s
o al ene gy consump ion and app oxima ely 35% o i s o al ca bon emissions as o 2020.
These s a is ics unde sco e he u gen need o a sus ainable shi (W. Chen e al., 2017;
Qin e al., 2021). In alignmen wi h he “Made in China 2025” s a egy, g een de elop-
men has been iden i ied as a ounda ional p inciple o he manu ac u ing indus y,
ad oca ing o he widesp ead implemen a ion o g een manu ac u ing p ac ices o
acili a e he indus y’s ecological ans o ma ion (Wübbeke e al., 2016).
A he hea o his ans o ma i e agenda is g een echnological inno a ion (Yuan &
Xiang, 2018), which, h ough he adop ion o eco- iendly p oduc ion echnologies, has
CONTACT Jing Cheng [email p o ec ed] School o Cons uc ion Enginee ing, Yunnan Ag icul u al Uni e si y,
No.452, Feng Yuan Road, Pan Long Dis ic , Kunming, China
JOURNAL OF APPLIED ECONOMICS
2024, VOL. 27, NO. 1, 2308951
h ps://doi.o g/10.1080/15140326.2024.2308951
© 2024 The Au ho (s). Published by In o ma UK Limi ed, ading as Taylo & F ancis G oup.
This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion-NonComme cial License (h p://
c ea i ecommons.o g/licenses/by-nc/4.0/), which pe mi s un es ic ed non-comme cial use, dis ibu ion, and ep oduc ion in any medium,
p o ided he o iginal wo k is p ope ly ci ed. The e ms on which his a icle has been published allow he pos ing o he Accep ed
Manusc ip in a eposi o y by he au ho (s) o wi h hei consen .
he po en ial o signi ican ly cu b ene gy consump ion and pollu ion. Such inno a ions
no only p omise enhanced ene gy e iciency and cos educ ions bu also bols e he
long- e m compe i i eness o he manu ac u ing sec o (He ing & Roy, 2007). In he
con ex o a de eloping na ion like China, he pi o owa ds g een inno a ion is key,
p omising en i onmen al imp o emen while also ca ing ou no el ma ke and g ow h
a enues (Baloch e al., 2021). The ansi ion o g een echnological inno a ion necessi-
a es a depa u e om adi ional esou ce-in ensi e me hods o hose ha p io i ize
echnology and in elligen inpu s. Inno a ion wi h an en i onmen al bias is essen ial in
s ee ing his change. He and Wang’s (2015) esea ch co obo a es ha inno a ion wi h
such a bias can imp o e co po a e en i onmen al pe o mance and os e sus ainable
de elopmen (He & Wang, 2015). In oday’s clima e, di ec ing inno a ion esou ces
owa d he de elopmen and implemen a ion o g een echnologies is impe a i e o he
g een economic ans o ma ion o China.
Mo eo e , he inno a ion o g een echnology in he manu ac u ing sec o is
embedded wi hin a dynamic spa ial ne wo k, in luenced by an in e play o policy,
ma ke , echnology, and en i onmen al ac o s (Malecki & Edwa d, 2008). A i m’s
inno a i e endea ou s a e signi ican ly shaped by i s egional inno a ion ecosys em
(Aud e sch e al., 2022; Radziwon e al., 2022). Wi h ad ancemen s in in o ma ion and
digi al echnologies, adi ional geog aphical limi a ions a e becoming less p onounced,
leading o a complex ne wo k o c oss- egional spa ial in e ac ions wi hin China. This
s udy hypo hesizes ha g een inno a ion in China’s manu ac u ing sec o is no only
con ingen upon he di ec ion o g een echnology inno a ion bias bu is also cha ac e -
ized by a p onounced spa ial dependency.
To sum up, his esea ch endea ou s o dissec he inclina ion owa ds g een echno-
logical inno a ion wi hin he manu ac u ing ealm, elucida ing i s spa ial ne wo k
dynamics and iden i ying he ca alys s behind i . By cla i ying hese aspec s, we seek o
p o ide a clea e pic u e o he oppo uni ies and challenges con on ing China’s man-
u ac u ing sec o on i s pa h o a g een u u e.
The s uc u e o he s udy is as ollows: Sec ion 2 p o ides a comp ehensi e li e a u e
e iew, con ex ualizing ou s udy wi hin he exis ing body o wo k on bias in echnolo-
gical inno a ion and he ne wo ks wi hin inno a ion spaces. Sec ion 3 ou lines he
esea ch me hodology and delinea es he da a sou ces u ilized. Sec ion 4 p esen s he
co e analy ical indings along wi h pe inen discussion, which ocuses on he spa ial
co ela ion o inpu -biased g een inno a ion in he manu ac u ing sec o . Sec ion 5
conduc s an empi ical analysis o d i e s. Finally, Sec ion 6 summa izes he p ima y
conclusions, discussions and o e s co esponding policy ecommenda ions.
2. Li e a u e e iew
2.1. Resea ch ela ed o biased echnological inno a ion
Da id and an de Klunde (1965) s udy was a seminal wo k in gauging he ajec o y o
echnological p og ess, using he CES p oduc ion unc ion o e alua e he ac o bias o
echnology (Da id & an de Klunde , 1965). La e , Klump in 2012 employed he
s anda dized supply-side sys em app oach o quan i y he o ien a ion o echnological
inno a ions (Klump e al., 2012). Howe e , as he complexi y o inpu ac o s g ew, he
2Y. LIU ET AL.
ixed elas ici y subs i u ion o he CES unc ion ell ou o a o , wi h he T anslog
p oduc ion unc ion becoming he p e e ed ool o schola s o measu e he di ec ion-
ali y o echnological p og ess. Fo example, Ka an il and Yeddi -Tamsamani (2010)
u ilized he T anslog p oduc ion unc ion o assess ene gy- ocused echnological inno a-
ions in F ance, pinpoin ing ene gy p ices as he key de e minan o di ec ional bias
(Ka an il & Yeddi -Tamsamani, 2010). Subsequen esea ch, s a ing wi h Fa e e al.,
began inco po a ing nonpa ame ic me hods o measu e di ec ional bias (Fä e e al.,
1997). This in ol ed analyzing shi s in he p oduc ion on ie o e ime ha a e non-
p opo ional, leading o al e a ions in he ma ginal ou pu a ios o a ious ac o s.
A common echnique is he DEA-Malmquis index me hod, which e alua es echnolo-
gical p og ess om bo h inpu and ou pu dimensions, examining he in luence o i s
di ec ional componen s on o al ac o p oduc i i y (P.-C. Chen & Yu, 2014; Fä e e al.,
1997; Peng e al., 2019; Webe & Domazlicky, 1999). The Malmquis index me hod
mi iga es he subjec i e bias ha can a ise om p oduc ion unc ion assump ions and
e eals he impac o he o ien a ion o echnological inno a ion on di e en inpu s.
Ye , when i comes o g een echnological inno a ion, i ’s impe a i e o acknowledge
he a ie y o inpu ac o s and he po en ial o en i onmen ally de imen al ou pu s.
Chinese esea che s, like Yang e al. (2019), ha e inno a ed by in eg a ing he SBM
di ec ional dis ance unc ion wi h he Malmquis -Luenbe ge index, o mula ing a no el
me hodology o gauge g een echnological p og ess.
Building upon his ounda ion, ou s udy ac o s in non-desi ed ou pu s and unde -
akes a comp ehensi e decomposi ion o he Malmquis -Luenbe ge index. This allows
us o measu e he in e g oup bias o g een echnological inno a ion wi hin he con ex o
esou ce and en i onmen al impac s, examining bo h he inpu s and he ou pu s. This
mul idimensional app oach p o ides a mo e nuanced unde s anding o he g een inno-
a ion ajec o y, e lec ing he in ica e balance be ween economic p og ess and en i -
onmen al s ewa dship.
2.2. Resea ch ela ed o inno a ion ne wo ks
The ield o spa ial ne wo k analysis eme ged om Cas ells’ Space o Flows heo y, which
posi s ha socie y is a spa ial o m cons i u ed by he lows o a ious elemen s (Cas ells
& Ca doso, 2006). Building on his, he globaliza ion and wo ld ci ies esea ch ne wo k
has deepened he s udy o spa ial ne wo ks by linking elemen s such as in as uc u e and
co po a e o ganiza ion wi h spa ial da a (De udde & Taylo , 2021). Schola s like Roge s
(2004) and Feldman (2016) ha e u he en iched his domain by in oducing inno a ion
in o spa ial ne wo k s udies, unco e ing he signi ican impac o ne wo k s uc u es on
he di usion o inno a ion. Fo ins ance, Huggins e al. (2023) examined he ela ionship
be ween he egional concen a ion and dispe sion o inno a i e agen s wi hin speci ic
indus ies and hei capaci y o inno a ion. Resea ch on he spa ial e ec s o g een
inno a ion equen ly employs spa ial econome ic me hods, u ilizing ools like he
Mo an’s I (Gai e al., 2022), ke nel densi y es ima ion (P. Zhao e al., 2023), and Theil
index (N. Zhao e al., 2021) o in es iga e he spa io empo al e olu ion and di e en ia-
ion o g een inno a ion (Liang e al., 2022; Xuhui & Yi ao, 2023; Zhou e al., 2021).
JOURNAL OF APPLIED ECONOMICS 3
Recen s udies on g een inno a ion p edominan ly ely on “a ibu e da a”,
which, as Z. Chen e al. (2022) no e, inadequa ely cap u e he in e ac i e mechan-
isms be ween egions. Analysing spa ial associa ion ne wo ks om a ela ional
iewpoin , as ad oca ed by CHONG and QIN (2017), yields mo e insigh ul
indings han me ely examining “a ibu e da a” h ough adi ional spa ial econo-
me ic me hods. This ela ional pe spec i e e eals a consensus in he li e a u e:
egions wi h highe inno a ion e iciency o en occupy cen al posi ions in hese
ne wo ks. These egions le e age loca ional ad an ages o access ex ensi e inno-
a i e esou ces, enhancing hei inno a i e e iciency (Feng e al., 2022; Guan
e al., 2015). In con as , egions wi h lowe inno a ion e iciency bene i less
om ne wo k e ec s, leading o une en egional inno a ion de elopmen (Min
e al., 2020).
In he con ex o China’s sus ainable de elopmen goals, he equi able egional
dis ibu ion o g een inno a ion e iciency is pa amoun . Howe e , esea ch by
SUN e al. (2022) indica es a pe sis en co e-pe iphe y s uc u e in g een echno-
logical inno a ion, poin ing o a signi ican ela ionship be ween di ec ed in es -
men in g een echnology inno a ion elemen s and inno a ion ou comes. This
ela ionship, coupled wi h he inc easing deg ee o in e egional connec i i y,
sugges s ha a egion’s g een inno a ion p opensi y is inc easingly in luenced
by i s neighbou s’ ac i i ies.
Despi e he ecogni ion o spa ial co ela ion in g een inno a ion, exis ing
esea ch has sho comings ha necessi a e comp ehensi e esolu ion. T adi ional
spa ial explo a o y analyses and econome ic models p ima ily ocus on he
he e ogenei y in spa ial dis ibu ion due o geog aphical p oximi y. These
app oaches, howe e , o e look he analysis o in e nal biases and egional holis ic
pe spec i es, he eby cons aining he unde s anding o he cha ac e is ics and
gene a i e mechanisms o spa ial co ela ion ne wo ks. The p edominan ocus
on a ibu e da a, wi h scan explo a ion o ela ional da a in complex spa ial
ne wo ks, leads o inadequa e di e en ia ion o he oles, unc ions, and mechan-
isms o di e en egions wi hin he g een inno a ion bias ne wo k. The e o e, in
he ace o une en g een inno a ion e iciency and s eng hened spa ial co ela-
ions, employing social ne wo k analysis o e s a mo e comp ehensi e explo a ion
o in e egional ela ionships. This me hod no only un a els he mechanisms o
spa ial ela ionships bu also suppo s mo e a ional inpu alloca ion, p o iding
insigh s o p omo ing balanced de elopmen in g een inno a ion.
This s udy examines 30 p o incial-le el adminis a i e egions, employing
a Malmquis -Luenbe ge decomposi ion index model o assess he bias in g een
inno a ion in manu ac u ing (GIBM) om 2012 o 2021. GIBM e e s o
a endency, inclina ion, o p e e ence ha is embedded wi hin eco-inno a ion.
Addi ionally, i adop s a modi ied g a i y model o cons uc an inpu biases o
g een echnological inno a ion (GIIM) spa ial ma ix, a spa ial ma ix designed o
measu e he egional dispa i ies in g een echnological inpu s biases. This esea ch
analyses he cha ac e is ics o he spa ial ne wo k s uc u e o GIIM using social
4Y. LIU ET AL.
ne wo k me hods. Fu he mo e, QAP eg ession is u ilized o in es iga e he
d i e s o GIIM spa ial co ela ion ne wo ks, he eby elucida ing he o ma ion
mechanism o hese spa ial co ela ion ne wo ks (K acka d , 1987).
3. Ma e ials and me hods
3.1. Measu emen o he GIBM
3.1.1. Cons uc ion o he indica o sys em
D awing upon ele an li e a u e (Egilmez e al., 2013; Sims e al., 1974; Zheng
e al., 2021), his s udy de ines he key componen s in he con ex o manu ac u -
ing p oduc ion as ollows: Inpu ac o s a e iden i ied as ou key elemen s –
capi al, land, labou , and ene gy. The desi ed ou pu is concep ualized p ima ily
as economic ou pu indica o s. Howe e , his s udy also includes echnological
inno a ion in he desi ed ou pu , e lec ing a mo e ealis ic app oach o con em-
po a y demands. The non-desi ed ou pu ypically encompasses was e emissions
(Kaneko & Managi, 2004), including a ange o pollu an s such as Chemical
Oxygen Demand (COD), ammonia emission, CO2 emissions, sul u dioxide,
ni ogen oxide, was ewa e , and solid was e (Liu e al., 2022; Yang e al., 2019;
B. Zhang e al., 2022). De ailed indica o s a e ou lined in Table 1.
3.1.2. Measu emen me hod
We assume he e a e N decision-making uni s (DMUs) in he manu ac u ing, each
u ilizing W ypes o inpu s, p oducing Q ypes o expec ed ou pu s, and O ypes o
unexpec ed ou pu s in each pe iod. Fo he n h n¼1...Nð Þ DMU in pe iod
¼1;2...;Tð Þ, he se o inpu ac o s is deno ed as x
n¼x
1n;...;x
Wn
�, he se o
expec ed ou pu ac o s as y
n¼y
1n;...;y
Qn
� �, and he se o unexpec ed ou pu ac o s
as b
n¼b
1n;...;b
On
�. I he Di ec ional Dis ance Func ion (DDF) sa is ies:
D
ox ;y ;b ;y ;b
�¼max β
Table 1. GIBM e alua ion index sys em.
Guideline Laye Indica o Laye P oxy Indica o s
Inpu ac o s Capi al inpu In es men in ixed asse s in manu ac u ing (billion yuan)
Labo inpu Employmen in manu ac u ing (million people)
Land inpu Indus ial land a ea (km
2
)
Ene gy inpu Comp ehensi e ene gy consump ion pe uni o indus ial added alue
( s anda d coal/million yuan)
Expec ed
Ou pu
Economic ou pu Indus ial added alue as a sha e o GDP (%)
Technology ou pu Numbe o g een pa en s o manu ac u ing companies
Unexpec ed
Ou pu
COD emissions COD emission pe uni o indus ial added alue ( /million yuan) (%)
Ammonia emission Ammonia ni ogen emission pe uni o indus ial added alue
( /100 million yuan) (%)
Sul u dioxide emission SO
2
emissions pe uni o indus ial added alue/( /billion yuan) (%)
Ni ogen oxide emissions NO
x
emission pe uni o indus ial added alue/( /billion yuan) (%)
Was ewa e emissions Was ewa e emission pe uni o indus ial added alue/( /billion yuan) (%)
Solid was e p oduc ion Solid was e gene a ion pe uni o indus ial added alue/( /yuan) (%)
JOURNAL OF APPLIED ECONOMICS 5
P
N
n¼1
z
nx
wn �x
wn;w¼1;2;���;W
P
N
n¼1
z
ny
wn �1þβð Þy
qn;q¼1;2;���;Q
P
N
n¼1
z
nb
wn ¼1βð Þb
on;o¼1;2;���;O
z
n�0;n¼1;2;. . . ;N
8
>
>
>
>
>
>
>
>
>
<
>
>
>
>
>
>
>
>
>
:
(1)
Whe e β gauges he dis ance o he e alua ed DMU o he p oduc ion on ie in he
di ec ion o y ;b
�:By in eg a ing he cons ain condi ion o Equa ion (1) wi h he
Malmquis -Luenbe ge (ML) index me hod, he ML index can be decomposed in o
e iciency change (ΔT) and echnological change (ΔTE). Fu he , he ΔTE index is
decomposed o ob ain he inpu -biased o g een echnological inno a ion in manu ac u -
ing (GIIM) and he ou pu bias o g een echnological inno a ion (GIOM) (Fä e e al.,
1997). The speci ic calcula ion me hod is as ollows:
M
0x ;y ;x þ1;y þ1
�¼D
0x þ1;y þ1
�=D
0x ;y
ð Þ; ¼1;...;T1
¼Δ T x þ1;y þ1
��Δ TE x ;y ;x þ1;y þ1
�
¼D
0x þ1;y þ1
ð Þ
D þ1
0x þ1;y þ1
ð Þ
� ��D þ1
0x þ1;y þ1
ð Þ
D
0x ;y
ð Þ
� �:
GIIM x ;y ;x þ1
�¼D þ1
iy ;x þ1
ð Þ
D
iy ;x þ1
ð Þ =D þ1
iy ;x
ð Þ
D
iy ;x
ð Þ
� � (2)
GIOM y ;x þ1;y þ1
�¼D
0x þ1;y þ1
ð Þ
D þ1
0x þ1;y þ1
ð Þ=D
0x þ1;y
ð Þ
D þ1
0x þ1;y
ð Þ
� �
Fac o inpu bias is measu ed based on he change in he ma ginal a e o subs i u ion
o ac o s (Webe & Domazlicky, 1999). This s udy examines he ene gy subs i u ion bias
(ESB) index o g een echnology inno a ion in manu ac u ing o de e mine he ne wo k
node a ibu es. The ESB is exp essed as ollow:
ESBC;E¼C þ1
E þ1=C
E 1
� ��GIIM 1ð Þ (3)
whe e C e e s o capi al inpu and E e e s o ene gy inpu . When he e is echnological
inno a ion om pe iod o þ1, C þ1
E þ1=C
E is he a io o he ma ginal subs i u ion a es o
ac o s C and E om s age o þ1, e lec ing he ac o changes in inno a ion. When
GIIM >1, C þ1
E þ1=C
E >1, ha is ESBI;J>0, i indica es he ene gy-sa ing g een echnology
inno a ion.
3.2. Ne wo k cons uc ion and ne wo k analysis me hods
3.2.1. Ne wo k cons uc ion
The iden i ica ion o spa ial linkages is c ucial o analyzing he spa ial linkage ne wo k o
GIBM in China using social ne wo k me hodologies (Bo ga i & Fos e , 2003; Cassi e al.,
6Y. LIU ET AL.
2012). Gi en ha spa ial linkages o en diminish wi h inc easing geog aphical dis ance –
a phenomenon known as “dis ance decay” highligh ed by Basile e al. (2012)— he close
he geog aphical p oximi y, he s onge he linkages end o be. The e o e, his s udy
p oposes he use o a modi ied g a i y model o cons uc he spa ial associa ion ne wo k.
This app oach in eg a es conside a ions o g een manu ac u ing wi h economic and
geog aphical dis ances, he eby o e ing a mo e nuanced unde s anding o he spa ial
associa ion cha ac e is ics. Addi ionally, he economic dis ance be ween egions is
a signi ican ac o in luencing hese linkages. Inco po a ing popula ion and economic
scale in o he g a i y model (Кузнецов e al., 2014), allows o a mo e accu a e depic ion
o he spa ial e olu ion ends. In his esea ch, we adap he g a i y model o de e mine
he spa ial co ela ion ne wo k o g een manu ac u ing among di e en egions. The
speci ic calcula ion o mula employed is ou lined as ollows:
Fij ¼Kij i i i i i i i i i i i i i i
PiGiMi
3
p i i i i i i i i i i i i i i
PjGjMj
3
p
D2
ij
;Kij ¼Mi
MiþMj
;Dij ¼dij
gigj
(4)
In Equa ion (4): Fij deno es he spa ial co ela ion s eng h (g a i a ional alue) o
GIBM be ween p o inces i and j; Mi and Mj deno e he g een manu ac u ing de elop-
men indexes o p o inces i and j, espec i ely; Kij deno es he con ibu ion a e o
p o ince i o Fij; dij is he geog aphic dis ance be ween p o inces i and j; Gi and Gj
deno e he le el o economic de elopmen o p o inces i and j, as measu ed by he o al
GDP; gi and gj a e he pe capi a GDP o he wo egions, espec i ely.
Building upon his amewo k, his s udy acknowledges he p esence o a h eshold
alue in he s eng h o spa ial ela ionships. Consequen ly, his s udy p oposes using he
a e age alue o each ow in he associa ion s eng h ma ix de i ed om he g a i y
model as his h eshold alue. Following his, a 0–1 ma ix is cons uc ed, de e mining
he p esence o absence o associa ion ela ionships. This p ocess culmina es in he
o ma ion o a di ec ed bina y spa ial associa ion ma ix.
3.2.2. Social ne wo k analysis me hods
This s udy del es in o he spa ial associa ion ne wo k o GIBM, employing social ne wo k
analysis as a amewo k (F i ze e al., 2018). In examining he o e all s uc u e o he
ne wo k, we ocus on se e al key cha ac e is ics. Ne wo k Densi y (ND) is used o gauge
he complexi y o he ela ionships wi hin he g een manu ac u ing associa ion ne wo k.
I e lec s he in icacy o connec ions among a ious nodes. The numbe o nodes and
he quan i y o ela ionships wi hin he ne wo k indica e he ne wo k’s ele ance (NR),
which is a measu e o he GIBM ne wo k s uc u e’s s abili y. Ne wo k E iciency (NE) is
employed o quan i y he cha ac e is ics o GIBM associa ion channels. Fu he mo e,
Ne wo k hie a chy (NH) is analysed o unde s and he deg ee o asymme ic accessibili y
wi hin he ne wo k. This me ic is c ucial in e alua ing he hie a chical s uc u e o
access o esou ces and in o ma ion. The p esence o small-wo ld cha ac e is ics wi hin
he ne wo k is sc u inized o assess he e iciency and accessibili y o esou ce dissemina-
ion ac oss he ne wo k. The calcula ions o ND, NR, NE and NH is as p esen ed in
Eq. 5–8:
ND ¼N=TN �TN 1ð Þ½ � (5)
JOURNAL OF APPLIED ECONOMICS 7
The indica o s in Figu e 7 collec i ely demons a e no only he quan i a i e g ow h o
he ne wo k bu also i s quali a i e de elopmen . The expansion in ne wo k densi y and
ela ionships signi ies a ma u ing landscape o GIIM in China, e lec ing a na ionwide
shi owa ds mo e collabo a i e and in e connec ed app oaches in his ield. This end
is a posi i e indica o o he p og ess in he applica ion and in eg a ion o g een
echnologies ac oss he p o inces, ma king a signi ican s ep owa ds achie ing sus ain-
able de elopmen goals.
The pape u he calcula es key me ics such as he deg ee o co ela ion,
ne wo k hie a chy, and ne wo k e iciency wi hin he GIIM spa ial co ela ion
ne wo k. As de ailed in Table 3, he ne wo k co ela ion deg ee was consis en ly
measu ed a 1 h oughou he s udy pe iod. This consis ency sugges s a ela i ely
s able ne wo k s uc u e, despi e mino luc ua ions in he numbe o ne wo k
ela ionships. Such s abili y indica es good ne wo k accessibili y among he p o-
inces, cha ac e ized by s ong spa ial co ela ions and spillo e e ec s, implying
ha mos p o inces ha e es ablished s able coope a i e ela ionships in he ealm
o GIIM.
Addi ionally, he ne wo k hie a chy deg ee, measu ed a 0, e eals ha he GIIM
spa ial co ela ion ne wo k ope a es on a ela i ely la hie a chical scale a he in e -
p o incial le el. This absence o a s ic hie a chical o de and connec i i y ba ie s
unde lines a signi ican syne gis ic e ec among he p o inces, acili a ing collabo a i e
and egali a ian in e ac ions in g een inno a ion.
Figu e 7. Ne wo k densi y and associa ion om 2012 o 2021.
Table 3. Connec edness, hie a chy and e iciency.
yea 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021
Connec edness 1 1 1 1 1 1 1 1 1 1
Hie a chy 0 0 0 0 0 0 0 0 0 0
E iciency 0.5394 0.4901 0.4557 0.4286 0.4187 0.4212 0.3768 0.3842 0.3227 0.3128
14 Y. LIU ET AL.
The o e all end o declining ne wo k e iciency, wi h dis inc phase cha ac e is ics,
indica es ha he channels o GIIM spa ial spillo e ha e expanded du ing he s udy
pe iod. This expansion has os ensibly enhanced he o e all s abili y o he ne wo k,
sugges ing a ma u ing and mo e in e connec ed GIIM landscape.
4.4. Ne wo ked small-wo ld ea u es
In his s udy, we cons uc ed a andom ne wo k ha ma ches he scale and densi y o
eal-wo ld co ela ion ne wo ks annually. We cha ac e ized he connec i i y be ween
ne wo k nodes using me ics such as ne wo k clus e ing coe icien , a e age dis ance, and
ne wo k diame e . These analyses a e illus a ed in Figu e 8.
F om 2012 o 2021, he ne wo k’s clus e ing coe icien emained ela i ely s able,
luc ua ing be ween 0.6 and 0.8. This indica es a s ong deg ee o connec i i y among
nodes and hei adjacen nodes. The spa ial co ela ion ne wo k o he GIIM main ained
a balanced s a e o connec i i y, aligning wi h he ends obse ed in he GIIM spa ial
co ela ion ne wo ks depic ed in Figu es 4–6.
The a e age pa h leng h a ied be ween 1.544 and 1.321, showing a yea ly dec easing
end. This dec ease sugges s ha p o inces ha e been le e aging he ne wo k s uc u e
o mo e e ec i e collabo a ion in GIIM, he eby educing ne wo k edundancy and
enhancing o e all e iciency.
Rema kably, he ne wo k diame e consis en ly emained a 2, signi ican ly lowe han
he o al numbe o 30 membe s in he o e all spa ial ne wo k. This consis ency high-
ligh s he ne wo k’s e icien s uc u e, whe e he dis ance be ween any wo nodes is
ela i ely small.
O e all, he GIIM spa ial co ela ion ne wo k exhibi s cha ac e is ics o a “small-
wo ld” phenomenon, indica i e o excellen ne wo k connec i i y. This phenomenon
allows o he apid dissemina ion o a p o ince’s le el o GIIM o o he s, he eby
Figu e 8. E olu ion o small-wo ld cha ac e is ics o spa ially linked ne wo ks.
JOURNAL OF APPLIED ECONOMICS 15
Figu e 9. Cen ali y analysis in 2012.
Figu e 10. Cen ali y analysis in 2016.
16 Y. LIU ET AL.
imp o ing he spa ial co ela ion ne wo k s a us o p o inces wi h lowe GIIM le els.
These indings unde sco e he i al ole o c oss-p o incial and c oss- egional collabo a-
ion in os e ing he de elopmen and dissemina ion o GIIM, highligh ing he in e -
connec ed na u e o g een inno a ion e o s ac oss China.
4.5. Ego-ne wo k cha ac e is ics and dynamic e olu ion end
4.5.1. Cen ali y analysis
To elucida e he posi ions and unc ions o each p o ince wi hin he GIIM spa ial
associa ion ne wo k, his s udy quan i a i ely assesses h ee pi o al ne wo k me ics :
deg ee cen ali y, be weenness cen ali y, and closeness cen ali y. The yea s 2012, 2016,
and 2021 we e speci ically chosen o his dynamic analysis, wi h Figu es 9–11 illus a ing
he dis ibu ion o indi idual ne wo k cha ac e is ics ela ed o spa ial associa ions ac oss
he p o inces.
In e ms o deg ee cen ali y, egions such as Beijing, Shanghai, Jiangsu, and Zhejiang
consis en ly eme ged as cen al playe s in he GIIM ne wo k h oughou he esea ch
pe iod. The s able ye no able inc ease in bo h in-deg ee and ou -deg ee o a ious
p o inces, pa icula ly in 2021, sugges s an in ensi ying spillo e e ec . No ably, Gansu’s
highe in-deg ee compa ed o i s ou -deg ee each yea implies i bene i s mo e om he
ne wo k han i con ibu es, while Beijing and Tianjin, wi h high in-deg ees, demons a e
a s ong capaci y o abso b esou ces and p omo e local GIIM de elopmen h ough
a “siphoning e ec ”.
Figu e 11. Cen ali y analysis in 2021.
JOURNAL OF APPLIED ECONOMICS 17
The a e age be weenness cen ali y alue dec eased o e he yea s, indica ing
a g adual educ ion in bipola iza ion wi hin he GIIM ne wo k among p o inces. This
end is bene icial o enhancing coope a i e e o s and communica ion in GIIM de el-
opmen . Jiangsu, Beijing, and Shanghai, consis en ly anking high in be weenness cen-
ali y, play c ucial oles due o hei s a egic posi ions in inland communica ion
channels.
Closeness cen ali y ankings closely align wi h deg ee cen ali y, sugges ing ha
p o inces wi h signi ican posi ions and close ies o o he s also in luence he GIIM o
neighbo ing p o inces. The no iceable upwa d end in closeness cen ali y, wi h mo e
p o inces exceeding he a e age alue each yea , indica es an enhanced abili y o connec
h ough sho e pa hs and access inno a ion esou ces e ec i ely.
By combining hese h ee cen ali y indices o examine g een inno a ion in he
manu ac u ing sec o s o a ious p o inces, an inc easingly s ong in es men bias is
e iden . Regions wi h ad anced manu ac u ing, such as Shanghai, Beijing, and Jiangsu,
con inue o exe s ong con ol o e GIIM- ela ed inno a ion esou ces. Howe e , he
in ensi ica ion o spa ial ne wo k connec ions also b ings geog aphically mo e emo e
a eas in o p ominence wi hin hese ne wo ks.
4.5.2. S uc u al hole analysis
Applying he heo y o s uc u al holes o he GIIM ne wo k, his s udy measu es he
posi ions o s uc u al holes wi hin he ne wo k, using wo indica o s: E ec i e Size and
Cons ain . “E ec i e Size” gauges he ex en o a p o ince’s in luence on he o e all
GIIM ne wo k, while “Cons ain ” assesses he likelihood o a p o ince occupying
a s uc u al hole posi ion (Y. Chen, 2015).
In a ep esen a ion (Figu e 12), he s udy examines ad an ageous and disad an a-
geous nodes in he con ex o g een inno a ion in 2021. I e eals ha egions like
Beijing, Zhejiang, and Fujian, wi h la ge E ec i e Sizes and low Cons ain s, a e
Figu e 12. S uc u al hole indica o o China’s GIIM in 2021.
18 Y. LIU ET AL.
posi ioned ad an ageously in s uc u al holes. This posi ioning implies a g ea e abili y o
access di e se in o ma ion and exe con ol wi hin he ne wo k.
Con e sely, egions such as Xinjiang, Hainan, and Jilin, cha ac e ized by small
E ec i e Sizes and high Cons ain s, demons a e weake ex e nal esou ce capabili ies
and a high dependency on p o inces occupying ad an ageous s uc u al hole posi ions.
These indings indica e a dispa i y in he capaci y o le e age ne wo k bene i s o g een
inno a ion.
The s udy sugges s ha egions in ad an ageous node posi ions should capi alize on
hei ne wo k bene i s in g een manu ac u ing inno a ion. By enhancing in o ma ion
exchange wi h egions in disad an aged node posi ions, hey can acili a e a mo e
balanced and inclusi e egional de elopmen . This s a egy is c ucial o ensu ing ha
all p o inces, ega dless o hei cu en ne wo k posi ion, can con ibu e o and bene i
om ad ancemen s in g een echnology inno a ion.
4.6. E olu ion o co e-pe iphe y s uc u e
U ilizing co e-pe iphe y analysis, i is e iden ha he spa ial co ela ion ne wo k o
China’s GIIM is cha ac e ized by an expanding co e and a diminishing pe iphe y, as
illus a ed in Figu e 13. F om 2012 o 2021, he numbe o p o inces in he co e egion
s eadily inc eased, om 17 in 2012 o 19 by 2021. This inc ease e lec s a signi ican end
o g adual ex ension om he sou h-eas e n coas al p o inces o mo e pe iphe al a eas,
ma king a shi om a single-co e agglome a ion o a mo e di e se, mul i-co e expansion.
The e ol ing dis ibu ion o he co e egion signi ies no only he geog aphic sp ead o
g een inno a ion bu also he inc easing in eg a ion and collabo a ion among p o inces.
This expansion demons a es he g owing in luence and each o g een inno a ion
ini ia i es, pa icula ly om he de eloped p o inces along he eas e n coas .
Concu en ly, he diminishing pe iphe y illus a es how p e iously pe iphe al p o-
inces a e becoming inc easingly in eg a ed in o he ne wo k. This in eg a ion is d i en
by enhanced economic de elopmen and in e -p o incial connec ions, leading o igh e
ies wi hin he densi y ma ix. As a esul , pe iphe al p o inces a e g adually becoming
ac i e pa icipan s in he GIIM ne wo k, con ibu ing o and bene i ing om g een
inno a ion ini ia i es.
Figu e 13. Co e-edge s uc u e o GIIM spa ial associa ions, 2012, 2016 and 2021.
JOURNAL OF APPLIED ECONOMICS 19
The de eloped p o inces along he eas e n coas play a pi o al ole in his p ocess.
Thei expansion in di usion and adia ion scope signi ican ly p omo es he imp o e-
men o he GIIM le el in su ounding p o inces. This end is indica i e o a b oade
shi owa ds a mo e inclusi e and in e connec ed app oach o GIIM, whe e p o inces
show inc easing ini ia i e and capabili y in con ibu ing o he ne wo k s uc u e.
O e all, while he spa ial co ela ion ne wo k o China’s GIIM con inues o exhibi
a co e-pe iphe y s uc u e, he dynamics wi hin his s uc u e a e apidly e ol ing. The
inc easing numbe o co e a eas and he in eg a ion o pe iphe al egions unde sco e
a na ional end owa ds mo e in eg a ed and collabo a i e g een inno a ion e o s.
4.7. Blocks analysis
The analysis o China’s GIIM, conside ing bo h o e all and indi idual cha ac e is ics,
e eals signi ican spa ial di e en ia ion. To u he depic he in e - egional in e ac ions
and dynamic changes, his s udy selec s da a om 2012, 2016, and 2021. An i e a i e
algo i hm, which conside s segmen a ion dep h and concen a ion s anda ds, ca ego izes
Figu e 14. Co ela ion among he ou GIIM blocks in 2021.
Table 4. Classi ica ion o GIBM space- ela ed ne wo k blocks in 2012.
Segmen
Segmen
Ma ix p o inces in
each segmen
o e low
ela ion-
ships
accep ed
ela ion-
ships
Desi ed in e nal
ela ionship a io
Ac ual in e nal
ela ionship a ioI II III IV
I 2 12 24 20 2 56 55 3.45% 3.45%
II 12 21 69 42 6 123 106 17.24% 4.65%
III 23 68 40 17 12 108 109 37.93% 10.00%
IV 20 26 16 15 10 62 79 31.03% 13.89%
20 Y. LIU ET AL.
30 egions in o ou majo blocks. The esul s show R
2
alues o 0.531 (2012), 0.557
(2016), and 0.519 (2021), indica ing a gene ally good i o each yea . Figu e 14 illus a es
he in e ac i e ela ionships be ween hese ou majo blocks o he GIIM spa ial co ela-
ion ne wo k in 2021.
The block model analysis esul s, as lis ed in Tables 4–6, e eal ha when spa ial
co ela ions a e conside ed, in e -block ela ionships a e signi ican ly mo e p e alen
han in a-block associa ions. This indica es no able spa ial co ela ions and spillo e
e ec s be ween GIIM blocks. By comp ehensi ely conside ing he numbe o ecei ing
blocks, spillo e ela ionship coun s, in e nal ela ionship coun s, and he p opo ion o
in e nal ela ionships, he a ibu es o he blocks o 2012, 2016, and 2021 we e ca ego -
ized. A close examina ion o hei dynamic changes e eals:
●Block I: Weak in e nal spillo e e ec s, signi ican spillo e o o he s, and ecep ion
o spillo e om o he blocks, playing an e ec i e “b idge” ole in he spa ial
co ela ion ne wo k. Th oughou he s udy pe iod, i was consis en ly ca ego ized
as he “B oke ” block.
●Block II: The numbe o spillo e ela ionships exceeds hose ecei ed, making i
a “Ne Spillo e ” block h oughou he s udy pe iod. This block no only mee s i s
own de elopmen needs bu also acili a es he de elopmen o o he block membe s
h ough elemen spillo e .
●Blocks III and IV: These blocks expe ienced ole e e sals du ing he s udy pe iod.
In 2012 and 2016, Block III, wi h a oughly equal numbe o ecei ed and spillo e
ela ionships and a ela i ely high p opo ion o in e nal ela ionships, was ca ego -
ized as a “Bidi ec ional Spillo e ” block, playing a “Bidi ec ional Guide” ole. In
2021, i became a “Ne In low” block, whe eas Block IV ansi ioned om a “Ne
In low”block in 2012 and 2016 o a “Bidi ec ional Spillo e ” block in 2021.
No only did he a ibu es o he blocks change, bu also he composi ion o p o inces
wi hin each block shi ed no ably. Fo ins ance, he numbe o p o inces in Block
Table 5. Classi ica ion o GIBM space- ela ed ne wo k blocks in 2016.
Segmen
Segmen
Ma ix p o inces in
each segmen
o e low
ela ion-
ships
accep ed
ela ion-
ships
Desi ed in e nal
ela ionship a io
Ac ual in e nal
ela ionship a ioI II III IV
I 7 24 39 18 3 81 81 6.90% 3.57%
II 24 33 97 33 8 154 142 24.14% 4.94%
III 39 97 57 5 13 141 139 41.38% 8.44%
IV 18 21 3 12 6 42 56 17.24% 12.50%
Table 6. Classi ica ion o GIBM space- ela ed ne wo k blocks in 2021.
Segmen
Segmen
Ma ix p o inces in
each segmen
o e low
ela ion-
ships
accep ed
ela ion-
ships
Desi ed in e nal
ela ionship a io
Ac ual in e nal
ela ionship a ioI II III IV
I 30 29 54 60 6 143 144 17.24% 4.03%
II 30 12 45 43 5 118 112 13.79% 4.07%
III 54 41 21 37 9 132 140 27.59% 6.38%
IV 60 42 41 22 10 143 140 31.03% 6.54%
JOURNAL OF APPLIED ECONOMICS 21
I inc eased, wi h Beijing, Tianjin, Jiangsu, Zhejiang, Fujian, and Shanghai being pa o he
“B oke ””block by 2021, indica ing an inc easing numbe o p o inces playing a“b idge”-
and“in e media y” ole. The a ia ions in he p o incial composi ion o he o he blocks
ac oss di e en yea s highligh he complexi y and dynamism o he GIIM spa ial co ela-
ion ne wo k.
4.8. Ne wo k s uc u e e ec analysis
To e eal he s uc u al cha ac e is ics o he GIIM spa ial co ela ion ne wo k, his s udy
empi ically examines he impac o ne wo k s uc u e on egional di e ences and le els
o GIIM om wo aspec s: he o e all ne wo k and indi idual ne wo k s uc u es. F om
he pe spec i e o he o e all ne wo k s uc u e, he a ia ion coe icien o GIIM le els
ac oss p o inces is used o measu e he in e -p o incial di e ences in GIIM le els. OLS
eg ession is conduc ed using ne wo k densi y and ne wo k e iciency, wi h he depen-
den a iables unde going loga i hmic ans o ma ion. Addi ionally, om he pe spec-
i e o indi idual ne wo k s uc u e, besides he deg ee cen ali y and be weenness
cen ali y used in he indi idual cha ac e is ics analysis, eigen ec o cen ali y is also
in oduced as an explana o y a iable. Balanced panel da a eg ession analysis is pe -
o med wi h GIIM as he dependen a iable. Loga i hmic ans o ma ion o he expla-
na o y a iables is ca ied ou o a oid ce ain mul icollinea i y issues.
4.8.1. Reg ession esul s o he whole ne wo k s uc u e
Table 7 e eals ha he eg ession coe icien s o ne wo k densi y and ne wo k e iciency
on he egional di e ences o GIIM a e 2.420 and −2.357, espec i ely, and bo h a e
signi ican a he 5% le el. This indica es ha an inc ease in ne wo k densi y and
a dec ease in ne wo k e iciency signi ican ly a ec he egional dispa i ies o GIIM,
leading o a mo e balanced spa ial dis ibu ion o GIIM le els. The po en ial easons a e
as ollows: Fi s ly, an inc ease in ne wo k densi y enhances he cohesion among p o-
inces in e ms o g een inno a ion o ien a ion, hus a oiding spa ial di e ences and
pola iza ion ends in GIIM. Secondly, a dec ease in ne wo k e iciency implies an
inc ease in c i ical nodes h oughou he ne wo k, leading o highe le els o GIIM
spillo e among p o inces. This enhances he o e all s abili y o he ne wo k, he eby
educing he ela i e di e ences in GIIM.
4.8.2. Reg ession esul s o ego-ne wo k s uc u e
Based on he cha ac e is ics o he da a, es ima ion was pe o med using panel
models, wi h e e ence o he esul s o he Hausman es o selec an app op ia e
Table 7. OLS eg ession esul s o he whole ne wo k s uc u e e ec .
Model (1) (2)
Ne wo k Densi y 2.420**(2.66)
Ne wo k E iciency −2.357**(−2.43)
In e cep ion −2.939***(−5.58) −0.569(−1.40)
R2 0.4697 9.4256
Adj.R2 0.435 0.3537
No es: (1)*, **, *** ep esen a 1%, 5%, and 10% signi icance le els, espec i ely; (2)
The igu es in () indica e he - alues.
22 Y. LIU ET AL.
andom e ec s model. The eg ession esul s a e shown in Table 8. The eg ession
coe icien s o deg ee cen ali y, be weenness cen ali y, and eigen ec o cen ali y
a e 0.109, 0.266, and 0.130, espec i ely, and a e posi i e a he 5% and 10% sig-
ni icance le els. This indica es ha he enhancemen o cen ali y in each p o ince
has a signi ican posi i e impac on he GIIM le el. The possible easons a e as
ollows: Fi s , he highe he deg ee cen ali y, he mo e di ec ela ionships
a p o ince has wi h o he p o inces, esul ing in a highe deg ee o local associa ion
and enabling p o inces o acqui e ele an esou ces mo e e icien ly, he eby
imp o ing he GIIM le el. Secondly, p o inces wi h highe be weenness cen ali y
can e ec i ely con ol he associa i e e ec s wi h o he p o inces, hus guiding he
a ional alloca ion o esou ce elemen s and enhancing he GIIM le el in he ne wo k
s uc u e egion. Finally, he highe he eigen ec o cen ali y, he close and mo e
in luen ial he co e in e -p o incial ela ionships a e, p omo ing in e -p o incial
communica ion and coope a ion, and s eadily imp o ing he GIIM le el.
5. QAP analysis o d i e s
5.1. Va iable design
The o ma ion and e olu ion o he GIIM spa ial co ela ion ne wo k a e a ibu ed o
he di e en ial esou ce endowmen s o ne wo k nodes and hei collec i e ac ions.
Resea ch, no ably by K. Gao and Yuan (2022), has e ealed ha he spa ial he e ogenei y
and egional a ia ions in China’s g een inno a ion a e linked o geog aphic spa ial
ac o s. The p oximi y in geog aphical dis ance acili a es easie ac o mobili y and
mu ual sha ing o ou comes. Ex ensi e esea ch has demons a ed he impac o go e n-
men al en i onmen al egula ion on g een inno a ion in manu ac u ing. Manu ac u ing
en e p ises a e equi ed o alloca e subs an ial esou ces o comply wi h manda o y
en i onmen al policies, ine i ably escala ing p oduc ion cos s and imposing
a signi ican bu den on g een inno a ion (Y. Zhang e al., 2018), he eby in luencing
he di ec ion o g een inno a ion ini ia i es. Howe e , wi h he inc easing in ensi y o
manda o y en i onmen al policies, manu ac u ing i ms a e compelled o adop mo e
en i onmen ally sus ainable echnologies. This shi enhances p oduc ion e iciency and
esou ce u iliza ion h ough GIIM, consequen ly educing p oduc ion cos s (J. Gao e al.,
2023). This complex in e play accen ua es he p o ound in luence o he le el o digi al
ans o ma ion and en i onmen al egula o y in ensi y on he dynamics be ween sec o s.
Table 8. Reg ession esul s o he ego-ne wo k s uc u e e ec .
Model (1) (2) (3)
Deg ee cen ali y 0.109**（2.30）
Be weenness cen ali y 0.266**（2.19）
Eigen ec o cen ali y 0.130*（1.75）
In e cep ion 0.502***（2.59）−0.196（-0.38）0.531**（2.24）
Wald 5.28** 4.82** 3.06*
R2 0.0509 0.0448 0.0427
Hausman s a is ic 0.36 0.09 1.03
FE/RE RE RE RE
No es: (1）*, **, *** ep esen a 1%, 5%, and 10% signi icance le els, espec i ely;（2）The igu es in () indica e
he - alues.
JOURNAL OF APPLIED ECONOMICS 23
Gao, J., Feng, Q., Guan, T., & Zhang, W. (2023). Unlocking pa hs o ans o ming g een
echnological inno a ion in manu ac u ing indus ies. Jou nal o Inno a ion & Knowledge, 8
(3), 100394. h ps://doi.o g/10.1016/j.jik.2023.100394
Gao, K., & Yuan, Y. (2022). Spa io empo al pa e n assessmen o China’s indus ial g een
p oduc i i y and i s spa ial d i e s: E idence om ci y-le el da a o e 2000–2017. Applied
Ene gy, 307, 118248. h ps://doi.o g/10.1016/j.apene gy.2021.118248
Guan, J., Zhang, J., & Yan, Y. (2015). The impac o mul ile el ne wo ks on inno a ion. Resea ch
Policy, 44(3), 545–559. h ps://doi.o g/10.1016/j. espol.2014.12.007
He ing, H., & Roy, R. (2007). Technological inno a ion, ene gy e icien design and he ebound
e ec . Techno a ion, 27(4), 194–203. h ps://doi.o g/10.1016/j. echno a ion.2006.11.004
He, G., & Wang, Z. (2015). Ene gy biased echnology p og ess and g een g ow h ans o ma ion
——an empi ical analysis based on 33 indus ies o China. China’s Indus ial Economy, 02,
50–62. h ps://doi.o g/10.19581/j.cnki.ciejou nal.2015.02.006
Huggins, R., Johns on, A., Munday, M., & Xu, C. (2023). Compe i ion, open inno a ion, and
g ow h challenges in he semiconduc o indus y: The case o Eu ope’s clus e s. Science and
Public Policy, 50(3), 531–547. h ps://doi.o g/10.1093/scipol/scad005
Kaneko, S., & Managi, S. (2004). En i onmen al p oduc i i y in China. Economics Bulle in, 17(2), 1–10.
Ka an il, F., & Yeddi -Tamsamani, Y. (2010). Is echnological change biased owa d ene gy? A
mul i-sec o al analysis o he F ench economy. Ene gy Policy, 38(4), 1842–1850. h ps://doi.
o g/10.1016/j.enpol.2009.11.061
Klump, R., McAdam, P., & Willman, A. (2012). The no malized ces p oduc ion unc ion: Theo y
and empi ics. Jou nal o Economic Su eys, 26(5), 769–799. h ps://doi.o g/10.1111/j.1467-6419.
2012.00730.x
K acka d , D. (1987). QAP pa ialling as a es o spu iousness. Social Ne wo ks, 9(2), 171–186.
h ps://doi.o g/10.1016/0378-8733(87)90012-8
Liang, Z., Chen, J., Jiang, D., & Sun, Y. (2022). Assessmen o he spa ial associa ion ne wo k o
g een inno a ion: Role o ene gy esou ces in g een eco e y. Resou ces Policy, 79, 103072.
h ps://doi.o g/10.1016/j. esou pol.2022.103072
Liu, S., Hou, P., Gao, Y., & Tan, Y. (2022). Inno a ion and g een o al ac o p oduc i i y in China:
A linea and nonlinea in es iga ion. En i onmen al Science and Pollu ion Resea ch, 29(9),
12810–12831. h ps://doi.o g/10.1007/s11356-020-11436-1
Malecki, P. O., & Edwa d, J. (2008). The e olu ion o echnologies in ime and space: F om
na ional and egional o spa ial inno a ion sys ems. In M. Ron (Ed.), Economy (1s ed., pp. 401–
430). Rou ledge.
Min, S., Kim, J., & Sawng, Y.-W. (2020). The e ec o inno a ion ne wo k size and public R&D
in es men on egional inno a ion e iciency. Technological Fo ecas ing and Social Change, 155,
119998. h ps://doi.o g/10.1016/j. ech o e.2020.119998
Peng, J., Xiao, J., Wen, L., & Zhang, L. (2019). Ene gy indus y in es men in luences o al ac o
p oduc i i y o ene gy exploi a ion: A biased echnical change analysis. Jou nal o Cleane
P oduc ion, 237, 117847. h ps://doi.o g/10.1016/j.jclep o.2019.117847
Qin, L., Ki ikkaleli, D., Hou, Y., Miao, X., & Tu ail, M. (2021). Ca bon neu ali y a ge o G7
economies: Examining he ole o en i onmen al policy, g een inno a ion and composi e isk
index. Jou nal o En i onmen al Managemen , 295, 113119. h ps://doi.o g/10.1016/j.jen man.
2021.113119
Radziwon, A., Rong, K., Lin, Y., Yu, J., & Zhang, Y. (2022). Explo ing egional inno a ion
ecosys ems: An empi ical s udy in China. In B. Ca he ine, B.-H. Thie y, & C. Pa ick (Eds.),
Inno a ion policies and p ac ices wi hin inno a ion ecosys ems (pp. 10–34). Rou ledge. h ps://
doi.o g/10.4324/9781003279464 .
Roge s, M. (2004). Ne wo ks, i m size and inno a ion. Small Business Economics, 22(2), 141–153.
h ps://doi.o g/10.1023/B:SBEJ.0000014451.99047.69
Sims, C. A., Lo ell, M. C., & Solow, R. M. (1974). Ou pu and labo inpu in manu ac u ing.
B ookings Pape s on Economic Ac i i y, 1974(3), 695–735. h ps://doi.o g/10.2307/2534251
30 Y. LIU ET AL.
SUN, Z., FAN, J., SUN, Y., & H, L. I. U. (2022). S uc u al cha ac e is ics and in luencing ac o s o
spa ial co ela ion ne wo k o g een science and echnology inno a ion e iciency in China.
Economic Geog aphy, 03, 33–43. h ps://doi.o g/10.15957/j.cnki.jjdl.2022.03.004
Wang, J., Wu, H., & Chen, Y. (2020). Made in China 2025 and manu ac u ing s a egy decisions
wi h e e se QFD. In e na ional Jou nal o P oduc ion Economics, 224, 107539. h ps://doi.o g/
10.1016/j.ijpe.2019.107539
Webe , W. L., & Domazlicky, B. R. (1999). To al ac o p oduc i i y g ow h in manu ac u ing:
A egional app oach using linea p og amming. Regional Science and U ban Economics, 29(1),
105–122. h ps://doi.o g/10.1016/S0166-0462(98)00013-1
Wübbeke, J., Meissne , M., Zenglein, M. J., I es, J., & Con ad, B. (2016). Made in china 2025.
Me ca o Ins i u e o China S udies Pape s on China, 2(74), 4.
Wu, F., Hu, H., Lin, H., & Ren, X. (2021). En e p ise digi al ans o ma ion and capi al ma ke
pe o mance: Empi ical e idence om s ock liquidi y (in Chinese). Jou nal o Managemen
Wo ld, 2021(7), 130–144+10.
Xuhui, L. I., & Yi ao, T. A. O. (2023). Measu emen , egional di e ences, and causes o China’
s g een and low?ca bon inno a ion de elopmen unde he “dual ca bon” goals. China
Popula ion, Resou ces & En i onmen , 33(1), 124–136.
Yang, X., Li, X., & Zhong, C. (2019). S udy on he e olu ion end and in luencing ac o s o
China’s indus ial di ec ed echnical change. The Jou nal o Quan i a i e & Technical
Economics, 04, 101–119. h ps://doi.o g/10.13653/j.cnki.jq e.2019.04.006
Yuan, B., & Xiang, Q. (2018). En i onmen al egula ion, indus ial inno a ion and g een de el-
opmen o Chinese manu ac u ing: Based on an ex ended CDM model. JOURNAL o CLEANER
PRODUCTION, 176, 895–908. h ps://doi.o g/10.1016/j.jclep o.2017.12.034
Zenglein, M. J., & Holzmann, A. (2019). In W. Claudia (Ed.), E ol ing made in China 2025 (Vol. 8,
pp. 78). Ge many: MERICS Pape s on China.
Zhang, Y., Wang, J., Xue, Y., & Yang, J. (2018). Impac o en i onmen al egula ions on g een
echnological inno a i e beha io : An empi ical s udy in China. Jou nal o Cleane P oduc ion,
188, 763–773. h ps://doi.o g/10.1016/j.jclep o.2018.04.013
Zhang, B., Yu, L., & Sun, C. (2022). How does u ban en i onmen al legisla ion guide he g een
ansi ion o en e p ises? Based on he pe spec i e o en e p ises’ g een o al ac o p oduc i i y.
Ene gy Economics, 110, 106032. h ps://doi.o g/10.1016/j.eneco.2022.106032
Zhao, N., Liu, X., Pan, C., & Wang, C. (2021). The pe o mance o g een inno a ion: F om an
e iciency pe spec i e. Socio-Economic Planning Sciences, 78, 101062. h ps://doi.o g/10.1016/j.
seps.2021.101062
Zhao, P., Lu, Z., Kou, J., & Du, J. (2023). Regional di e ences and con e gence o g een inno a ion
e iciency in China. Jou nal o En i onmen al Managemen , 325, 116618. h ps://doi.o g/10.
1016/j.jen man.2022.116618
Zheng, W.-L., Wang, J.-W., Jiang, A.-D., Rehman Khan, S. A., Yang, X.-Q., Zhang, X., &
Zhang, Z.-Y. (2021). S udy on en i onmen al pe o mance e alua ion o di e en linkage
de elopmen ypes o he logis ics and manu ac u ing indus ies conside ing he unexpec ed
ou pu . Jou nal o he Ai & Was e Managemen Associa ion, 71(8), 1025–1038. h ps://doi.o g/
10.1080/10962247.2021.1920516
Zhou, X., Yu, Y., Yang, F., & Shi, Q. (2021). Spa ial- empo al he e ogenei y o g een inno a ion in
China. Jou nal o Cleane P oduc ion, 282, 124464. h ps://doi.o g/10.1016/j.jclep o.2020.124464
Кузнецов, С., Турдаков, Д., Назар, Б., & Валерий, А. (2014). Social ne wo k analysis: Me hods
and applica ions. h ps://xueshu.baidu.com/use cen e /pape /show?pape id=
32e66bde64 b 74278db7569b85702dc&si e=xueshu_se
JOURNAL OF APPLIED ECONOMICS 31
