Integrating higher education strategies into urban cluster development: Spatial agglomeration analysis of China's key regions

Hu, Yangguang; Yang, Chuang; Ma, Jun eng
A icle
In eg a ing highe educa ion s a egies in o u ban clus e
de elopmen : Spa ial agglome a ion analysis o China's
key egions
Economies
P o ided in Coope a ion wi h:
MDPI – Mul idisciplina y Digi al Publishing Ins i u e, Basel
Sugges ed Ci a ion: Hu, Yangguang; Yang, Chuang; Ma, Jun eng (2025) : In eg a ing highe
educa ion s a egies in o u ban clus e de elopmen : Spa ial agglome a ion analysis o China's key
egions, Economies, ISSN 2227-7099, MDPI, Basel, Vol. 13, Iss. 6, pp. 1-24,
h ps://doi.o g/10.3390/economies13060167
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/329447
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/4.0/
Academic Edi o : Ral Fendel
Recei ed: 28 Ma ch 2025
Re ised: 29 May 2025
Accep ed: 4 June 2025
Published: 10 June 2025
Ci a ion: Hu, Y., Yang, C., & Ma, J.
(2025). In eg a ing Highe Educa ion
S a egies in o U ban Clus e
De elopmen : Spa ial Agglome a ion
Analysis o China’s Key Regions.
Economies,13(6), 167. h ps://doi.o g/
10.3390/economies13060167
Copy igh : © 2025 by he au ho s.
Licensee MDPI, Basel, Swi ze land.
This a icle is an open access a icle
dis ibu ed unde he e ms and
condi ions o he C ea i e Commons
A ibu ion (CC BY) license
(h ps://c ea i ecommons.o g/
licenses/by/4.0/).
A icle
In eg a ing Highe Educa ion S a egies in o U ban Clus e
De elopmen : Spa ial Agglome a ion Analysis o China’s
Key Regions
Yangguang Hu 1, Chuang Yang 2,* and Jun eng Ma 3
1School o Economics and S a is ics, Guangzhou Uni e si y, Guangzhou 510006, China; [email p o ec ed]
2School o Educa ion, Guangzhou Uni e si y, Guangzhou 510006, China
3School o Teache Educa ion, Fujian No mal Uni e si y, Fuzhou 350117, China
*Co espondence: [email p o ec ed]
Abs ac : As u baniza ion accele a es globally, highe educa ion agglome a ion (HEA)
eme ges as a c i ical mechanism o in eg a ing egional economic heo ies wi h p ac ical
s a egies, d i ing inno a ion and sus ainable de elopmen . This pape examines how HEA
p omo es inno a ion, human capi al accumula ion, indus ial es uc u ing, and equi able
income dis ibu ion ac oss 193 ci ies in he “Two T ans e se and Th ee Leng hways” u ban
clus e s om 2006 o 2020. Using dynamic panel eg ession and spa ial econome ic models,
he esul s show ha HEA yields signi ican local and spa ial spillo e bene i s, pa icula ly
in co e ci ies ha acili a e knowledge di usion and esou ce sha ing. He e ogenei y
analysis e eals ha hese posi i e spillo e s a e s onges in i s - ie , highly de eloped
clus e s and hi d- ie , ea ly-s age clus e s bu weake o e en nega i e in second- ie ,
apidly expanding egions. These spa ial e ec s g ow o e ime, e lec ing he e ol ing
pa e ns o egional in eg a ion. Theo e ically, he pape ad ances he unde s anding o
spa ial syne gy and spillo e mechanisms in HEA in u ban clus e s. P ac ically, he indings
highligh he need o ailo highe educa ion s a egies o he de elopmen al s age o each
u ban clus e o op imize esou ce alloca ion and os e inclusi e g ow h. This pape
p o ides policy insigh s o using HEA as a ca alys o coo dina ed u ban de elopmen .
Keywo ds: highe educa ion; u ban clus e ; agglome a ion economy; spa ial spillo e
e ec ; demons a ion e ec
1. In oduc ion
Despi e di e gen u baniza ion ajec o ies and de elopmen al s ages wo ldwide,
he spa ial e olu ion o u ban agglome a ions consis en ly demons a es he clus e ing o
highe educa ion ac i i ies, which ac as ca alys s o spa ial spillo e e ec s, enhancing
egional inno a ion and d i ing indus ial es uc u ing (Bo si e al.,2022;Mok & Ma gin-
son,2021;Sla o a e al.,2025). Acco ding o da a om China’s Minis y o Educa ion, he
en olmen in highe educa ion su passed 40 million s uden s by 2020
1
, wi h he majo i y
concen a ed in u ban clus e s such as he Yang ze Ri e Del a, he Pea l Ri e Del a,
and he Beijing–Tianjin–Hebei egion (Yin e al.,2022). This unde lines he c ucial ole o
u ban clus e s in he alloca ion o highe educa ion esou ces ac oss China. Simila ly, in
mos egions o Asia o Eu ope, op uni e si ies a e loca ed wi hin he main me opoli an
a eas, p o iding sus ained impe us o egional inno a ion sys ems (Chao,2014;B ennan &
Coch ane,2019;Peng & Xu,2024).
Economies 2025,13, 167 h ps://doi.o g/10.3390/economies13060167
Economies 2025,13, 167 2 o 24
As pla o ms o socioeconomic p og ess, ci ies se e as c i ical hubs o he agglom-
e a ion o highe educa ion ins i u ions. Resea che s o en emphasize he ans o ma i e
po en ial o highe educa ion agglome a ion in d i ing inno a ion and economic ansi ions
wi hin u ban clus e s (Addie e al.,2015;Lendel & Qian,2017;X. Liu e al.,2024). Some
s udies ha e placed g ea e emphasis on he in eg a ion o highe educa ion wi h local
indus ies, aiming o enhance egional coo dina ion and balanced de elopmen (J. F. Gu,
2012a;Ji e al.,2023;Zibe i e al.,2022). Fo example, leading u ban agglome a ions in de-
eloped na ions, such as he Bos on and Silicon Valley clus e s in he Uni ed S a es, le e age
uni e si ies o ancho high- ech indus ial ecosys ems, he eby eme ging as global cen e s
o inno a ion (Be ilacqua e al.,2023;Fang & Yu,2017;Ma e al.,2023;Pink-Ha pe ,2015).
F om a heo e ical s andpoin , he li e a u e equen ly d aws upon new economic geog a-
phy o in es iga e knowledge di usion, alen ne wo ks, and hei syne gis ic linkages wi h
egional de elopmen (Acs e al.,2002;Chou e al.,2023;Flo ida & Adle ,2020;Lee,2014;
Vaz,2024). Collec i ely, hese p ocesses p omo e egional educa ional in eg a ion, which in
u n os e s inno a ion, d i es indus ial ans o ma ion, and suppo s high-quali y eco-
nomic de elopmen . While subs an ial esea ch has been conduc ed on u ban clus e s and
highe educa ion, he mechanisms h ough which he spa ial e olu ion o u ban clus e s
acili a es he alignmen o highe educa ion agglome a ion wi h egional economic g ow h
and scien i ic inno a ion demands emain unde explo ed. Addi ionally, he e is limi ed
empi ical e idence ega ding he spa ial spillo e e ec s o highe educa ion agglome a ion
on socioeconomic de elopmen .
The e o e, his pape u ilizes panel da a om 193 ci ies wi hin China’s “Two T ans-
e se and Th ee Leng hways” u ban clus e s
2
o he pe iod om 2006 o 2020, aiming
o empi ically in es iga e he pa hways h ough which highe educa ion agglome a ion
con ibu es o egional socioeconomic de elopmen . The ocus is on examining he spillo e
and demons a ion e ec s in p omo ing inno a ion, enhancing human capi al, ad ancing
indus ial upg ading, and op imizing income dis ibu ion. Employing dynamic panel
eg ession and spa ial econome ic analysis, his pape seeks o add ess he ollowing co e
ques ions: wha is he impac o highe educa ion agglome a ion on egional inno a ion
and economic de elopmen , and h ough wha mechanisms does his impac mani es ? The
indings o his pape a e an icipa ed o o e aluable insigh s o he o mula ion o highe
educa ion s a egies and egional coo dina ion policies. Speci ically, he esul s unde sco e
he impo ance o aligning highe educa ion ini ia i es wi h he de elopmen al p io i ies
o speci ic egions, enabling u ban clus e s o ealize hei ans o ma i e po en ial and
p omo ing inclusi e egional p ospe i y. This s a egic alignmen is essen ial o le e aging
he capaci y o u ban agglome a ions o d i e comp ehensi e economic and social p og ess.
2. Li e a u e Re iew and Hypo hesis De elopmen
2.1. Highe Educa ion Agglome a ion in U ban Clus e s
U ban clus e s unc ion as cen al nodes o highe educa ion agglome a ion, a ac ing
g adua es o skilled p o essionals ia supe io economic oppo uni ies and public se ices,
he eby ca alyzing knowledge c ea ion and inno a ion (Go don & Kou i ,2020). Agglome -
a ion economies, as a consequence o highe educa ion clus e ing, can enhance p oduc i i y
and in igo a e egional inno a ion sys ems (H. Gu e al.,2020). While some s udies ha e
highligh ed ha high le els o concen a ion may exace ba e egional dispa i ies, a mo e
nuanced spa ial–economic lens indica es ha agglome a ion is also a dynamic mechanism
o long- e m spa ial equilib ium (Ha ison e al.,2017). This mechanism enables lagging
egions o bene i om sus ained knowledge di usion and esou ce ealloca ion, mi iga ing
inequali y o e ime.
Economies 2025,13, 167 3 o 24
Highe educa ion se es as a ounda ional componen o he knowledge economy,
shaping alen o ma ion and inno a ion capaci y. The p ocess o highe educa ion agglom-
e a ion is mani es ed h ough h ee p ima y ea u es: he spa ial concen a ion o educa-
ional esou ces, he gene a ion o economic and social bene i s, and he acili a ion o spa ial
spillo e e ec s ha ansmi inno a ion and knowledge o neighbo ing a eas (H. Liu e al.,
2024;Mohan y & Bhanumu hy,2018). These spillo e s a e c i ical channels h ough which
highe educa ion agglome a ion ad ances egional socioeconomic de elopmen .
The empi ical li e a u e u ilizes di e se me hodologies o examine hese phenomena,
ocusing on he spa ial dis ibu ion o highe educa ion alen , de e minan s o agglome a-
ion, and implica ions o egional o u ban economic dispa i ies (Aldie i e al.,2018;H.
Gu e al.,2020;Ts e ko a & Lome ,2019). O he esea ch has in es iga ed he impac s o
highe educa ion clus e ing on egional inno a ion, indus ial ans o ma ion, and u ban-
iza ion (Alde e al.,2016;Lyu e al.,2019;Pablo-Ma í & A auzo-Ca od,2020). Despi e
hese ad ances, he e is a no able gap ega ding sys ema ic empi ical in es iga ion in o he
unique spa ial ea u es o highe educa ion agglome a ion and he explici mechanisms
by which i in luences egional de elopmen ajec o ies and educes spa ial inequali y
(D ucke ,2016;Kim & Lee,2024).
Me hodologically, highe educa ion agglome a ion wi hin u ban clus e s o e s mul-
iple bene i s, such as a ac ing high-le el alen , encou aging academic ib ancy, and
acili a ing egional indus ial upg ading and compe i i eness. S a egic clus e ing allows
highe educa ion ins i u ions o op imize esou ce dis ibu ion, coun e ac ing ine iciencies
associa ed wi h dispe sed o undi e en ia ed expansion (Tleppaye e al.,2025). Howe e ,
eliance on c oss-sec ional analysis o insu icien spa ial analysis has cons ained deepe
insigh s in o hese complex dynamics (De Falco,2021). Fu he mo e, ambigui y and incon-
sis ency in empi ical conclusions ha e limi ed ac ionable policy guidance (T ipa hi,2021).
While he ex an s udies ha e con ibu ed impo an heo e ical and empi ical pe -
spec i es, he e emains a lack o sys ema ic assessmen o he scale and spillo e e ec s o
highe educa ion agglome a ion wi hin u ban clus e s. Wi hou such empi ical cla i y, i is
challenging o design e ec i e policies ha le e age highe educa ion agglome a ion as
a ca alys o egionally balanced inno a ion, economic ans o ma ion, and sus ainable
socioeconomic ad ancemen .
2.2. Spa ial Agglome a ion o Highe Educa ion
Mains eam economic heo ies s ess ha u ban agglome a ions o igina e om he
spa ial concen a ion o key p oduc ion ac o s such as labo and capi al wi hin speci ic
geog aphic uni s (Fang & Yu,2017;Zhao e al.,2025). This phenomenon, known as “ag-
glome a ion economies”, is undamen ally d i en by mechanisms o sha ing, ma ching,
and lea ning among spa ially p oxima e ac o s wi hin an u ban a ea (Billings & Johnson,
2016). T adi ional economic heo ies, based p ima ily on balanced g ow h assump ions and
diminishing e u ns, o en neglec ed he decisi e in luence o spa ial cha ac e is ics on eco-
nomic pa e ns (O a iano & Thisse,2004;T idico,2011). The eme gence o new economic
geog aphy (NEG), pionee ed by K ugman, ma ked a signi ican shi by inco po a ing
inc easing e u ns o scale and spa ial disequilib ium as cen al componen s in egional
de elopmen analysis (K ugman,1991). NEG, in pa icula , emphasizes how loca ional
ad an ages lead o he nonlinea e olu ion o “agglome a ion–dispe sion– eagglome a ion”
cycles, esul ing in mul i-equilib ium and “lumpy” spa ial con igu a ions o economic ac i -
i y (Fuji a & K ugman,2003). Cen al o his amewo k a e mechanisms o sel - ein o cing
ex e nali ies and lock-in e ec s, which explain he pe sis en economic dominance o u -
ban clus e s and co e nodes, and hei capaci y o a ac key esou ces, including highe
educa ion ins i u ions (Fuji a & Mo i,1996;Z. Yang e al.,2015;Yu e al.,2024).
Economies 2025,13, 167 4 o 24
Building on hese ounda ions, agglome a ion economics and knowledge ex e nali ies
heo y p o ide he co e analy ical logic o unde s anding highe educa ion agglome a ion.
The o me , ypi ied by Ma shall and ex ended by u ban economis s such as Glaese e al.
(1992), a gues ha spa ial concen a ion o indus ial o ganiza ion and skilled labo educes
ansac ion cos s, enhances labo ma ke pooling, and os e s mo e specialized supply
chains, he eby op imizing esou ce alloca ion and d i ing p oduc i i y gains. The la e ,
g ounded in endogenous g ow h heo y (Lucas,1988;Rome ,1986) and he geog aphy o
inno a ion (B eschi & Lissoni,2001), highligh s he cen al ole o uni e si ies and esea ch
cen e s in gene a ing knowledge spillo e s, de eloping human capi al, and s imula ing
en ep eneu ial ac i i y (Al-Youbi e al.,2021;Chen & Kenney,2007). These posi i e ex-
e nali ies, in u n, acili a e spa ial knowledge di usion and ex end he de elopmen al
bene i s o agglome a ion well beyond immedia e locales.
The spa ial s uc u e o highe educa ion is hus shaped by he agg ega ion, in e ac-
ion, and mo emen o ins i u ions, alen , disciplines, and esea ch esou ces be ween
egions (Chao,2014;Jongbloed e al.,2008). Highe educa ion agglome a ion is no only
e lec ed by he geog aphic concen a ion o educa ional and esea ch asse s bu also by he
ex ensi e knowledge spillo e s, echnological di usion, and inno a ion ne wo k e ec s
ha subs an ially impac egional g ow h and indus ial upg ading (Rod íguez-Pose,2001;
Ve spagen,2010). NEG, oge he wi h u ban sys ems and co e–pe iphe y models, u he
p o ides heo e ical pe spec i es o analyzing bo h he hie a chical and unc ional di e -
en ia ion o u ban clus e s as well as he he e ogeneous spa ial impac s o highe educa ion
agglome a ion (Addie,2017;Vale o & Van Reenen,2019).
In China, key na ional ini ia i es such as “P ojec 211”, “P ojec 985”, and he “Dou-
ble Fi s -Class” p og am ha e s a egically concen a ed highe educa ion esou ces and
inno a ion capaci ies in coas al and co e u ban clus e s, unde lining he join impac o
ma ke o ces and policy in e en ion in spa ial es uc u ing (J. F. Gu,2012b;X. Y. Li e al.,
2020;X. Liu e al.,2024;Yu e al.,2024). These ends align wi h NEG’s asse ions ega ding
ini ial ad an age, spa ial sel - ein o cemen , and he Ma hew e ec while also empha-
sizing he long- e m ole o knowledge di usion and collec i e lea ning in na owing
egional dispa i ies.
2.3. Spa ial Mechanisms o Highe Educa ion Agglome a ion Dynamic
The concep o “agglome a ion” was i s sys ema ically a icula ed by Al ed Ma shall
in he la e nine een h cen u y, es ablishing a ounda ional pilla in egional economics. He
in oduced no only he concep o agglome a ion economies bu also he idea o “indus ial
dis ic s” (Konzelmann e al.,2025;Ma shall,2013). Building on his ounda ional wo k,
economis s u he de eloped he heo y and highligh ed i s implica ions o spa ial dy-
namics, speci ically h ough spillo e and demons a ion e ec s (K ugman,1991;Ma shall,
2013;Po e ,2000;Rosen eld e al.,2007). Spa ial agglome a ion is now widely ecognized
as a mul i ace ed p ocess ha acili a es in o ma ion exchange, collabo a ion, specializa ion,
and echnological spillo e s. These p ocesses enhance egional p oduc i i y and p omo e
inno a ion (Fuji a,2008). Va ia ions in agglome a ion pa e ns p oduce di e ing e ec s on
esou ce alloca ion and echnological ad ancemen (Mo oyama,2008;San oalha e al.,2018).
Empi ical esea ch has demons a ed ha specialized agglome a ion p omo es signi ican
posi i e e ec s and spa ial spillo e s on u ban echnological inno a ion (Gaspa ,2020;
Munnich & Iacono,2016). By syn hesizing exis ing esea ch and heo e ical insigh s, he
spa ial spillo e and demons a ion e ec s associa ed wi h highe educa ion agglome a ion
wi hin u ban clus e s can be iden i ied in se e al key a eas.
(1) The spa ial agglome a ion o highe educa ion ins i u ions wi hin u ban clus e s
enhances egional inno a ion h ough knowledge spillo e mechanisms. Compe i i e

Economies 2025,13, 167 5 o 24
and communica i e in e ac ions among uni e si ies, s uden s, and indus ies os e he
di usion o knowledge, he eby s eng hening egional inno a ion capabili ies (Y. Li e al.,
2021). The Bologna P ocess se es as a benchma k ini ia i e, exempli ying e o s o build
symbio ic pa ne ships be ween uni e si ies, indus ies, and esea ch ins i u es o ad ance
coo dina ed inno a ion ne wo ks (Woldegio gis e al.,2015;A bo & Bennewo h,2007).
Empi ical s udies ha e consis en ly con i med he subs an ial spa ial spillo e e ec s o
highe educa ion on economic de elopmen including inc eased en ep eneu ial ac i i ies
and ele a ed inno a ion le els (Caniëls & an den Bosch,2011;D ucke ,2016). Howe e ,
while esea ch ex ensi ely explo es echnical suppo p o ided by uni e si ies o egional
de elopmen , ewe s udies examine hei c i ical ole in shaping egional inno a ion
sys ems h ough highe educa ion agglome a ion.
(2) The spa ial agglome a ion o highe educa ion ampli ies “human capi al ex e -
nali ies”, ans o ming accumula ed human capi al in o measu able alen di idends ia
economies o scale. Concen a ing highe educa ion esou ces, including uni e si ies, s u-
den s, and acul y, wi hin u ban clus e s gene a es a obus pool o skilled alen , acili a ing
he con e sion o esea ch ou pu s in o p oduc i e o ces and ein o cing egional inno-
a ion capabili ies (Vale o & Van Reenen,2019). S a egic in es men in human capi al
h ough spa ial agglome a ion p inciples can elimina e ins i u ional ba ie s o alen mo-
bili y, enabling he e icien low o highly skilled labo ac oss egions and maximizing
socie al bene i s. Empi ical e idence has demons a ed ha he spa ial concen a ion o
high-skilled labo inc eases he a e o e u n on educa ion and c ea es a demons a ion
e ec , aising wages and boos ing a ci y’s o e all social a e o e u n (Czalle ,2017;Xiong
e al.,2022).
(3) Spa ial agglome a ion in highe educa ion acili a es esou ce alloca ion and in e-
g a ion, o e coming ine iciencies caused by agmen ed ins i u ional s uc u es. U ban
clus e s in de eloped egions, such as he No heas Coas o he Uni ed S a es, exempli y
his e iciency, whe e sho commu ing dis ances among uni e si ies enhance academic
collabo a ion and p o ide ample s uden and acul y esou ces o ins i u ional de elop-
men (Bayuo e al.,2020;Hu man & Quigley,2002). A oiding he edundancies associa ed
wi h “colossal and comp ehensi e” models, spa ial agglome a ion enables egional highe
educa ion sys ems o op imize esou ce alloca ion. In egions like he Yang ze Ri e Del a,
Pea l Ri e Del a, and Shandong Peninsula, he spa ial e olu ion o highe educa ion has
ca alyzed en e p ise agglome a ion, capi al in es men , alen cul i a ion, and echnological
enewal (Wilck,2020;T. Zhang e al.,2023). These enhancemen s in esou ce e iciency and
in eg a ion accele a e human capi al accumula ion, scien i ic p og ess, and socioeconomic
de elopmen , making u ban clus e s pi o al in d i ing coo dina ed egional g ow h.
The global pu sui o high-quali y de elopmen places inc easing emphasis on egional
inno a ion and social inclusi eness. Highe educa ion agglome a ion, as concep ualized in
agglome a ion economics and knowledge ex e nali ies, is ecognized as a key mechanism
o p omo ing egional inno a ion, alen concen a ion, and s uc u al ans o ma ion.
D awing on agglome a ion economies, he spa ial concen a ion o highe educa ion acil-
i a es in o ma ion exchange, labo ma ke ma ching, and knowledge di usion, c ea ing
a o able condi ions o inno a ion and indus ial upg ading. Knowledge ex e nali y
u he highligh s he ole o uni e si ies in acili a ing c oss-sec o al knowledge spillo e s
and demons a ion e ec s ha bene i su ounding a eas while new economic geog aphy
emphasizes he ampli ying in luence o scale and spa ial p oximi y. Acco dingly, he pape
iden i ies highe educa ion agglome a ion as he co e explana o y a iable and in oduces a
comp ehensi e heo e ical amewo k ( e e o Figu e 1) ha elucida es he in insic mecha-
nisms o spa ial and demons a ion e ec s. Based on his heo e ical logic, he ollowing
hypo heses a e p oposed:
Economies 2025,13, 167 6 o 24
H1: Highe educa ion agglome a ion in u ban clus e s gene a es spa ial spillo e s ha d i e egional
socioeconomic de elopmen h ough inno a ion, alen a ac ion, and economic ans o ma ion.
H2: The spa ial concen a ion o highe educa ion, as explained by new economic geog aphy and
agglome a ion economies, enhances egional g ow h ia knowledge spillo e s, inno a ion, and
coo dina ed u baniza ion.
H3: The impac s o highe educa ion agglome a ion exhibi empo al and spa ial he e ogenei y
ac oss u ban clus e s, wi h he e ec s di e ing by egion and e ol ing o e ime.
Figu e 1. The spa ial mechanism o u ban clus e in highe educa ion agglome a ion.
3. Empi ical S a egy
3.1. Backg ound
China is unde going one o he wo ld’s mos apid u baniza ion p ocesses, cha ac-
e ized by signi ican spa ial expansion and econ igu a ion. The collabo a i e sha ing,
syne gis ic coope a ion, and adia ing in luence among ci ies ha e in ensi ied, ma king
a shi om unidi ec ional u ban de elopmen o a spa ially coo dina ed model d i en
by u ban clus e s, which exempli ies a amewo k o achie ing common p ospe i y (Yin
e al.,2022). In alignmen wi h China’s “14 h Fi e-Yea Planning”, he “Two T ans e se
and Th ee Leng hways” u ban clus e amewo k is poised s a egically o accele a e he
eme gence o i al o ces o high-quali y de elopmen . This ini ia i e aims h ough a
bel - ie ed app oach, which includes 193 ci ies and 19 clus e s, he eby enhancing egional
economic e iciency and ensu ing ha he bene i s o de elopmen a e mo e equi ably
sha ed among he populace (Fang & Yu,2020). The spa ial e olu ion o u ban clus e s
dic a es he edis ibu ion o esou ces, wi h he coo dina ed ad ancemen o hese clus e s
expec ed o unlock he demog aphic di idend h ough imp o ed labo alloca ion, he eby
d i ing egional inno a ion and acili a ing China’s ansi ion o high-quali y economic
g ow h (C. Zhang e al.,2024). To sus ain his ajec o y, i is essen ial o mo e beyond
adi ional adminis a i e esou ce alloca ion and o econ igu e highe educa ion esou ces
in line wi h he spa ial e olu ion o u ban clus e s. This equi es explo ing inno a i e
ins i u ional a angemen s and op imizing egional highe educa ion layou s o ca alyze
socioeconomic p og ess and suppo sus ainable de elopmen .
3.2. Me hod
In conco dance wi h Joseph Schumpe e ’s concep ualiza ion o inno a ion and i s
implica ions o economic g ow h, Paul Rome ’s endogenous g ow h heo y iden i ies
echnological p og ess as an in insic d i e o sus ained economic expansion. This heo y
Economies 2025,13, 167 7 o 24
accen ua es he escala ing bene i s o knowledge agg ega ion, echnological inno a ion, and
human capi al in es men . The op imal condui o knowledge accumula ion is h ough
s a egic in es men in human capi al, sugges ing a go e nmen al impe a i e o bols e he
spa ial con igu a ion o highe educa ion. Such in es men s a e ins umen al in amassing a
knowledge base and p opelling egional echnological ad ancemen s h ough enhanced
scien i ic and echnological endea o s (Chand a,2022;Ziemnowicz,2020). The p esence
o knowledge spillo e and demons a ion e ec s indica es ha u ban clus e s, wi h hei
concen a ed highe educa ion ins i u ions, can os e condi ions o mul iple equilib ium
in knowledge and inno a ion dynamics. As he p eeminen o m o spa ial layou , he
ascendancy o u ban clus e has inc easingly su passed he adi ional economic s uc u es
wi hin p o inces/s a es, becoming synonymous wi h egional economies. Based on his
ac , ou esea ch manages o examine he impac o highe educa ion in u ban clus e s,
seeking o examine whe he hese a angemen s in luence inno a ion, human capi al, and
socioeconomic de elopmen . In ligh o his, ou empi ical s a egy is ancho ed in a e ined
knowledge p oduc ion unc ion, which se es as he ounda ion o a baseline eg ession
model. I is subjec ed o a igo ous empi ical examina ion o asce ain i s alidi y and
applicabili y. The benchma k model’s o mula ion is a icula ed as ollows:
Ou pu i =α0+β1HEAi +L·β2L·HEAi +β3Xi +µi+γ +εi (1)
Equa ion (1) is he dynamic panel model wi h wo-way in e ac ion ixed e ec , whe e
i and e e o he ci y and he yea , espec i ely. The dependen a iable, Ou pu , encap-
sula es he spec um o inno a ion capaci y, human capi al, indus ial composi ion, and
pe capi a disposable income o ci ies ac oss a ious yea s. The co e explana o y a iable,
HEA, cap u es he spa ial agglome a ion o highe educa ion con igu a ion. The con ol
a iables, X, encompass a sui e o ac o s ha a ec a ci y’s inno a ion capabili ies, indus-
ial s uc u e, esiden ial income le els, and human capi al accumula ion.
α
is he cons an
e m while
β
ep esen s he ec o s o coe icien s o he explana o y and con ol a iables.
The ci y ixed e ec ,
µ
, mi iga es he in luence o unobse ed, ci y-speci ic a iables ha a e
in a ian o e ime bu di e among ci ies. The ime ixed e ec ,
γ
, add esses unobse ed
elemen s ied o he empo al accumula ion o e ec s and empo al a ia ions, e lec ing
he mac oeconomic shi s and inc emen al changes aced by highe educa ion de elopmen
in u ban clus e s. The andom dis u bance e m,
ε
, accoun s o he s ochas ic elemen s
wi hin he model.
The a ionale behind he design o his model lies in i s conside a ion o he cumula-
i e e ec s o clus e ing o e ime and he egional selec ion mechanisms, a oiding he
po en ial endogenei y dis u bances. By inco po a ing lagged e ms o independen a i-
ables and employing a wo-way-in e ac ion ixed-e ec s app oach, he model in eg a es
mul idimensional shocks p esen in eali y and cap u es he empo al he e ogenei y o
how a ious ci ies espond o hese shocks. This me hodology minimizes selec ion bias
a ising om pa h dependence and mi iga es he in e e ence o endogenei y issues as
much as possible.
To add ess po en ial endogenei y, e e se causali y, and sample selec ion bias in he
model, his pape adop s mul iple obus app oaches. Fi s , p opensi y sco e ma ching is
applied o educe sample selec ion bias. Speci ically, he ke nel ma ching me hod is u ilized
o pai he ea men and con ol g oups based on co a ia es om he baseline eg ession.
This app oach educes he in luence o ex eme p opensi y sco es and imp o es he alidi y
o sample ma ching. Second, o add ess possible e e se causali y, an ins umen al a iable
app oach is u ilized, wi h he s ock o highe educa ion ins i u ions in each ci y selec ed
as an ins umen . This selec ion is jus i ied because he numbe o ins i u ions is s ongly
co ela ed wi h highe educa ion de elopmen bu is no di ec ly a ec ed by he ou come
Economies 2025,13, 167 8 o 24
a iables, hus ensu ing i s exogenei y. The IV es ima ion p o ides u he con ol o e
endogenei y conce ns. Thi d, obus ness checks a e conduc ed by subs i u ing he p ima y
independen a iable wi h an al e na i e indica o , speci ically he numbe o u ban esi-
den s holding an associa e deg ee o highe . This a iable also e lec s highe educa ion
agglome a ion a he ci y le el. Collec i ely, hese me hods imp o e he c edibili y and
explana o y powe o he empi ical esul s.
To u he pinpoin he spa ial and demons a ion e ec s o highe educa ion agglom-
e a ion, his pape employs he Global Mo an’s I o es i y spa ial au oco ela ion ac oss
ci ies and applies he spa ial panel Du bin model o econome ic es ima ion, ollowing he
analy ical pa adigm o spa ial econome ic me hodologies. Spa ial panel models gene ally
include key o ms such as he spa ial Du bin model (SDM), he Spa ial Au o eg essi e
Model (SAR), and he Spa ial Au oco ela ion Model (SAC). The SAR and SAC models
impose ela i ely s ic assump ions whe eas he SDM is mo e lexible. I accoun s o
spa ial co ela ions among he dependen a iable, independen a iables, and e o e m
while also p o iding in e ac ion e ec s ac oss a ious spa ial weigh ma ices (Anselin &
Anil,1998). This e sa ili y makes he SDM well-sui ed o p ac ical scena ios and a a o ed
op ion in empi ical esea ch. The models a e as ollows:
Ou pu i =α1+ρ1WijOu pu ij +β1WijHEAij +β2WijXij +β3HEAi +β4Xi +µi+γ +εi (2)
GlobalMo an′s I =
∑n
i=1∑n
j=1WijYi−YYj−Y
S2∑n
i=1∑n
j=0Wij
(3)
Equa ion (2) is he spa ial econome ic eg ession model, whe e
ρ
and
β
e e o he
spa ial in e ac ion coe icien s o he dependen and co e explana o y a iables, espec-
i ely. I
ρ
is ze o o insigni ican , i indica es ha he e is no spa ial associa ion. I his
alue is posi i e signi ican ly, i sugges s ha he spa ial and demons a ion e ec o he ag-
glome a ion o highe educa ion among ci ies has a signi ican posi i e impac . Con e sely,
a nega i e alue would imply he opposi e. In Equa ion (3), Y
i
and Y
j
ep esen he ob-
se ed alues o ci ies i and j conce ning knowledge p oduc ion, indus ial ans o ma ion,
disposable income, and human capi al. The spa ial weigh ma ix, W
ij
, illus a es he spa ial
ela ionship be ween ci ies i and j. Gi en ha in e -ci y economic and social ac i i ies a e
in luenced no only by geog aphical p oximi y bu also by economic de elopmen , his
pape u ilizes spa ial weigh ma ices ha accoun o bo h geog aphical and economic
dis ances; hese wo dis ances a e c ucial o spa ial agglome a ion and spillo e e ec s
(Maggioni & Ube i,2009). The Global Mo an’s I is employed o de ec pa e ns o spa ial
clus e ing and dispe sion among ci ies. The emaining ec o s a e consis en wi h hose
p esen ed in Equa ion (1).
3.3. Va iables
3.3.1. Dependen Va iables
This pape selec s he numbe o pa en g an s, he le el o human capi al, he p o-
po ion o seconda y and e ia y indus ies, and egional a e age wages as indica o s
o assess he impac o he agglome a ion o highe educa ion on inno a ion capabili ies,
human capi al, indus ial s uc u e, and disposable income. The a ionale o he choice o
he dependen a iable is mul i ace ed:
(1) The le el o scien i ic and echnological inno a ion is o en gauged by o iginal
achie emen s o science and echnology in es men s. Howe e , elying solely on he
numbe o o iginal achie emen s o esea ch expendi u e can o e look he syne gy be ween
inno a ion and egional economic and social de elopmen (Capone e al.,2021). Pa en
g an s se e as an ou come indica o o inno a ion le els, e lec ing bo h he ans o ma ion
Economies 2025,13, 167 15 o 24
The second me hod employs he ins umen al a iable app oach o add ess po en ial
e e se causali y be ween he a iables. The pape u ilizes he s ock o highe educa ion
ins i u ions (‘School’) ac oss a ious ci ies as an objec i e ins umen al a iable. This
selec ion is unde pinned by se e al conside a ions. Fi s ly, he e is ypically a s ong
co ela ion be ween he numbe o highe educa ion ins i u ions and he ex en o highe
educa ion de elopmen . A g ea e numbe o ins i u ions gene ally signi ies an abundance
o deg ee esou ces and educa ional oppo uni ies, indica ing a highe agglome a ion o
highe educa ion. This co ela ion ul ills he equi emen o a s ong associa ion be ween
he ins umen al a iable and he independen a iable. Secondly, inno a ion capabili y is
con ingen upon ac o s such as esea ch in es men and collabo a ion, wi h he numbe o
ins i u ions se ing as a ounda ional condi ion. Human capi al le els a e limi ed by a ious
ac o s, including educa ional quali y; he indus ial s uc u e is in luenced by esou ces
and policy; and esiden disposable income is ied o he economy and employmen , none
o which a e di ec ly de e mined by he numbe o ins i u ions. Hence, he e is no di ec
causal ela ionship wi h he e o e m, and i main ains a deg ee o exogenei y ela i e
o he dependen a iable. The esul s show ha ( e e o Table 6) he Kleibe gen–Paap
k LM s a is ic is signi ican a he 1% le el, allowing he ejec ion o he null hypo hesis
o he unde -iden i ica ion o he ins umen al a iable. The C agg–Donald Wald F and
Kleibe gen–Paap k Wald F s a is ics signi ican ly exceed he c i ical alue o 16.38, ejec ing
he hypo hesis o weak ins umen s. Addi ionally, ou es ima ion esul s align closely wi h
he benchma k eg ession, con i ming he alidi y o ou ins umen al a iable selec ion
and a es ing o he obus ness and eliabili y o he pape ’s indings.
Table 6. Resul s o ins umen al a iable es .
Fi s -S age Second-S age
Va iables
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
School PAT HC IND SAL
STU 0.234 ***
(0.046)
0.517 **
(0.243)
2.621 ***
(0.560)
0.099 **
(0.078)
0.038
(0.085)
Kleibe gen–Paap k LM = 17.110 ***
C agg–Donald Wald F = 90.292, Kleibe gen–Paap k Wald F = 26.238
UNI 0.015 ***
(0.001)
7.890 *
(4.091)
39.977 ***
(4.494)
1.503 **
(1.134)
0576
(1.275)
Kleibe gen–Paap k LM = 25.852 ***
C agg–Donald Wald F = 1663.86, Kleibe gen–Paap k Wald F = 109.46
Con ols YES YES YES YES YES YES YES YES YES YES
TWFE YES YES YES YES YES YES YES YES YES YES
No e: *** p< 0.01, ** p< 0.05, * p< 0.1; obus ness s anda d e o s a e in pa en heses.
The hi d me hod in ol es he subs i u ion o he co e independen a iable (as
de ailed in Table 7). In his pa , he a iable is eplaced wi h he numbe o u ban esiden s
wi h junio college educa ion o abo e (STU2), an indica o ha simila ly ep esen s he
spa ial agglome a ion o highe educa ion among ci ies and is concep ually and s a is ically
simila o he co e explana o y a iable used in he benchma k model. The esul s a e
consis en wi h he eg ession esul s abo e, wi h he di ec ion and signi icance le el o he
es ima ed coe icien s o each a iable in each column being simila and wi hou signi ican
di e ences, u he con i ming he obus ness o he empi ical es ima ion esul s.

Economies 2025,13, 167 16 o 24
Table 7. Robus ness es esul s wi h co e explana o y a iables eplaced.
Va iables
Spa ial Weigh Ma ix in Geog aphical Dis ance Spa ial Weigh Ma ix in Economic Dis ance
(1)
PAT
(2)
HC
(3)
IND
(4)
SAL
(5)
PAT
(6)
HC
(7)
IND
(8)
SAL
ρ0.877 ***
(0.002)
0.942 *
(0.025)
0.900 ***
(0.023)
0.710 ***
(0.049)
0.293 ***
(0.029)
0.700 **
(0.033)
0.571 ***
(0.020)
0.436 ***
(0.025)
W×STU2 0.027 **
(0.0.014)
0.079 ***
(0.010)
0.007 *
(0.010)
0.006 *
(0.042)
0.013 ***
(0.005)
0.002 ***
(0.003)
0.001 *
(0.001)
0.001 *
(0.001)
Con ols YES YES YES YES YES YES YES YES
W×Con ols YES YES YES YES YES YES YES YES
TWFE YES YES YES YES YES YES YES YES
No e: *** p< 0.01, ** p< 0.05, * p< 0.1; obus ness s anda d e o s a e in pa en heses.
4.3. He e ogenei y Analysis
Conside ing he he e ogenei y o spa ial and demons a ion e ec s a di e en s ages
o u ban clus e s, we conduc a sample-speci ic he e ogenei y analysis based on he “Two
T ans e se and Th ee Leng hways” Re o m ou lined in China’s “14 h Fi e-Yea Planning”
o he 19 na ional u ban clus e s di ided in o h ee g adien s (see No e 3), as shown in
Table 8.
Table 8. Spa ial econome ic es ima ion esul s o he e ogenei y analysis (by egion).3
Va iables
“High-Quali y Upg ading”
U ban Clus e s
“Vigo ous G ow h and Expansion”
U ban Clus e s
“Nu u ing and De eloping”
U ban Clus e s
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)
PAT HC IND SAL PAT HC IND SAL PAT HC IND SAL
ρ(STU)
0.306 ***
(0.109)
2.141 ***
(0.439)
0.282 ***
(0.102)
0.267 *
(0.136)
0.058
(0.156)
0.544 **
(0.268)
0.271 **
(0.110)
0.347
(0.220)
−
0.058
(0.167)
−
0.456 *
(0.265)
−
0.099
(0.178)
0.284 **
(0.133)
W * STU −
0.124
(0.585)
0.778 *
(0.398)
0.200 **
(0.146)
−
0.086
(0.125)
−
0.270
(0.279)
0.216 *
(0.176)
0.032
(0.082)
0.095
(0.108)
1.614 **
(0.777)
1.204 **
(0.548)
0.325 *
(0.253)
0.157 *
(0.268)
Con ols YES YES YES YES YES YES YES YES YES YES YES YES
W * Con ols YES YES YES YES YES YES YES YES YES YES YES YES
TWFE YES YES YES YES YES YES YES YES YES YES YES YES
No e: *** p< 0.01, ** p< 0.05, * p< 0.1; obus ness s anda d e o s a e in pa en heses.
In e ms o he dependen a iable, he demons a ion e ec o in e -ci y highe educa-
ion spa ial agglome a ion is pa icula ly p ominen wi hin he “High-Quali y Upg ading”
and “Nu u ing and De eloping” u ban clus e s, signi ican ly posi i ely spilling o e o
he inno a ion capabili ies, human capi al le els, indus ial s uc u e ans o ma ion and
upg ading, and ini ial dis ibu ion income le els o adjacen ci ies wi hin he clus e s. How-
e e , in he “Vigo ous G ow h and Expansion” u ban clus e s, he spa ial agglome a ion
o highe educa ion shows a ying deg ees o nega i e spillo e e ec s. These indings
pa ially con i m Resea ch Hypo hesis 3. This ale s policymake s ha in a eas wi h a
weak ounda ion o highe educa ion, a “sp inkling peppe ” app oach o highe educa ion
con igu a ion is no ad isable. Ins ead, e o s should be concen a ed on signi ican ini ia-
i es o ully le e age he spa ial-demons a ion e ec s and di usion unc ions o highe
educa ion agglome a ion, he eby p omo ing coo dina ed egional economic and social
de elopmen . Rega ding he co e independen a iable, he pape unco e s ha he es i-
ma ed coe icien s o he deg ee o highe educa ion spa ial agglome a ion di e ma kedly
among u ban clus e s. Speci ically, in he “Nu u ing and De eloping” u ban clus e s, an
inc ease in he deg ee o highe educa ion spa ial agglome a ion has a signi ican ly posi i e
demons a ion e ec , which in ensi ies wi h geog aphical p oximi y. The o de o impac is
as ollows: he numbe o pa en au ho iza ions > he le el o human capi al > he upg ade
o indus ial s uc u e > he egional a e age wage le el. Addi ionally, he “High-Quali y
Upg ading” and “Vigo ous G ow h and Expansion” u ban clus e s also indica e a ying
Economies 2025,13, 167 17 o 24
deg ees o spa ial spillo e . The esul s e eal ha unde he in luence o spa ial ac o s, he
posi i e demons a ion e ec s o echnological inno a ion and socioeconomic ac i i ies a e
p edominan ly p opelled by he “High-Quali y Upg ading” u ban clus e s. Howe e , he e
is an oppo uni y: by ma u ing he spa ial scale and concen a ion o highe educa ion
agglome a ion wi hin he “Nu u ing and De eloping” u ban clus e s, he demons a ion
e ec can be signi ican ly ampli ied, he eby enhancing he o e all de elopmen al ajec o y
o hese egions.
The pape u he examines he dynamics o highe educa ion agglome a ion in spa ial
spillo e e ec s and he e olu ion o spa ial concen a ion o e ime. The he e ogenei y
analysis based on empo al changes in Table 9 e eals a posi i e co ela ion in he spa-
io empo al lag e ms o key independen a iables, indica ing ha as ime p og esses,
signi ican and sus ained spa ial spillo e and demons a ion e ec s eme ge in highe
educa ion agglome a ion. Gene ally, he spa ial e ec s do no immedia ely mani es in
economic ou comes bu equi e ime o become appa en , o e ing signi ican e idence in
suppo o Resea ch Hypo hesis 3. In he ea ly phases o u ban clus e ing, di usion and
adia ion e ec s a e no p ominen , likely due o policy suppo and egional e o s, along
wi h syne gies and s uc u al adjus men s wi hin ci ies ha ha e os e ed a obus educa-
ional ecosys em (Ma e al.,2023), acili a ing he egional expansion o highe educa ion
agglome a ion. This p ocess con ibu es o he con e gence o socioeconomic de elop-
men gaps be ween ci ies a speci ic geog aphical and economic dis ances. Howe e , as
changes o he agglome a ion o e yea s, socioeconomic ac i i ies be ween geog aphically
p oxima e and simila ly de eloped ci ies become inc easingly ib an . This is speci ically
e iden in u ban expansion, inc eased popula ion mobili y, and adjus men s in uni e si ies’
disciplina y and p o essional s uc u es (Duan e al.,2022), ad ancing economic and social
p og ess ac oss a b oade egion. This de elopmen p o ides highe educa ion ins i u ions
wi h g ea e space and esou ce suppo , encou aging ci y clus e s wi h simila indus ial
and socioeconomic de elopmen le els o o e come hei o iginal pa h dependencies,
he eby in luencing he pa e n o highe educa ion agglome a ion. Mo e impo an ly, he
dissemina ion o knowledge and echnology u he ele a es egional scien i ic inno a ion
le els and d i es economic ans o ma ion and upg ading, enhancing he spa ial spillo e
and demons a ion e ec s o his agglome a i e layou .
Table 9. Spa ial econome ic es ima ion esul s o he e ogenei y analysis (by ime).
Va iables
Spa ial Weigh Ma ix in Geog aphical Dis ance Spa ial Weigh Ma ix in Economic Dis ance
(1)
PAT
(2)
HC
(3)
IND
(4)
SAL
(5)
PAT
(6)
HC
(7)
IND
(8)
SAL
ρ(STU) 0.212 ***
(0.057)
−0.042
(0.076)
0.257 ***
(0.054)
0.236 ***
(0.055)
0.304 ***
(0.064)
0.023
(0.141)
0.796 ***
(0.025)
0.377 ***
(0.139)
W * LAG1 −0.439 ***
(0.031)
−0.157
(0.169)
−0.303 ***
(0.023)
−0.148 ***
(0.037)
−0.427 ***
(0.028)
−0.656 ***
(0.118)
−0.363 ***
(0.026)
−0.225 ***
(0.045)
W * LAG2 −0.061 **
(0.024)
0.572 ***
(0.152)
−0.002
(0.019)
−0.017
(0.014)
−0.091 ***
(0.027)
0.080
(0.150)
0.016
(0.027)
−0.052 ***
(0.014)
W * LAG3 0.156 ***
(0.024)
−0.143
(0.169)
0.081 ***
(0.025)
0.013
(0.010)
0.084 ***
(0.021)
0.012
(0.109)
0.083 ***
(0.027)
−0.007
(0.015)
W * LAG4 0.093 ***
(0.015)
0.004
(0.060)
−0.008
(0.017)
0.052 ***
(0.009)
0.082 ***
(0.015)
−0.199 ***
(0.067)
−0.007
(0.018)
0.039 ***
(0.010)
W * LAG5 0.060 ***
(0.016)
0.081
(0.110)
0.101 ***
(0.034)
0.092 ***
(0.015)
0.078 ***
(0.019)
−0.081
(0.071)
0.074 *
(0.039)
0.062 ***
(0.016)
Con ols YES YES YES YES YES YES YES YES
W * Con ols YES YES YES YES YES YES YES YES
TWFE YES YES YES YES YES YES YES YES
No e: *** p< 0.01, ** p< 0.05, * p< 0.1; obus ness s anda d e o s a e in pa en heses.
Economies 2025,13, 167 18 o 24
5. Discussion and Conclusions
5.1. Discussion
This pape has comp ehensi ely examined he mul idimensional spa ial spillo e
and demons a ion e ec s o highe educa ion agglome a ion on egional inno a ion,
human capi al accumula ion, indus ial upg ading, and income dis ibu ion wi hin Chi-
nese u ban clus e s. The empi ical indings ha e e ealed ha highe educa ion agglom-
e a ion exe s a c i ical in luence on inno a ion-led and coo dina ed egional de elop-
men , wi h no able spa ial he e ogenei y. These esul s a e la gely consonan wi h p io
empi ical esea ch while he p esen pape ad ances bo h heo e ical amewo ks and
me hodological app oaches.
The esul s a e consis en wi h mains eam s udies, which iden i ied he signi ican
con ibu ion o spa ial agglome a ion o economic g ow h (Fang & Yu,2017;Zhao e al.,
2025). These s udies ha e a i med ha spa ial concen a ion, whe he o highe educa ion
ins i u ions, popula ion, o indus y, enhances mac oeconomic pe o mance ia media ing
mechanisms such as echnological inno a ion and indus ial es uc u ing. Mo eo e , his
pape has elucida ed he cen ali y o knowledge spillo e s induced by highe educa ion
agglome a ion in augmen ing egional inno a ion capaci y, a ac ing high-quali y human
capi al, and s eng hening economic coo dina ion. This obse a ion is co obo a ed by
Liang e al. (2025), who documen ed ha polycen ic u ban spa ial s uc u es and dual-co e
coupling mechanisms signi ican ly p omo e economic esilience h ough posi i e spa ial
ex e nali ies. Unlike analyses ha ocus solely on indus ial o demog aphic ac o s, he
cu en esea ch has emphasized he composi e e ec o highe educa ion agglome a ion as
a s a egic spa ial esou ce in enhancing bo h egional esilience and inno a i e capaci y.
No ably, his pape has u he inno a ed by in eg a ing dynamic panel eg ession and
spa ial econome ic models o cap u e he empo al e olu ion and spa ial he e ogenei y o
spillo e e ec s, allowing o he iden i ica ion o pa e ns p e iously o e looked in c oss-
sec ional analyses. Addi ionally, in line wi h Yu e al. (2024), his pape has ecognized he
go e nmen ’s pi o al ole and egional he e ogenei y in shaping he s a i ied unc ions
and di e en ia ed de elopmen pa hs o Chinese u ban agglome a ions. Dis inc i ely, his
esea ch has de eloped a comp ehensi e amewo k ha explici ly links highe educa ion
esou ce spa ial alloca ion o he mul ilaye ed g ow h ajec o ies o u ban clus e s, includ-
ing he iden i ica ion o unc ional egions and inno a ion hubs, hus en iching he cu en
unde s anding o u ban agglome a ion ypologies in de eloping economies.
In e na ionally, a compa able posi i e impac o highe educa ion spa ial concen a-
ion on egional inno a ion sys ems and echnological di usion has been subs an ia ed
in de eloped economies, no ably hose in Eu ope and No h Ame ica (D ucke ,2016;
Ebe sbe ge e al.,2013;Vale o & Van Reenen,2019). None heless, Chinese u ban agglom-
e a ions a e dis inc o hei policy-d i en dynamics, apid indus ial ans o ma ion, and
p onounced egional di e si y (H. Yang e al.,2025). This pape ’s nuanced analysis o
in eg a ion be ween s a e-d i en and ma ke -d i en mechanisms, and he obse ed shi o-
wa d di e si ied inno a ion models, ep esen s an o iginal con ibu ion o he compa a i e
li e a u e on global agglome a ion economies. F om a heo e ical iewpoin , he in eg a ion
o Ma in’s egional esilience amewo k and mul i-scala spa ial he e ogenei y en iches
he heo e ical landscape o new economic geog aphy and agglome a ion economics (Feng,
2021;Ma in,2011), o e ing c i ical insigh s in o he unique spillo e pa hways and in-
dus ial di e en ia ion associa ed wi h highe educa ion agglome a ion. In pa icula ,
by employing a spa ially explici analy ical pe spec i e, his esea ch has unpacked he
mul iple channels, such as knowledge ne wo ks and ins i u ional syne gies, h ough which
highe educa ion agglome a ion shapes complex u ban- egional inno a ion ecosys ems.
Economies 2025,13, 167 19 o 24
5.2. Conclusions
G ounded in he e ol ing phenomenon and s uc u e o u ban clus e s, his pape has
sough o elucida e he mechanisms by which highe educa ion agglome a ion s imula es
egional socioeconomic p og ess. U ilizing panel da a om 193 ci ies ac oss he “Two
T ans e se and Th ee Leng hways” u ban clus e s om 2006 o 2020, we applied a dynamic
panel eg ession o sc u inize he in luence o highe educa ion agglome a ion on in e -ci y
inno a ion capaci y, human capi al de elopmen , indus ial es uc u ing, and equi able
ini ial income dis ibu ion. Fu he mo e, spa ial Du bin models, in o med by spa ial
weigh s de i ed om geog aphic and economic dis ances, we e employed o con i m he
demons a ion e ec s o highe educa ion agglome a ion and i s ca aly ic pa hways o
egional de elopmen . Ou o e a ching conclusion is ha he cu en agglome a ion o
highe educa ion is e ec i ely synch onized wi h he s a egic u ban clus e de elopmen
a he na ional le el. This alignmen gene a es a no able agglome a ion e ec on inno a ion
capabili ies, human capi al accumula ion, indus ial adjus men , and income le els. The
key ole o co e ci ies wi hin u ban clus e s is pa icula ly p onounced, wi h he spa ial and
demons a ion e ec o highe educa ion agglome a ion becoming inc easingly e iden ,
especially in e ms o human capi al accumula ion.
This esea ch sys ema ically analyzes he spa ial e olu ion o u ban clus e s and high-
ligh s he ole o highe educa ion agglome a ion in egional de elopmen . The indings
indica e ha egional highe educa ion s a egies should be mo e closely in eg a ed wi h
he spa ial and socioeconomic amewo ks o u ban clus e s. Policy o mula ion should be
adap ed o he speci ic condi ions o di e en egions o add ess di e se de elopmen al
needs. Rele an concep s om new economic geog aphy, such as mul iple equilib ium
and spa ial equilib ium, emphasize he impo ance o spa ial in e dependence. Regional
policies should acili a e he low o educa ional esou ces and popula ion in de eloped
a eas, while ad ancing a ge ed highe educa ion e o ms in less-de eloped egions, and
accoun o di e ences in echnological and esou ce alloca ion among di e en ci ies. The
heo e ical con ibu ion o his pape lies in supplemen ing and alida ing exis ing heo ies
on highe educa ion agglome a ion and egional coo dina ion om a spa ial pe spec i e
and demons a ing he applicabili y o new economic geog aphy app oaches in he Chi-
nese con ex . In p ac ical e ms, he esul s sugges ha highe educa ion policies should
be di e en ia ed based on he ype and de elopmen al s age o u ban clus e s. Fo co e
ci ies, e o s should ocus on he u he concen a ion o highe educa ion esou ces and
enhancing inno a ion capaci y; o eme ging and seconda y clus e s, he emphasis should
be placed on close in eg a ion be ween highe educa ion and local indus y; and o pe-
iphe al o less-de eloped a eas, a en ion should be di ec ed o he equi able dis ibu ion
o educa ional esou ces and imp o ing access o disad an aged g oups.
The limi a ions o his pape s em p ima ily om i s eliance on o icial s a is ical
da a, which do no ully cap u e ac o s such as he low o educa ional esou ces o
in o mal educa ion, and om he es ic ed ange o a iables and me hodologies applied.
Fu u e esea ch could inco po a e mic o-le el da a and mo e ad anced spa ial analy ical
echniques o conduc mo e de ailed compa a i e s udies ac oss a ious ypes o u ban
clus e s and popula ion g oups.
Au ho Con ibu ions: Concep ualiza ion, Y.H., C.Y. and J.M.; me hodology, Y.H.; alida ion, Y.H. and
C.Y.; o mal analysis, Y.H.; da a cu a ion, Y.H.; w i ing—o iginal d a p epa a ion, Y.H.; w i ing— e iew
and edi ing, C.Y. and J.M.; supe ision, C.Y.; p ojec adminis a ion, C.Y.; unding acquisi ion, J.M.
All au ho s ha e ead and ag eed o he published e sion o he manusc ip .
Funding: This esea ch was unded by ollowing p og ams: Pos doc o al Resea ch P ojec s o
Guangzhou (G an No. 624021-71); Fujian P o incial Social Science Fund Gene al P ojec (G an
Economies 2025,13, 167 20 o 24
No. FJ2024B138). Majo P ojec o he Minis y o Educa ion’s Resea ch on Philosophy and Social
Sciences (G an No. 21JZD057).
Ins i u ional Re iew Boa d S a emen : No applicable.
In o med Consen S a emen : No applicable.
Da a A ailabili y S a emen : The o iginal da a p esen ed in his s udy a e included in he a icle.
Fu he inqui ies can be di ec ed o he co esponding au ho .
Con lic s o In e es : The au ho s decla e no con lic o in e es .
No es
1
The da a on en olmen in highe educa ion ha e been de i ed om he “Na ional Educa ional S a is ics Bulle in” issued by he
Minis y o Educa ion o China. Rela ed links: h p://www.moe.go .cn/jyb
_
sjzl/sjzl
_
z jgb/202108/ 20210827
_
555004.h ml,
accessed on 10 Ma ch 2025.
2
The na ional-le el u ban clus e s o he “Two T ans e se and Th ee Leng hways” amewo k in China a e hie a chically classi ied
as ollows: he i s ie o “High-Quali y Upg ading” U ban Clus e s (Beijing–Tianjin–Hebei, he Yang ze Ri e Del a, he Pea l
Ri e Del a, Chengdu–Chongqing, and he Middle Reaches o he Yang ze Ri e ), he second ie o “Vigo ous G ow h and
Expansion” U ban Clus e s ( he Shandong Peninsula, he Coas al A ea o Guangdong–Fujian–Zhejiang, he Cen al Plains, he
Guanzhong Plain, and he Beibu Gul ), and he hi d ie o “Nu u ing and De eloping” U ban Clus e s (Ha bin–Changchun,
Cen al and Sou he n Liaoning, Cen al Shanxi, Cen al Guizhou, Cen al Yunnan, Hohho –Bao ou–Yulin, Lanzhou–Xining,
Ningxia along he Yellow Ri e , and he no he n slope o he Tianshan Moun ains).
3
Gi en ha he esul s o benchma k eg ession ha e alida ed he in e changeabili y o “numbe o highe educa ion ins i u ions
pe million people” as a me ic o assessing he spa ial agglome a ion o highe educa ion dis ibu ion, and ha e shown
signi ican spa ial spillo e e ec s unde bo h he geog aphical dis ance and economic dis ance spa ial weigh ma ices, in
Tables 8and 9, i is hence o h su icien o use only “numbe o unde g adua e s uden s” as a p oxy a iable, based on he spa ial
weigh ma ix o geog aphical dis ance and he dynamic panel analysis, o examine he he e ogenei y o he spa io empo al
e ec s o highe educa ion agglome a ion in u ban clus e s. The comple e esul s a e also a ailable upon eques .
Re e ences
Abu Al oul, M. N., Bazhai , A. H., Kha a beh, I. N., A ian, A. G., & Abu Al-Foul, M. N. (2024). The e ec o educa ion on economic
g ow h in Sub-Saha an A ican coun ies: Do ins i u ions ma e ? Economies,12(11), 300. [C ossRe ]
Acs, Z. J., Anselin, L., & Va ga, A. (2002). Pa en s and inno a ion coun s as measu es o egional p oduc ion o new knowledge.
Resea ch Policy,31(7), 1069–1085. [C ossRe ]
Adamakou, M., Kallio as, D., & Pe akos, G. (2021). De ec ing con e gence ends among EU uni e si ies. Economies,9(2), 42.
[C ossRe ]
Addie, J. P. D. (2017). F om he u ban uni e si y o uni e si ies in u ban socie y. Regional S udies,51(7), 1089–1099. [C ossRe ]
Addie, J. P. D., Keil, R., & Olds, K. (2015). Beyond own and gown: Uni e si ies, e i o iali y and he mobiliza ion o new u ban
s uc u es in Canada. Te i o y Poli ics Go e nance,3(1), 27–50. [C ossRe ]
Alde , S., Shao, L., & Zilibo i, F. (2016). Economic e o ms and indus ial policy in a panel o Chinese ci ies. Jou nal o Economic G ow h,
21(4), 305–349. [C ossRe ]
Aldie i, L., Ko semi , M. N., & Vinci, C. P. (2018). Knowledge spillo e e ec s: Empi ical e idence om Russian egions. Quali y &
Quan i y,52(5), 2111–2132. [C ossRe ]
Al-Youbi, A. O., Zahed, A. H. M., Nahas, M. N., & Hegazy, A. A. (2021). The Roles o Uni e si ies in De elopmen . In The leading
wo ld’s mos inno a i e uni e si ies (pp. 1–8). Sp inge In e na ional Publishing. [C ossRe ]
Anselin, L. B., & Anil, K. (1998). Spa ial dependence in linea eg ession models wi h an in oduc ion o spa ial econome ics. In
A. Ullah (Ed.), Handbook o applied economic s a is ics. CRC P ess. [C ossRe ]
A bo, P., & Bennewo h, P. (2007). Unde s anding he egional con ibu ion o highe educa ion ins i u ions (OECD Educa ion Wo king
Pape s, No. 9). OECD Publishing. [C ossRe ]
A i enang, A. F., & Chand amidi, A. N. (2023). The spa ial e ec s o iscal decen aliza ion on egional con e gence: The case o egions
in indonesia. GeoJou nal,88(2), 2011–2030. [C ossRe ]
Bayuo, B. B., Chaminade, C., & Gö ansson, B. (2020). Unpacking he ole o uni e si ies in he eme gence, de elopmen and impac o
social inno a ions—A sys ema ic e iew o he li e a u e. Technological Fo ecas ing and Social Change,155, 120030. [C ossRe ]
Belde bos, R., Cas ellani, D., Du, H. S., & Lee, G. H. (2024). In e nal e sus ex e nal agglome a ion ad an ages in in es men loca ion
choice: The ole o global ci ies’ in e na ional connec i i y. Jou nal o In e na ional Business S udies,55(6), 745–763. [C ossRe ]

Economies 2025,13, 167 21 o 24
Be ilacqua, C., Pizzimen i, P., & Ou, Y. (2023). Ci ies in ansi ion and u ban inno a ion ecosys ems: Place and inno a ion dynamics in
he case o Bos on and Camb idge (USA). Sus ainabili y,15(18), 13346. [C ossRe ]
Billings, S. B., & Johnson, E. B. (2016). Agglome a ion wi hin an u ban a ea. Jou nal o U ban Economics,91, 13–25. [C ossRe ]
Bo si, M. T., Vale io Mendoza, O. M., & Comim, F. (2022). Measu ing he p o incial supply o highe educa ion ins i u ions in China.
China Economic Re iew,71, 101724. [C ossRe ]
B ennan, J., & Coch ane, A. (2019). Uni e si ies: In, o , and beyond hei ci ies. Ox o d Re iew o Educa ion,45(2), 188–203. [C ossRe ]
B eschi, S., & Lissoni, F. (2001). Knowledge spillo e s and local inno a ion sys ems: A c i ical su ey. Indus ial and Co po a e Change,
10(4), 975–1005. [C ossRe ]
Caniëls, M. C. J., & an den Bosch, H. (2011). The ole o highe educa ion ins i u ions in building egional inno a ion sys ems. Pape s
in Regional Science,90(2), 271–287. [C ossRe ]
Capone, F., Lazze e i, L., & Innocen i, N. (2021). Inno a ion and di e si y: The ole o knowledge ne wo ks in he in en i e capaci y
o ci ies. Small Business Economics,56(2), 773–788. [C ossRe ]
Chand a, R. (2022). Paul Rome and mode n endogenous g ow h heo y. In R. Chand a (Ed.), Endogenous g ow h in his o ical pe spec i e:
F om Adam Smi h o Paul Rome (pp. 251–281). Sp inge In e na ional Publishing. [C ossRe ]
Chao, R. Y. (2014). Pa hways o an Eas Asian highe educa ion a ea: A compa a i e analysis o Eas Asian and Eu opean egionaliza ion
p ocesses. Highe Educa ion,68(4), 559–575. [C ossRe ]
Chen, K., & Kenney, M. (2007). Uni e si ies/ esea ch ins i u es and egional inno a ion sys ems: The cases o Beijing and Shenzhen.
Wo ld De elopmen ,35(6), 1056–1074. [C ossRe ]
Chou, M. H., E kkilä, T., & Mölsä, J. (2023). C a ing inno a ion hubs: Fu u e ci ies and global challenges. The B i ish Jou nal o Poli ics
and In e na ional Rela ions,26(3), 694–717. [C ossRe ]
Czalle , L. (2017). Inc easing social e u ns o human capi al: E idence om Hunga ian egions. Regional S udies,51(3), 467–477.
[C ossRe ]
De Falco, S. (2021). Spa ial dynamics ega ding geog aphical concen a ion o economic ac i i ies. I alian NUTS-3 analysis. Applied
Spa ial Analysis and Policy,14(4), 795–825. [C ossRe ]
D ucke , J. (2016). Reconside ing he egional economic de elopmen impac s o highe educa ion ins i u ions in he Uni ed S a es.
Regional S udies,50(7), 1185–1202. [C ossRe ]
Duan, Y., Zhang, H., Wang, W., & Ao, X. (2022). The e ec s o China’s highe educa ion expansion on u ban and u al in e gene a ional
mobili y. China Economic Re iew,73, 101793. [C ossRe ]
Ebe sbe ge , B., He s ad, S. J., & Al mann, A. (2013). Highe educa ion ins i u ions and egional de elopmen . In A. Al mann, & B.
Ebe sbe ge (Eds.), Uni e si ies in change: Managing highe educa ion ins i u ions in he age o globaliza ion (pp. 311–321). Sp inge
New Yo k. [C ossRe ]
Fang, C., & Yu, D. (2017). U ban agglome a ion: An e ol ing concep o an eme ging phenomenon. Landscape and U ban Planning,162,
126–136. [C ossRe ]
Fang, C., & Yu, D. (2020). O e iew o he undamen al conno a ion and s a egic posi ion o China’s u ban agglome a ions. In China’s
u ban agglome a ions (pp. 1–63). Sp inge Singapo e. [C ossRe ]
Feng, L. (2021). In luence o spa io empo al di e ence and inpu o highe educa ion on egional economy. Disc e e Dynamics in Na u e
and Socie y,2021(1), 9515285. [C ossRe ]
Flo ida, R., & Adle , P. (2020). Economic geog aphy. In A. Kobayashi (Ed.), In e na ional encyclopedia o human geog aphy (2nd ed.,
pp. 25–28). Else ie . [C ossRe ]
Fuji a, M. (2008). Fo ma ion and g ow h o economic agglome a ions and indus ial clus e s: A heo e ical amewo k om he
iewpoin o spa ial economics. In A. Kuchiki, & M. Tsuji (Eds.), The lowcha app oach o indus ial clus e policy (pp. 18–37).
Palg a e Macmillan UK. [C ossRe ]
Fuji a, M., & K ugman, P. (2003). The new economic geog aphy: Pas , p esen and he u u e. Pape s in Regional Science,83(1), 139–164.
[C ossRe ]
Fuji a, M., & Mo i, T. (1996). The ole o po s in he making o majo ci ies: Sel -agglome a ion and hub-e ec . Jou nal o De elopmen
Economics,49(1), 93–120. [C ossRe ]
Gaspa , J. M. (2020). New economic geog aphy: Economic in eg a ion and spa ial imbalances. In S. Colombo (Ed.), Spa ial economics
olume I: Theo y (pp. 79–110). Sp inge In e na ional Publishing. [C ossRe ]
Glaese , E. L., Kallal, H. D., Scheinkman, J. A., & Shlei e , A. (1992). G ow h in ci ies. Jou nal o Poli ical Economy,100(6), 1126–1152.
[C ossRe ]
Go don, P., & Kou i , K. (2020). Agglome a ion and clus e s nea and a o egional de elopmen : A c i ical assessmen . Regional
Science Policy & P ac ice,12(3), 387–396. [C ossRe ]
Gu, H., Meng, X., Shen, T., & Wen, L. (2020). China’s highly educa ed alen s in 2015: Pa e ns, de e minan s and spa ial spillo e
e ec s. Applied Spa ial Analysis and Policy,13(3), 631–648. [C ossRe ]
Economies 2025,13, 167 22 o 24
Gu, J. F. (2012a). Ha monious expansion o China’s highe educa ion: A new g ow h pa e n. Highe Educa ion,63(4), 513–528.
[C ossRe ]
Gu, J. F. (2012b). Spa ial ec ui ing compe i ion in Chinese highe educa ion sys em. Highe Educa ion,63(2), 165–185. [C ossRe ]
Ha ison, J., Smi h, D. P., & Kin on, C. (2017). Rela ional egions “in he making”: Ins i u ionalizing new egional geog aphies o highe
educa ion. Regional S udies,51(7), 1020–1034. [C ossRe ]
Hu man, D., & Quigley, J. M. (2002). The ole o he uni e si y in a ac ing high ech en ep eneu ship: A Silicon Valley ale. Annals o
Regional Science,36(3), 403–419. [C ossRe ]
Ji, M., Jiao, Y. W., & Cheng, N. (2023). An Inno a i e decision-making scheme o he high-quali y economy de elopmen d i en by
highe educa ion. Jou nal o Inno a ion & Knowledge,8(2), 100345. [C ossRe ]
Jongbloed, B., Ende s, J., & Sale no, C. (2008). Highe educa ion and i s communi ies: In e connec ions, in e dependencies and a
esea ch agenda. Highe Educa ion,56(3), 303–324. [C ossRe ]
Kim, Y. J., & Lee, C.-I. (2024). Space and echnology in ca ching-up economies: “The ci y as a labo a o y o inno a ion”. The Annals o
Regional Science,73(1), 205–239. [C ossRe ]
Konzelmann, S. J., Pi elis, C. N., & Tomlinson, P. R. (2025). Is i all in Ma shall, s ill? An app ecia ion o Ma shall’s con ibu ion o
mode n economics. Camb idge Jou nal o Economics,49(3), 385–404. [C ossRe ]
K ugman, P. (1991). Inc easing e u ns and economic geog aphy. Jou nal o Poli ical Economy,99(3), 483–499. [C ossRe ]
Lao, X., Gu, H., Yu, H., & Xiao, F. (2021). Explo ing he spa ially- a ying e ec s o human capi al on u ban inno a ion in China. Applied
Spa ial Analysis and Policy,14(4), 827–848. [C ossRe ]
Lee, J. T. (2014). Educa ion hubs and alen de elopmen : Policymaking and implemen a ion challenges. Highe Educa ion,68(6),
807–823. [C ossRe ]
Lendel, I., & Qian, H. F. (2017). Inside he g ea ecession: Uni e si y p oduc s and egional economic de elopmen . G ow h and Change,
48(1), 153–173. [C ossRe ]
Li, X. Y., Li, J. L., & Wu, X. J. (2020). Uni e si y spillo e s, spa ial dis ance, and i m inno a ion: E idence a Chinese lis ed i ms.
Eme ging Ma ke s Finance and T ade,56(7), 1504–1519. [C ossRe ]
Li, Y., Zhang, J. N., Yang, X. D., Wang, W. L., Wu, H. T., Ran, Q. Y., & Luo, R. D. (2021). The impac o inno a i e ci y cons uc ion on
ecological e iciency: A quasi-na u al expe imen om China. Sus ainable P oduc ion and Consump ion,28, 1724–1735. [C ossRe ]
Liang, W., Wang, D., Gao, L., & Li, C. (2025). S udy on he impac o China’s u ban agglome a ions’ ie ed spa ial s uc u e on egional
economic esilience. PLoS ONE,20(3), e0314538. [C ossRe ]
Liu, H., Wang, L., Wang, J., Ming, H., Wu, X., Xu, G., & Zhang, S. (2024). Mul idimensional spa ial inequali y in China and i s
ela ionship wi h economic g ow h. Humani ies and Social Sciences Communica ions,11(1), 1415. [C ossRe ]
Liu, X., Wu, X., & Zhang, W. (2024). A new DEA model and i s applica ion in pe o mance e alua ion o scien i ic esea ch ac i i ies in
he uni e si ies o China’s double i s -class ini ia i e. Socio-Economic Planning Sciences,92, 101839. [C ossRe ]
Lucas, R. E. (1988). On he mechanics o economic de elopmen . Jou nal o Mone a y Economics,22(1), 3–42. [C ossRe ]
Lyu, L., Sun, F., & Huang, R. (2019). Inno a ion-based u baniza ion: E idence om 270 ci ies a he p e ec u e le el o abo e in China.
Jou nal o Geog aphical Sciences,29(8), 1283–1299. [C ossRe ]
Ma, C. Y., Wu, H. C., & Li, X. H. (2023). Spa ial spillo e o local gene al highe educa ion expendi u es on sus ainable egional
economic g ow h: A spa ial econome ic analysis. PLoS ONE,18(11), e0292781. [C ossRe ]
Maggioni, M. A., & Ube i, T. E. (2009). Knowledge ne wo ks ac oss Eu ope: Which dis ance ma e s? The Annals o Regional Science,
43(3), 691–720. [C ossRe ]
Ma shall, A. (2013). The g ow h o weal h. In A. Ma shall (Ed.), P inciples o economics (pp. 183–199). Palg a e Macmillan UK.
[C ossRe ]
Ma in, R. (2011). Regional economic esilience, hys e esis and ecessiona y shocks. Jou nal o Economic Geog aphy,12(1), 1–32.
[C ossRe ]
Mi ze, T., & Makkonen, T. (2020). When in e ac ion ma e s: The con ingen e ec s o spa ial knowledge spillo e s and in e nal R&I on
i m p oduc i i y. The Jou nal o Technology T ans e ,45(4), 1088–1120. [C ossRe ]
Mohan y, B., & Bhanumu hy, N. R. (2018). Regional g ow h policy expe ience in India: The spa ial dimension. Asia-Paci ic Jou nal o
Regional Science,2(2), 479–505. [C ossRe ]
Mok, K. H., & Ma ginson, S. (2021). Massi ica ion, di e si ica ion and in e na ionalisa ion o highe educa ion in China: C i ical
e lec ions o de elopmen s in he las wo decades. In e na ional Jou nal o Educa ional De elopmen ,84, 102405. [C ossRe ]
Mo oyama, Y. (2008). Wha was new abou he clus e heo y?: Wha could i answe and wha could i no answe ? Economic
De elopmen Qua e ly,22(4), 353–363. [C ossRe ]
Munnich, L., & Iacono, M. (2016). Compe i i e indus y clus e s and anspo a ion in Minneso a. Compe i i eness Re iew,26, 25–40.
[C ossRe ]
O a iano, G., & Thisse, J. F. (2004). Agglome a ion and economic geog aphy. In J. V. Hende son, & J. F. Thisse (Eds.), Handbook o
egional and u ban economics (Vol. 4, pp. 2563–2608). Else ie . [C ossRe ]
Economies 2025,13, 167 23 o 24
Pablo-Ma í, F., & A auzo-Ca od, J.-M. (2020). Spa ial dis ibu ion o economic ac i i ies: A ne wo k app oach. Jou nal o Economic
In e ac ion and Coo dina ion,15(2), 441–470. [C ossRe ]
Peng, C., & Xu, H. (2024). Uni e si ies and ci ies: The impac o highe educa ion on u ban inno a ion, en ep eneu ship, and economic
g ow h. Asian Economic Pape s,23(2), 33–56. [C ossRe ]
Pink-Ha pe , S. A. (2015). Educa ional a ainmen : An examina ion o i s impac on egional economic g ow h. Economic De elopmen
Qua e ly,29(2), 167–179. [C ossRe ]
Po e , M. E. (2000). Loca ion, compe i ion, and economic de elopmen : Local clus e s in a global economy. Economic De elopmen
Qua e ly,14(1), 15–34. [C ossRe ]
Rod íguez-Pose, A. (2001). Is R&D in es men in lagging a eas o Eu ope wo hwhile? Theo y and empi ical e idence. Pape s in
Regional Science,80(3), 275–295. [C ossRe ]
Rome , P. M. (1986). Inc easing e u ns and long- un g ow h. Jou nal o Poli ical Economy,94(5), 1002–1037. [C ossRe ]
Rosen eld, M. T. W., F anz, P., & Heimpold, G. (2007). Economic ‘clus e s’ in Eas Ge many: E idence on he loca ion and cha ac e is ics
o spa ially concen a ed indus ies. Pos -Communis Economies,19(1), 73–92. [C ossRe ]
San oalha, A., Biscaia, R., & Teixei a, P. (2018). Highe educa ion and i s con ibu ion o a di e se egional supply o human capi al:
Does he bina y/uni a y di ide ma e s? Highe Educa ion,75(2), 209–230. [C ossRe ]
Sla o a, S., Rubalcaba, L., & F anco-Riquelme, J. N. (2025). Unde s anding imbalanced ansmission om R&D inpu s in o inno a ion
ou pu s and impac s: E idence om Kazakhs an. Economies,13(2), 25. [C ossRe ]
Sun, Y. M., Lu, Y. L., Wang, T. Y., Ma, H., & He, G. Z. (2008). Pa e n o pa en -based en i onmen al echnology inno a ion in China.
Technological Fo ecas ing and Social Change,75(7), 1032–1042. [C ossRe ]
Tleppaye , A., Zeinolla, S., Tyulyubaye a, D., & Aben, A. (2025). Educa ion, ins i u ions, and in es men as de e minan s o economic
g ow h in Cen al Asia and he Caucasus: A panel da a analysis. Economies,13(3), 78. [C ossRe ]
T idico, P. (2011). Explaining economic de elopmen : Old and new heo ies. In Ins i u ions, human de elopmen and economic g ow h in
ansi ion economies (pp. 181–194). Palg a e Macmillan UK. [C ossRe ]
T ipa hi, S. (2021). Do mac oeconomic ac o s p omo e u baniza ion? E idence om BRICS coun ies. Asia-Paci ic Jou nal o Regional
Science,5(2), 397–426. [C ossRe ]
Ts e ko a, E., & Lome , S. (2019). Academic excellence as “compe i i eness enhancemen ” in Russian highe educa ion. In e na ional
Jou nal o Compa a i e Educa ion and De elopmen ,21(2), 127–144. [C ossRe ]
Vale o, A., & Van Reenen, J. (2019). The economic impac o uni e si ies: E idence om ac oss he globe. Economics o Educa ion Re iew,
68, 53–67. [C ossRe ]
Vaz, E. (2024). Spa ial spillo e s and knowledge s uc u es. In Regional knowledge economies: Explo ing he in e sec ion o echnology,
geog aphy, and inno a ion in he digi al e a (pp. 7–21). Sp inge Na u e Swi ze land. [C ossRe ]
Ve spagen, B. (2010). The spa ial hie a chy o echnological change and economic de elopmen in Eu ope. The Annals o Regional
Science,45(1), 109–132. [C ossRe ]
Wang, C., Wang, L., Xue, Y., & Li, R. (2022). Re ealing spa ial spillo e e ec in high- ech indus y Agglome a ion om a high-skilled
labo low ne wo k pe spec i e. Jou nal o Sys ems Science and Complexi y,35(3), 839–859. [C ossRe ]
Wilck, J. H. (2020). In oduc ion o he special issue on enginee ing economy educa ion. The Enginee ing Economis ,65(3), 177–178.
[C ossRe ]
Woldegio gis, E., Jonck, P., & Goujon, A. (2015). Regional highe educa ion e o m ini ia i es in A ica: A compa a i e analysis wi h
he Bologna P ocess. In e na ional Jou nal o Highe Educa ion,4, 241–253. [C ossRe ]
Wu, W. W., & Zhu, J. D. (2021). A e he e demons a ion e ec s o iscal expendi u es on highe educa ion in China? An empi ical
in es iga ion. In e na ional Jou nal o Educa ional De elopmen ,81, 102345. [C ossRe ]
Xiong, W., Yang, J., & Shen, W. (2022). Highe educa ion e o m in China: A comp ehensi e e iew o policymaking, implemen a ion,
and ou comes since 1978. China Economic Re iew,72, 101752. [C ossRe ]
Yang, H., Ma, G., & Tan, B. (2025). The impac o highe educa ion e i alisa ion on egional inno a ion capabili y. Finance Resea ch
Le e s,75, 106908. [C ossRe ]
Yang, Z. (2023). Human capi al space: A spa ial pe spec i e o he dynamics o people and economic ela ionships. Humani ies and
Social Sciences Communica ions,10(1), 145. [C ossRe ]
Yang, Z., Hao, P., & Cai, J. (2015). Economic clus e s: A b idge be ween economic and spa ial policies in he case o Beijing. Ci ies,42,
171–185. [C ossRe ]
Yin, X. S., Hussain, J., & Kui, Z. (2022). The spa ial dis ibu ion pa e n and o ma ion mechanism o he spa ial agglome a ion o
Chinese highe educa ion popula ion: A c oss-p o ince compa ison. Applied Spa ial Analysis and Policy,15(2), 315–338. [C ossRe ]
Yu, J., Hu, W., & Deng, T. (2024). Towa ds mo e esilien economy—Analyzing he impac o new- ype u baniza ion on u ban economic
esilience: Mechanisms and spa ial spillo e bounda ies. Economic Change and Res uc u ing,57(5), 158. [C ossRe ]
Zhang, C., Tian, L., & Fang, G. (2024). Analysis o sus ainable ans o ma ion de elopmen pa e ns and he e ogenei y o Chinese ci ies
based on spa ial gene al equilib ium model. En i onmen , De elopmen and Sus ainabili y,26(10), 25689–25715. [C ossRe ]
Economies 2025,13, 167 24 o 24
Zhang, Q. (2023). U ban– u al gap and egional inequali y o local household income in Jiangsu P o ince, China. GeoJou nal,88(6),
6589–6601. [C ossRe ]
Zhang, T., Lei, Y., & Zhang, Q. (2023). The con ibu ion o en e p ises, uni e si ies, and esea ch ins i u es o egional inno a ion
e iciency in China. Technology Analysis & S a egic Managemen , 1–12. [C ossRe ]
Zhao, J., Chen, N., & Jin, Z. (2025). The mo e cen alized he spa ial s uc u e is, he g ea e he economic g ow h? E idence om u ban
agglome a ions in wes e n China. Humani ies and Social Sciences Communica ions,12(1), 483. [C ossRe ]
Zibe i, B. F., Rexha, D., Ib aimi, X., & A diaj, B. (2022). Empi ical analysis o he impac o educa ion on economic g ow h. Economies,
10(4), 89. [C ossRe ]
Ziemnowicz, C. (2020). Joseph A. Schumpe e and inno a ion. In E. G. Ca ayannis (Ed.), Encyclopedia o c ea i i y, in en ion, inno a ion
and en ep eneu ship (pp. 1517–1522). Sp inge In e na ional Publishing. [C ossRe ]
Disclaime /Publishe ’s No e: The s a emen s, opinions and da a con ained in all publica ions a e solely hose o he indi idual
au ho (s) and con ibu o (s) and no o MDPI and/o he edi o (s). MDPI and/o he edi o (s) disclaim esponsibili y o any inju y o
people o p ope y esul ing om any ideas, me hods, ins uc ions o p oduc s e e ed o in he con en .