Evaluation of operational efficiency in China's pharmaceutical industry and analysis of environmental impacts

Sun, Jiaqiang; Rosli, Ani a Bin i; Daud, Ad ian; Yan, Xia
A icle
E alua ion o ope a ional e iciency in China's
pha maceu ical indus y and analysis o en i onmen al
impac s
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: Sun, Jiaqiang; Rosli, Ani a Bin i; Daud, Ad ian; Yan, Xia (2025) : E alua ion o
ope a ional e iciency in China's pha maceu ical indus y and analysis o en i onmen al impac s,
Economies, ISSN 2227-7099, MDPI, Basel, Vol. 13, Iss. 4, pp. 1-35,
h ps://doi.o g/10.3390/economies13040090
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/329370
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 : Angeliki N.
Menegaki
Recei ed: 10 Feb ua y 2025
Re ised: 17 Ma ch 2025
Accep ed: 20 Ma ch 2025
Published: 27 Ma ch 2025
Ci a ion: Sun, J., Rosli, A. B., Daud, A.,
& Yan, X. (2025). E alua ion o
Ope a ional E iciency in China’s
Pha maceu ical Indus y and Analysis
o En i onmen al Impac s. Economies,
13(4), 90. h ps://doi.o g/10.3390/
economies13040090
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
E alua ion o Ope a ional E iciency in China’s Pha maceu ical
Indus y and Analysis o En i onmen al Impac s
Jiaqiang Sun 1,2 , Ani a Bin i Rosli 1,* , Ad ian Daud 1,3 and Xia Yan 1
1Depa men o Social Science & Managemen , Facul y o Humani ies, Managemen & Science, Uni e si i
Pu a Malaysia, Bin ulu Campus, Nyabau Road, Bin ulu 97008, Sa awak, Malaysia;
[email p o ec ed] (J.S.); [email p o ec ed] (A.D.); [email p o ec ed] (X.Y.)
2S a egic Resea ch Cen e o Chengdu Maidison Pha maceu ical Technology Co., L d., No. 733, Eas Sec ion,
Hubin Road, Xinglong Subdis ic , Tian u New A ea, Chengdu 610000, China
3Ins i u e o Ecosys em Science Bo neo, Uni e si i Pu a Malaysia, Bin ulu Sa awak Campus, Nyabau Road,
Bin ulu 97008, Sa awak, Malaysia
*Co espondence: ani a [email p o ec ed]
Abs ac : The pha maceu ical indus y is a co ne s one o na ional economies and plays a
c i ical ole in public heal h. Howe e , China’s pha maceu ical indus y aces signi ican
challenges, including egional dispa i ies in de elopmen . The exis ing esea ch on ope a-
ional e iciency e alua ion p ima ily ocuses on inancial o inno a ion me ics, lacking a
comp ehensi e app oach. Mo eo e , s udies on he en i onmen al impac on ope a ional
e iciency o en ely on a limi ed se o indica o s, ailing o o e a holis ic unde s anding
o how en i onmen al ac o s in luence e iciency. This s udy aims o add ess hese gaps by
comp ehensi ely e alua ing ope a ional e iciency and analyzing he impac o b oade
en i onmen al ac o s on e iciency. To achie e hese objec i es, he s udy employs a Th ee-
S age Da a En elopmen Analysis me hod combined wi h P incipal Componen Analysis
o e alua e he ope a ional e iciency o he pha maceu ical indus y ac oss 31 p o inces
in China, conside ing bo h inancial and inno a ion dimensions.The indings e eal ha
o e all e iciency has imp o ed annually, wi h egional dispa i ies g adually na owing.
Speci ically, inno a ion capabili y and inno a ion en i onmen ha e a posi i e impac
on ope a ional e iciency, while li ing s anda ds and openness exhibi a nega i e co ela-
ion. Addi ionally, he cu en en i onmen al condi ions in he no hwes e n egion a e
ound o be conduci e o he de elopmen o he pha maceu ical indus y. This s udy
is he i s o in eg a e h ee-s age da a en elopmen analysis wi h p incipal componen
analysis, cons uc ing a comp ehensi e amewo k o analyzing he ela ionship be ween
en i onmen al ac o s and ope a ional e iciency. The esul s p o ide empi ical e idence
o policymake s aiming o enhance he e iciency o he pha maceu ical indus y.
Keywo ds: e iciency esea ch; en i onmen al impac ; e iciency dis ibu ion; egional
dispa i ies
1. In oduc ion
Since he Policy o China’s Re o m and Opening-Up policy, he pha maceu ical in-
dus y (PI) in China has expe ienced apid de elopmen . Pa icula ly since 2000, he
o al ou pu alue o he pha maceu ical indus y has inc eased by 13 imes, wi h he
compound annual g ow h a e o 12.85% (Na ional Bu eau o S a is ics o China,2023).
Cu en ly, China has become one o he signi ican componen s o he supply chain in he
pha maceu ical indus y. Howe e , despi e he ema kable achie emen s in ecen yea s,
Economies 2025,13, 90 h ps://doi.o g/10.3390/economies13040090
Economies 2025,13, 90 2 o 35
he pha maceu ical indus y o China s ill aces se e e challenges in e ms o inno a ion,
scale e iciency, low p o i , and sus ainable de elopmen (Bha dwaj,2024;Bo ja Reis &
Pin o,2022).
Fi s ly, China’s PI s a ed la e and in es ed less in esea ch and de elopmen (R&D),
esul ing in a lowe sha e o he in e na ional pha maceu ical ma ke and he in e na ional
compe i i eness o he pha maceu ical indus y (H. Guo & Shi,2021;Jia,2022;G. Wang
& Zhang,2023). In addi ion, China’s egional economic de elopmen is no balanced.
The e iciency o he PI in a ious egions a ies signi ican ly; whe he in he inno a ion
capaci y o he o e all le el o ope a ion, he e is an une en phenomenon. A he same
ime, because pha maceu ical companies, as pa o a highly pollu ing indus y, along wi h
he o e all de elopmen o China’s economy, some de eloped egions (such as Beijing,
Shanghai, e c.) a e g adually shi ing om he p oduc ion o aw ma e ials and chemical
d ugs o he de elopmen o biopha maceu icals and o he ields. Many pha maceu ical
companies a e, he e o e, eloca ing o mo e en i onmen ally a o able egions in sea ch
o u he de elopmen (Bai e al.,2022;Dou & Han,2019;Lai e al.,2020;T. Liu e al.,
2020). I is impo an o no e ha he daily ope a ions o an indus y o en e p ise a e
o en cons ained by egional poli ical, economic, cul u al, and echnological condi ions.
The e o e, when selec ing a de elopmen a ea, indus ies ypically p io i ize egions wi h
compa a i e ad an ages. A he same ime, he o ma ion o indus ial clus e s has a
signi ican impac on he ope a ional e iciency o local indus ies. This egional selec ion
and en i onmen al adap a ion no only de e mine he compe i i eness o en e p ises bu
also p o oundly in luence he o e all de elopmen pa e n o he indus y (P. Chen e al.,
2025;Im an e al.,2024a,2017;W. Wang e al.,2024;M. Zhang e al.,2024).
To add ess hese challenges men ioned abo e, a se ies o policies ha e been p oposed
a he na ional le el o p omo e he high-quali y de elopmen o he pha maceu ical
indus y. Fo ins ance, he 14 h Fi e-Yea Plan o he De elopmen o he Pha maceu ical
Indus y emphasizes he need o enhance esou ce u iliza ion e iciency, educe pollu ion
emissions, and achie e a g een and low-ca bon ans o ma ion. Simul aneously, local
go e nmen s a e explo ing di e en ia ed s a egies aligned wi h egional de elopmen
objec i es, aiming o s eng hen policy suppo and en i onmen al egula ion o os e he
g ow h o he pha maceu ical indus y. Howe e , he implemen a ion o hese policies
aces se e al challenges. No ably, he e is a lack o sys ema ic quan i a i e esea ch on he
speci ic impac o en i onmen al cons ain s on indus ial e iciency. Addi ionally, policy
o mula ion o en lacks a su icien ly a ge ed and e idence-based ounda ion. Add essing
how o comp ehensi ely imp o e he e iciency o he egional pha maceu ical indus y
unde policy guidance has become a c i ical and u gen issue equi ing esolu ion (Y. Liu
e al.,2022;Xu e al.,2022).
Despi e he impo ance o hese endea o s, he e a e s ill signi ican esea ch gaps on
how o assess he e ec i eness o he implemen a ion o hese policies, and in pa icula ,
how hese policies speci ically a ec he de elopmen o he pha maceu ical indus y. Mos
o he cu en e alua ions o i m e iciency ocus on a single dimension such as inance o
inno a ion, neglec ing he ac ha he pha maceu ical indus y is a combina ion o bo h
dimensions. In addi ion, he exis ing s udies mainly limi he en i onmen al impac o a
single indica o , adop ing one indica o o ep esen a ce ain e alua ion dimension and
lacking a sys ema ic s udy o he o e all en i onmen al impac on e iciency.
To ill he cu en esea ch gaps desc ibed abo e, his s udy e alua ed PI e iciency
om a comp ehensi e pe spec i e and e eals he in luence o comp ehensi e en i on-
men al ac o s on e iciency.
To achie e he esea ch objec i es, his s udy ocused on he pha maceu ical indus y ac oss
31 p o incial-le el adminis a i e uni s in China. I employed a h ee-s age da a en elopmen
Economies 2025,13, 90 3 o 35
analysis (DEA) me hod combined wi h p incipal componen analysis (PCA) o comp ehensi ely
e alua e he ope a ional e iciency o China’s pha maceu ical indus y in bo h bo h inancial
and inno a ion dimensions. PCA is used o ex ac key componen s om a la ge numbe o
en i onmen al ac o s, which a e hen used as en i onmen al a iables o e eal he impac o
he comp ehensi e ope a ional en i onmen on he e iciency o he pha maceu ical indus y.
The inno a ions o his s udy a e e lec ed in he ollowing aspec s: Fi s , he exis ing
esea ch on he impac o en i onmen al ac o s on e iciency ypically elies on single
indica o s o subs i u e o speci ic dimensions in eg ession analysis. In con as , his
s udy employs PCA o ex ac p incipal componen s om a la ge numbe o en i onmen al
indica o s, he eby enhancing he scien i ic igo and easibili y o he conclusions. Sec-
ond, he in eg a ion o PCA e ec i ely elimina es mul icollinea i y issues, imp o ing he
obus ness and explana o y powe o he eg ession model. Finally, in he e alua ion o
pha maceu ical-indus y ope a ional e iciency, p e ious s udies ha e p edominan ly o-
cused on a single dimension, such as inancial o inno a ion indica o s. This s udy adop ed
a mul idimensional, comp ehensi e e alua ion app oach, p o iding a mo e holis ic and
p ecise assessmen o pha maceu ical-indus y e iciency.
This s udy has heo e ical and p ac ical signi icance. By in eg a ing mul iple e alu-
a ion dimensions and employing he combina ion o he h ee-s age DEA and PCA, his
esea ch p o ides a mo e comp ehensi e assessmen o he cha ac e is ics o he pha -
maceu ical indus y and i s in luencing ac o s, add essing he limi a ions o adi ional
single-dimensional analysis. Fu he mo e, he s udy conduc ed an in-dep h analysis o
he d i e s o e iciency changes, e ealing he e ec s o ac o s such as pu e echnical
e iciency and economies o scale on pe o mance and he eby o e ing a e ined heo e ical
amewo k o e iciency analysis. F om an economic pe spec i e, he esea ch u he
in es iga es egional in luencing ac o s, explo ing how economic elemen s such as he local
economy and local inno a ion in luence he e iciency o he PI. These analyses no only
enhance he unde s anding o e iciency dynamics in he pha maceu ical sec o bu also
p o ide a heo e ical ounda ion o o mula ing a ge ed indus y policies and op imizing
esou ce alloca ion, unde sco ing he s udy’s academic and economic signi icance.
This s udy is o ganized in o six sec ions. Sec ion 1ou lines he esea ch backg ound,
iden i ies he esea ch p oblem, and de ines he s udy objec i es. Sec ion 2conduc s
a comp ehensi e li e a u e e iew, summa izing ele an heo e ical amewo ks and
e iciency measu emen me hodologies, jus i ying he adop ion o he h ee-s age app oach,
and highligh ing gaps in he exis ing li e a u e and he concep ual amewo k o his s udy.
Sec ion 3de ails he esea ch me hodology, encompassing da a sou ces, heo e ical models,
and he selec ion o a iables. Sec ion 4p esen s he empi ical esul s, beginning wi h
da a alida ion, ollowed by he i s -s age esul , p incipal componen ex ac ion and
eg ession analysis (second s age), and DEA-based e iciency measu emen ( hi d s age).
Sec ion 5discusses he dis ibu ion o e iciency, i s empo al e olu ion o e a decade,
egional e iciency dispa i ies, he in luence o en i onmen al ac o s on e iciency ac oss
egions, and a compa a i e analysis be ween he indings o his s udy and p io esea ch.
Finally, Sec ion 6concludes he s udy by summa izing he main indings, he implica ion
o policy-make s, and he limi a ions o his s udy, p oposing di ec ions o u u e esea ch.
2. Li e a u e Re iew
2.1. The Basic Theo y and Gene al Measu emen Me hods o E iciency
Value maximiza ion heo y: This heo y posi s ha an indus y o en e p ises should
balance inancial pe o mance wi h he conside a ion o a ious s akeholde s’ in e es s and
he equilib ium be ween long- e m and sho - e m goals. Speci ically, a comp ehensi e
e alua ion o he indus y o companies should in eg a e bo h inancial ou comes and inno-
Economies 2025,13, 90 4 o 35
a ion, e lec ing he balance o sho - e m and long- e m in e es s. The e o e, his heo y
emphasizes he necessi y o e alua ing he indus y using he inancial and inno a ion
dimensions (Colm,1960;F iedman,2007;G ossman & S igli z,1977).
Regional clus e economic heo y: Michael Po e ’s egional clus e economic heo y
sugges s ha he compe i i eness o an indus y is in luenced by mul iple ac o s such as
echnology, go e nmen policies, and local esou ces. Al hough en i onmen al egula ions
may pose challenges, hey can os e echnological inno a ion, enhancing p oduc i i y and
compe i i eness. Fu he mo e, egional clus e s acili a e coope a ion, knowledge sha ing,
and echnological ad ancemen among i ms, he eby s eng hening he compe i i e ad an-
age o en e p ises and p omo ing o e all egional economic g ow h. The e o e, he s udy
o he impac o egional en i onmen s on indus y e iciency is o signi ican heo e ical
and p ac ical impo ance (Ma in & Sunley,2003;Po e ,1990b,1998).
Resou ce-based iew (RBV) heo y: The esou ce-based iew (RBV) heo y asse s ha
he compe i i e ad an age o i ms and indus ies s ems om hei unique esou ces and ca-
pabili ies, pa icula ly hose ha possess alue, a i y, inimi abili y, and non-subs i u abili y
(VRIN). E ec i e esou ce alloca ion, especially he in eg a ion and dynamic adjus men o
esou ces, is c ucial o main aining compe i i e ad an age. This heo y p o ides a heo-
e ical ounda ion o analyzing local indus y esou ces and hei impac on compe i i e
ad an age (J. Ba ney,1991;J. B. Ba ney & A ikan,2005;Ge e al.,2024;Kau & Singh,2024;
Teece e al.,1997;M. Zhang e al.,2024).
E iciency, as a concep , was i s a icula ed by he I alian economis Pa e o, who
amed i as an op imal s a e o esou ce alloca ion, commonly called “Pa e o op imali y”. In
his s a e, no ealloca ion o esou ces can inc ease he bene i o one pa y wi hou educing
he bene i o o he s. Pa e o op imali y undamen ally e eals he economic meaning o
e iciency, which is o achie e op imal economic ou pu h ough he a ional dis ibu ion o
esou ces. Building on his heo y, Fa ell (1957) u he de eloped he heo y o e iciency
by decomposing i in o echnical e iciency and alloca i e e iciency. Fa ell’s wo k laid
a solid ounda ion o mode n e iciency e alua ion heo y and p o ided a heo e ical
basis o he de elopmen o e iciency measu emen me hods. E iciency measu emen
me hods a e mainly di ided in o wo ca ego ies: pa ame ic and non-pa ame ic me hods.
Among hese, s ochas ic on ie analysis (SFA), as a ep esen a i e o pa ame ic me hods,
is commonly used, while da a en elopmen analysis (DEA) is a ypical non-pa ame ic
me hod (G eene,2008;Lampe & Hilge s,2015;Res i,2000).
2.2. E iciency Measu emen Me hods: A Compa ison o SFA and DEA
SFA, a classical pa ame ic me hod, was p oposed by Aigne e al. (1977). The basic
p inciple o SFA is o decompose he e iciency o decision-making uni s (DMUs) in o wo
pa s: andom e o and an ine iciency e m. The andom e o e lec s he impac o
uncon ollable ac o s such as he ex e nal en i onmen , such as wea he , policy changes, o
ma ke luc ua ions; and he ine iciency e m e lec s he loss o e iciency caused by he
in e nal managemen o echnical le el o he i m (Aigne e al.,1977;Li & Fan,2009). SFA
e alua es he echnical e iciency and he o al e iciency by cons uc ing a p oduc ion on ie
unc ion combined wi h maximum likelihood es ima ion echniques. The ad an age o his
me hod is ha i can dis inguish be ween he impac o ex e nal uncon ollable ac o s and
in e nal managemen ac o s on e iciency, hus making he assessmen esul s close o he
ac ual si ua ion. Due o i s wide applicabili y, SFA has been widely used in many ields such
as ag icul u e, indus y and en i onmen al esea ch (Kumbhaka & Tsionas,2011;Lee & Jeon,
2023;Ðoki´c e al.,2022;Singh e al.,2020;J. Wang e al.,2020;R. Wang & Duan,2023).
Howe e , as a pa ame ic me hod, SFA equi es assump ions abou he unc ional
o m o he p oduc ion unc ion and he dis ibu ion o ine iciency and andom e o . I

Economies 2025,13, 90 5 o 35
he assump ions de ia e om he ac ual si ua ion, i may lead o biased e iciency es ima es.
Mo eo e , he applica ion o SFA in ol es ce ain limi a ions when dealing wi h inpu s
and ou pu s ha ha e di e en uni s and dimensions (Ahmed & Melesse,2018;Kali ajan &
Shand,1994;Madaleno & Mou inho,2023;Moulay Ali e al.,2024).
DEA is a me hod ha cons uc s an bes -p ac ice on ie based on he DMUs and
hen compa es o he DMUs wi h his o de e mine hei ela i e e iciency. This me hod
o e s se e al ad an ages in e iciency e alua ion. Fi s , i does no equi e p ede ined
assump ions abou he unc ional o m o he p oduc ion unc ion, hus a oiding he model
bias ha may esul om inco ec unc ion speci ica ion (Cha nes e al.,1978). Second, DEA
can handle mul i-inpu , mul i-ou pu e iciency issues simul aneously wi hou he need o
weigh hese a iables, which gi es i an ad an age when dealing wi h mul idimensional
da a (Banke e al.,1984). Mo eo e , DEA can iden i y sou ces o ine iciency, such as scale
ine iciency o insu icien echnical e iciency, p o iding decision-make s wi h di ec ions
o imp o emen (Coope e al.,2007). In e ms o sample size equi emen s, DEA is well
sui ed o small sample da a, making i pa icula ly use ul o indus y s udies o case
analyses (Zhu,2009). Addi ionally, DEA can p o ide imp o emen pa hs o non-DEA
e icien DMUs by analyzing e e ence se s and cla i ying he di ec ions o adjus ing
inpu s o ou pu s (Tone,2001). Finally, he lexibili y o DEA models is conside able, wi h
he abili y o inco po a e a ious ex ensions, such as h ee-s age DEA, weigh - es ic ed
models, and indi ec economic e iciency models (Camanho e al.,2024;Me goni e al.,
2024). Wi h he de elopmen o DEA esea ch, i has ound wide applica ion in e iciency
es ing ac oss a ious indus ies (And é e al.,2024;C. Guo e al.,2024;Y. Guo e al.,2024;
Rashid e al.,2024;Sun e al.,2024).
Howe e , DEA also aces ce ain limi a ions. Fi s , i canno sepa a e he in luence o
ex e nal en i onmen al ac o s and andom noise on e iciency, which may lead o e iciency
es ima es being a ec ed by ex e nal condi ions o da a luc ua ions. Second, DEA is unable
o u he analyze he speci ic sou ces o ine iciency o non-DEA e icien DMUs, such as
whe he i is due o managemen issues, en i onmen al cons ain s, o andom ac o s (Coelli
e al.,2005;Dyson e al.,2001;Hjalma sson e al.,1996;Joe & Wu,1996;Ma,2010).
2.3. The Me hod o Th ee-S age DEA and Ad an age
To add ess he limi a ions o adi ional da a en elopmen analysis (DEA) in handling
e iciency measu emen e o s caused by en i onmen al ac o s and he cons ain s o
s ochas ic on ie analysis (SFA), which equi es assump ions abou e o dis ibu ion
and s uggles wi h mul iple inpu s and ou pu s, F ied e al. (2002) p oposed he h ee-
s age DEA me hod. This app oach in eg a es DEA and SFA o imp o e he accu acy o
e iciency measu emen . In he i s s age, a con en ional DEA model is used o compu e
ini ial e iciency sco es and iden i y slack a iables. In he second s age, SFA is applied o
eg ess he slack a iables agains en i onmen al ac o s, allowing o he sepa a ion o
en i onmen al in luences and s ochas ic dis u bances on e iciency. Finally, in he hi d
s age, a e elimina ing he e ec s o en i onmen al and andom ac o s, he adjus ed inpu
da a a e used in a inal DEA calcula ion o ob ain a mo e p ecise measu e o echnical
e iciency (Cha nes e al.,1978;F ied e al.,2002;Luo,2012).
Compa ed o adi ional DEA me hods, he h ee-s age DEA app oach o e s signi -
ican ad an ages. I e ains he lexibili y o DEA, such as no equi ing a p ede ined
p oduc ion unc ion, while also isola ing he e ec s o ex e nal en i onmen al ac o s
and s ochas ic noise, he eby enhancing he accu acy o e iciency e alua ion (A ki an
& Rowlands,2008;F ied e al.,2002;Na e al.,2019;Z. Wang e al.,2017). Consequen ly,
his me hod has been widely applied in a ious ields, including ene gy managemen ,
en i onmen al go e nance, public se ices, he pha maceu ical indus y, and inno a ion
Economies 2025,13, 90 6 o 35
e iciency assessmen , p o iding a mo e eliable ool o e iciency measu emen (G. Chen
& Chen,2024;Guanglan & Zhening,2024;Shi e al.,2025;Song & Ma,2024;Z. Wang e al.,
2024;Wei & Zhao,2024;L. Zhang & Cui,2024).
2.4. The Cu en Resea ch on E iciency in he PI
Cu en ly, he e is limi ed esea ch on he e iciency o he pha maceu ical indus y
(PI), and exis ing s udies end o ocus on a single dimension, such as inno a ion e iciency
o inancial e iciency.
2.4.1. Inno a ion E iciency Resea ch
Inno a ion is one o he key cha ac e is ics o he pha maceu ical indus y. Howe e ,
esea ch on inno a ion e iciency is s ill e y limi ed. In s udies o lis ed pha maceu ical
companies o he pha maceu ical indus y, Xiong and Meng (2019) used DEA analysis
o ind ha , in China’s PI, he e iciency o biopha maceu icals is he highes , wi h he
main ine iciency s emming om pu e echnical ine iciency. The e iciency o ansla ing
inno a ion in o e enue is also low, p ima ily due o excessi e esea ch and de elopmen
in es men . Mo eo e , in China, inno a ion e iciency is no only low bu also une enly
dis ibu ed (Hao & Ruan,2022;Lai e al.,2020;Qiu e al.,2023).
In e na ional esea ch has also been conduc ed on pha maceu ical inno a ion. SFA
and mul iple- on ie analysis we e applied o e alua e he 705 pha maceu ical companies’
e iciency in Ame ica, e ealing ha di e en open inno a ion app oaches had dis inc im-
pac s on pe o mance (Shin e al.,2018). In e ms o he e iciency o la ge global en e p ises,
i was ound ha e en globally enowned companies ace e iciency challenges (Gascón
e al.,2017;Schuhmache e al.,2023,2021).
2.4.2. Pha maceu ical Companies’ Financial E iciency E alua ion
Financial e iciency e alua ion is a common me hod o assessing companies. In 2013,
a me hod combining da a en elopmen analysis (DEA) and s ochas ic on ie analysis
(SFA) was applied, and he esul s e ealed ha inc eased echnical knowledge ese es
signi ican ly imp o ed company e enue, wi h simila conclusions d awn using di e en
DEA models (Cai & Sun,2013). Xia e al. (2022) applied he BCC-DEA o e alua e he
ope a ional e iciency o public pha maceu ical i ms in China. Thei analysis indica ed a
gene al decline in o e all inancial e iciency, wi h he excep ion o he biopha maceu ical
sec o , which exhibi ed g ow h. In con as , he chemical pha maceu icals and adi ional
Chinese medicine sec o s expe ienced a educ ion in e iciency. Simila ly, Lin e al. (2021)
u ilized a wo-s age ne wo k DEA app oach combined wi h Malmquis indices o assess
he impac o go e nmen subsidies. Thei indings sugges ed ha such subsidies had no
signi ican e ec on inancial e iciency. Fu he esea ch by Yang (2024) applied a h ee-
s age DEA model and Malmquis indices, e ealing ha he o e all e iciency o Chinese
pha maceu ical companies emains ela i ely low, wi h no able yea - o-yea a iabili y.
Rega ding he selec ion o indica o s, domes ic esea che s ypically ely on ope a ing
income and p o i s as ou pu me ics due o he accessibili y o hese da a. In con as ,
in e na ional schola s o en inco po a e a wide a ay o indica o s, including human capi al
e iciency, s uc u al capi al e iciency, in ellec ual capi al e iciency, ea nings pe sha e,
di idends pe sha e, and e u ns on equi y (Hamad & Ta noczi,2021;Riaz e al.,2023).
2.5. Resea ch Gap
Up o now, he esea ch on he e iciency o pha maceu ical companies and he pha -
maceu ical indus y has been limi ed bo h in e ms o he quan i y o s udies and he dep h
o hei in eg a ion.
Economies 2025,13, 90 7 o 35
(1) Cu en e alua ions o he ope a ional e iciency o he PI a e based solely on inancial
o inno a ion dimensions, lacking comp ehensi e esea ch.
In he inno a ion dimension, schola s ypically use indica o s such as sales o new
p oduc s and he numbe o pa en applica ions (Hao & Ruan,2022;Lai e al.,2020;
Qiu e al.,2023). O he schola s also e alua e he ope a ional e iciency use he new
molecula en i y (NME) and impac ac o s o he publica ion (Gascón e al.,2017;
Schuhmache e al.,2023,2021). The inancial e alua ion dimension includes a b oade
ange o indica o s, wi h ope a ion e enue and ope a ion p o i being he mos widely
used (Gascón e al.,2017;Lin e al.,2021;Yang,2024). O he inancial indica o s, such
as asse u no e , e u ns on equi y (ROE), and ea nings pe sha e (EPS), a e also
employed (Hamad & Ta noczi,2021;Riaz e al.,2023;Xia e al.,2022). To da e, he only
s udy ha has combined bo h inancial and inno a ion dimensions in e alua ing he
ope a ional e iciency o pha maceu ical companies is Gascón e al. (2017). No o he
esea ch has conduc ed a combina ion e alua ion o bo h dimensions. Howe e , he
inancial indica o s p ima ily e lec a company’s sho - e m p o i abili y; inno a ion
se es as a measu e o i s long- e m g ow h po en ial. The e o e, a sepa a e e alua ion
o he inancial and inno a ion pe o mance o e looks he balance be ween sho - e m
and long- e m in e es .
(2)
The cu en esea ch on he impac o en i onmen al ac o s on e iciency is limi ed.
Al hough p e ious s udies ha e employed h ee-s age DEA o analyze he impac o he
en i onmen on e iciency, he use o a single indica o o ep esen a dimension lacks a
comp ehensi e unde s anding o he en i onmen . Qiu e al. (2023) used he numbe o
employees o ep esen company size, he numbe o employees wi h a bachelo ’s deg ee
o highe o ep esen employee quali y, and e u ns on equi y (ROE) and he a io o o al
liabili ies o o al asse s (LEV) as en i onmen al indica o s. Yang (2024) used go e nmen
subsidies, pe capi a GDP, and he yea s ha a company has been es ablished o measu e
and explain he en i onmen al in luence on e iciency. Sun e al. (2024), cons uc
en i onmen al indica o s using pe capi a disposable income o ep esen weal h le els,
he wo king-age popula ion o ep esen he labo supply, and local GDP o ep esen
he local economic le el. Al hough hese s udies explo e he impac o en i onmen al
ac o s on e iciency om di e en pe spec i es, hey emain inadequa e. En i onmen al
ac o s a e complex, and using a single indica o o eg ession analysis canno ully
cap u e he o e all en i onmen al impac . Mo eo e , when oo many en i onmen al
indica o s a e included in eg ession, po en ial mul icollinea i y issues may a ise, which
could comp omise he accu acy o eg ession esul s (Bai d & Biebe ,2016;Hai o sky,
1969;Sh es ha,2020).
In conclusion, he e a e no able esea ch gaps in he cu en esea ch ega ding he
comp ehensi e e iciency e alua ion o he PI and he in-dep h explo a ion o how he
en i onmen al ac o s impac e iciency. Add essing hese gaps will no only con ibu e o
he heo e ical amewo k bu also p o ide a mo e scien i ic implica ion o policy-make s
and companies in he PI, imp o ing hei ope a ion e iciency.
2.6. The Concep ual F amewo k o This S udy
Based on he esea ch gap, he s udy had wo p ima y objec i es: i s , o accu a ely
measu e he ope a ional e iciency o he PI in bo h he inancial and inno a ion dimensions
and, second, o e eal he impac o comp ehensi e en i onmen al ac o s on ope a ional
e iciency. To achie e hese esea ch objec i es, he s udy employed a combined me hod o
h ee-s age DEA and PCA. The p ocess can be di ided in o h ee s ages.
(1)
S age 1: ini ial e iciency measu emen .
Economies 2025,13, 90 8 o 35
In his s age, DEA was used o calcula e he e iciency alues o 31 p o incial-le el
egions in China. The main objec i e was o measu e he cu en e iciency dis ibu ion
and calcula e he sales o inpu a iables.
(2)
S age 2: en i onmen al ac o analysis and adjus men .
A he beginning o his s age, PCA was employed o ex ac p inciple componen s
om mul iple po en ial en i onmen al a iables. The en i onmen al a iable alues
we e calcula ed based on he a iance con ibu ion o each componen . The ex ac ed
en i onmen al a iables we e used as independen a iables (IV), and he slacks o he
inpu s om S age 1 we e used as he dependen a iable (DV) in he SFA eg ession.
This s age aimed o e eal how he comp ehensi e en i onmen in luences e iciency
and hen elimina e he impac o en i onmen al ac o s and andom dis u bances on
he inpu a iables by adjus ing he inpu indica o s.
(3)
S age 3: e iciency adjus men and ecalcula ion.
Adjus ed inpu s we e ob ained in S age 2, and he o iginal ou pu was employed in
his s age o ecalcula e he inal and accu a e e iciency using he DEA model. This
s age aimed o measu e he accu a e e iciency o PI in China while emo ing he
e ec s o en i onmen al ac o s.
The concep ual amewo k o his s udy was p ima ily based on he heo e ical ame-
wo ks o Zhao e al. (2019) and L. Zhang and Cui (2024), which o m he analy ical model
o his esea ch, as well as he PCA-DEA app oach p oposed by S e i´c e al. (2022). Figu e 1
illus a es he concep ual amewo k o his s udy.
Figu e 1. The concep ual amewo k o his s udy.
Economies 2025,13, 90 15 o 35
encompass aspec s such as economic ounda ions, consump ion le els, o eign in es men ,
local iscal e enue, in es men in echnology and educa ion, and he composi ion o he
labo o ce.
(1)
S ep 1: Me hod sui abili y e i ica ion.
The sui abili y o he da a o p incipal componen analysis (PCA) was e i ied using
he Kaise –Meye –Olkin (KMO) measu e, and cumula i e a iance was explained.
When he KMO alue is g ea e han 0.6, i indica es s ong co ela ions be ween he
a iables, making he da a sui able o PCA. I he KMO alue is below 0.6, i sugges s
weak co ela ions be ween he a iables, making he da a unsui able o PCA. Addi-
ionally, when he cumula i e a iance explained is 70% o highe , i indica es ha
he ex ac ed p incipal componen s e ec i ely explain he majo i y o he a ia ion in
he o iginal da a, making he da a sui able o u he analysis. Con e sely, i he cu-
mula i e a iance explained is below 70%, i sugges s ha he ex ac ed componen s
do no adequa ely cap u e he main in o ma ion in he da a (Hai e al.,2010). Table 4
shows he KMO measu e and Ba le ’s es esul s, while Table 5p esen s he analysis
o cumula i e a iance be o e and a e o a ion.
As shown in Table 4, he KMO alue is 0.902, which exceeds he accep able h eshold
o 0.60, indica ing ha hese en i onmen al ac o s a e well sui ed o he p incipal
PCA me hod. Fu he mo e, Ba le ’s es o sphe ici y esul s (
χ2=15737, d =253,
p<0.001
) demons a e signi ican co ela ions be ween he a iables, u he alida -
ing he app op ia eness o conduc ing PCA.
As shown in Table 5, a e PCA was applied, only ou main p inciple componen s
we e ex ac ed om he 23 po en ial en i onmen al ac o s. These p incipal compo-
nen s collec i ely al eady explain 89.35% o he o al a iance, indica ing a e y high
le el o explana o y powe in he da ase .
Table 4. The esul o KMO and Ba le ’s es .
KMO measu e o sampling adequacy 0.902
Ba le ’s es o sphe ici y app oxima e Chi-Squa e 15,737
Deg ees o eedom 253
Signi icance 0.000
Table 5. Eigen alues and a iance explained be o e and a e o a ion.
Componen Ini ial Eigen alues Ex ac ion Sums o Squa ed Loadings Ro a ion Sums o Squa ed Loadings
To al % Va iance Cum. % To al % o Va iance Cum. % To al % o Va iance Cum. %
1 15.18 66.00 66.00 15.18 66.00 66.00 11.82 51.37 51.37
2 2.95 12.83 78.83 2.95 12.83 78.83 4.19 18.21 69.58
3 1.40 6.07 84.90 1.40 6.07 84.90 2.37 10.29 79.87
4 1.02 4.43 89.33 1.02 4.43 89.33 2.18 9.46 89.33
5 0.58 2.56 90.81 - - - - - -
... - - - - - - - - -
23 0.03 0.02 100.00 - - - - - -
(2)
S ep 2: Ro a ed componen .
In PCA, he ini ially ex ac ed componen s may be complex, wi h a iable loadings
ha a e di icul o clea ly dis inguish, which hinde s he in e p e a ion o he esul s.
The e o e, o a ing he componen s helps simpli y he s uc u e o a iable loadings,
allowing each a iable o concen a e on a ew main componen s. This inc eases
he in e p e abili y o he p incipal componen s, cla i ies hei ac ual meaning, and
enhances he explana o y powe ega ding he s udy subjec . Table 6p esen s he
esul s o he o a ed componen loading ma ix.

Economies 2025,13, 90 16 o 35
Based on he esul s o he o a ed componen ma ix in Table 6, ou p incipal compo-
nen s we e ex ac ed. To be e in e p e he en i onmen al ac o s e lec ed by each
p incipal componen , only a iables wi h loadings g ea e han 0.5 we e e ained. Fo
he pu pose o he subsequen e iciency analysis, he ex ac ed ac o s we e classi ied
and named acco ding o he cha ac e is ics o he p incipal componen s.
Table 6. Ro a ed componen loadings
Componen P inciple
Componen 1
P inciple
Componen 2
P inciple
Componen 3
P inciple
Componen 4
Wa e Pollu ion Equi alen - - 0.769 -
Ai Pollu ion Equi alen - - 0.807 -
Full- ime R&D Hou s 0.927 - - -
Annual Numbe o R&D P ojec s 0.916 - - -
Annual R&D In es men 0.899 - - -
New P oduc P ojec s 0.95 - - -
New P oduc In es men 0.933 - - -
Annual New P oduc Sales 0.922 - - -
Au ho ized In en ions 0.76 0.516 -
Au ho ized U ili y Models 0.898 - - -
Au ho ized Designs 0.885 - - -
Pe Capi a Disposable Income - 0.905 - -
Pe Capi a Consump ion Le el - 0.908 - -
Numbe o Fo eign-Funded En e p ises Regis e ed 0.723 - - -
Fo eign In es men Amoun - - - 0.895
Regis e ed Capi al o Fo eign In es men - - 0.964
Highe Educa ion En ollmen 0.727 - - -
Local Gene al Budge Re enue 0.777 - - -
Go e nmen Suppo o Educa ion 0.809 - - -
Go e nmen Suppo o Science and Technology 0.809 - - -
Go e nmen Suppo o En i onmen al P o ec ion 0.563 - - -
Regional GDP 0.832 - - -
Pe Capi a GDP - 0.880 - -
(3)
S ep 3: Name componen .
P incipal Componen 1: economic and echnological ounda ion le el (Z1).
The i s p incipal componen includes a iables such as egional economic le els,
go e nmen iscal e enue, and go e nmen in es men s in echnology, educa ion,
inno a ion, and echnological ou pu s. The e o e, his componen e lec s he egion’s
economic de elopmen and inno a ion capaci y, and i is named “Economic and
echnological ounda ion le el”.
P incipal Componen 2: esiden s’ li ing s anda ds (Z2).
The second p incipal componen comp ises a iables like pe capi a GDP, esiden in-
come, and consump ion le els, which e lec he li ing quali y and economic s a us o
he egion’s esiden s. Thus, his componen is named “Residen s’ Li ing S anda ds”.
P incipal Componen 3: local pollu ion le els (Z3).
The hi d p incipal componen p ima ily consis s o was e emissions and pollu ion
le els in he na u al en i onmen . The highe he pollu an emissions, he highe he
local pollu ion le el, which is why his componen is named “Local Pollu ion Le el”.
P incipal Componen 4: openness o he o eign ma ke (Z4).
The ou h p incipal componen is mainly composed o a iables ela ed o o eign
in es men , e lec ing he egion’s le el o openness and i s abili y o a ac o eign
capi al. Hence, his componen is named “Openness o he o eign ma ke ”.
Economies 2025,13, 90 17 o 35
4.3.2. The Resul o he SFA Reg ession
The objec i e o his s age was o e eal how en i onmen al a iables a ec e iciency.
To achie e his objec i e, he p incipal componen s ex ac ed om he PCA in he p e ious
s age we e used as he IV, while he inpu slacks ob ained in he i s s age we e used as he
DV, and SFA eg ession was hen conduc ed using F on ie 4.1 o examine he impac o
en i onmen al a iables on e iciency. The eg ession esul s a e p esen ed in Table 7.
Table 7. Resul s o SFA on he impac o en i onmen al a iables on inpu slacks.
Independen Va iable Dependen Va iable
Slack o To al Asse Slack o Employees Numbe Slack o R&D In es men
Cons an Te m β0−17973.69 *** −2638.33 *** −32.16 ***
- a io −14.03 −23.19 −5.09
Economic and echnological
ounda ion (Z1)
β1−6568.12 *** 218.76 4.51 *
- a io −2.68 0.70 1.77
Residen s’ li ing
s anda ds (Z2)
β216,586.74 *** 2117.79 *** 40.48 ***
- a io 25.28 5.75 10.56
Local pollu ion le els (Z3)β37163.00 *** 2942.94 ** −42.76 **
- a io 39.44 2.01 −2.41
Openness o he
o eign ma ke (Z4)
β445,187.43*** −4467.38 *** 4.66
- a io 9.31 −4.63 0.10
σ23,560,829,700 57,626,093.00 33,054.70
γ0.999 0.999 0.999
LR es o he one-sided e o 204.38 204.72 310.85
No e: *, **, and *** indica e signi icance a he 10%, 5%, and 1% le els, espec i ely. Only coe icien s wi h
a signi icance le el o 5% o lowe (
p≤
0.05) a e included in subsequen calcula ions, while hose abo e 5%
(p>0.05) a e excluded.
As shown in Table 7, he likelihood a io (LR) es alues o he h ee inpu slack
a iables a e 204.38, 204.72, and 310.85, espec i ely, all signi ican ly exceeding he c i ical
alue o he mixed chi-squa e dis ibu ion a he 1% signi icance le el (10.83). This con i ms
he sui abili y o he model and he alidi y o he assump ions. A he same ime, he
γ
alues a e close o 1, indica ing ha he inpu slack is p ima ily caused by manage ial
ine iciency, a he han andom noise. This sugges s ha he e is signi ican po en ial o
managemen op imiza ion o imp o e ope a ional e iciency. By op imizing managemen
and adjus ing s a egies, inpu slack can be e ec i ely educed, u he enhancing he
ope a ional e iciency o he pha maceu ical indus y.
F om Table 7, i can be seen ha ex e nal en i onmen al ac o s ha e a complex
impac on he ope a ional e iciency o he pha maceu ical indus y. Fo example, he local
economic and echnological inno a ion capabili ies can educe asse alloca ion h ough
inno a ion, he eby educing he en e p ise’s inpu . On he o he hand, he s anda d o
li ing inc eases inpu om a ious aspec s, indica ing ha , in egions wi h highe li ing
s anda ds, inpu in he pha maceu ical indus y inc eases, which is no conduci e o i s
de elopmen . Regions wi h a highe en i onmen al capaci y end o inc ease in es men
in labo and asse s, bu hey can educe R&D in es men . In con as , egions wi h highe
openness o he ou side wo ld inc ease in es men in asse s while educing he numbe o
employees, indica ing ha o eign in es men upg ades asse alloca ion and educes labo
inpu . A de ailed analysis o hese impac s is p esen ed in Sec ion 5.
Ano he esea ch objec i e o his s age was o exclude he in luence o en i onmen al
ac o s and andom noise on e iciency o p o ide a mo e equi able en i onmen and place
Economies 2025,13, 90 18 o 35
all DMUs a he same le el o andomness, ul ima ely calcula ing he adjus ed inpu alues.
The speci ic calcula ion p ocess was as ollows.
A e F on ie was un, he en i onmen al alues we e calcula ed using he coe icien s
in Table 7, and he p incipal componen sha es we e ob ained h ough PCA. Subsequen ly,
Fo mulas
(4)
–
(9)
we e applied o calcula e manage ial ine iciency, and andom e o s
we e es ima ed using Fo mula
(10)
. Finally, adjus ed inpu alues we e ob ained h ough
Fo mula
(11)
. This se ies o calcula ions elimina ed he in e e ence o en i onmen al
impac s and andom e o s, ensu ing ha he e iciency alues e lec only he manage ial
issues wi hin he en e p ises hemsel es.
4.4. The Resul o S age 3
In S age 3, he BCC-DEA model was employed, wi h he adjus ed inpu s ob ained in
S age 2 and ini ial ou pu s. All he esul s a e p esen ed in Table 2.
As shown in Table 2, he o e all a e age echnical e iciency alues in he i s and
hi d s ages do no di e signi ican ly, wi h he mos no able di e ence obse ed in pu e
echnical e iciency. Since his s udy uses c oss-sec ional da a, he e iciency alues ob ained
ep esen he ela i e e iciency be ween he DMUs. In he hi d s age, Tianjin emains he
egion wi h he highes pha maceu ical-indus y e iciency, while Liaoning and Jiangxi
ha e shown signi ican imp o emen s. This sugges s ha he e iciency o hese wo egions
was unde es ima ed in he p e ious s age and indica es ha he en i onmen al condi ions
in hese egions a e no conduci e o he de elopmen o he pha maceu ical indus y. The
egions wi h he lowes echnical e iciency a e he same as in he i s s age, indica ing ha
he de elopmen o he pha maceu ical indus y in hese egions emains subop imal.
An analysis o e iciency ends e eals no able ad ancemen s in a eas like Hei-
longjiang, Beijing, and Sichuan. This implies ha he ini ial assessmen s may ha e un-
de alued he e iciency le els in hese egions. Mo eo e , i unde sco es he challenging
en i onmen al ci cums ances ha a e less conduci e o he g ow h o he pha maceu ical
sec o in hese locales. In con as , he egions ha expe ienced signi ican declines in
e iciency include Tibe , Ningxia, Fujian, and Qinghai. Mo eo e , all p o inces in he no h-
wes ha e seen a decline in e iciency, sugges ing ha he ex e nal en i onmen is cu en ly
mo e a o able o he de elopmen o he pha maceu ical indus y in o he egions.
As shown in he able, he a e age e iciency has shown a consis en upwa d end o e
he yea s, inc easing om 0.642 in 2013 o 0.767 and indica ing a con inuous imp o emen
in e iciency le els. Addi ionally, he p obabili y dis ibu ion o e iciency o each yea
also shows some imp o emen . The Gaussian dis ibu ion g aph based on he e iciency
dis ibu ion ac oss 31 p o inces each yea is shown in Figu e 2.
As shown in Figu e 2, he e was a no iceable change in he o e all e iciency dis ibu-
ion. In 2013, he e iciency alues we e concen a ed in he lowe ange, wi h a ela i ely
b oad dis ibu ion, indica ing low e iciency and signi ican dispa i ies. O e he yea s,
he cu e g adually shi ed o he igh , e lec ing an inc ease in e iciency le els, and he
dis ibu ion became mo e concen a ed, showing highe e iciency alues. By 2022, he
e iciency le el had signi ican ly imp o ed, wi h a s eepe dis ibu ion, indica ing educed
e iciency dispa i ies and a mo e balanced o e all e iciency. This change e lec s he g ad-
ual imp o emen in he e iciency o he pha maceu ical indus y and he na owing o
e iciency gaps ac oss di e en egions.
Meanwhile, he coe icien o a ia ion (CV, calcula ed as a e age e iciency/SD)
dec eased om 0.352 o 0.191, indica ing a educ ion in he dispa i y o e iciency le els.
The decline in he coe icien o a ia ion sugges s ha he e iciency le els ac oss di e en
egions a e becoming mo e balanced, which may e lec an inc ease in he ai ness o
esou ce alloca ion o as e e iciency imp o emen s in unde de eloped egions.
Economies 2025,13, 90 19 o 35
Figu e 2. Gaussian ke nel densi y dis ibu ion o TE.
This sec ion applies a h ee-s age DEA me hod combined wi h PCA o measu e he
ope a ional e iciency o he pha maceu ical indus y ac oss 31 p o inces in China in he
holis ic dimension and o e eal he impac o comp ehensi e ex e nal ac o s on ope a ional
e iciency. Howe e , he causes o hese di e ences a e mul i ace ed, and he cu en sec ion
will p o ide a de ailed discussion o he ac o s con ibu ing o hese dispa i ies.
5. Discussion
In he empi ical analysis o he p e ious chap e , i was ound ha he o e all e iciency
o China’s pha maceu ical indus y is g adually inc easing, wi h he gap be ween egions
na owing. Howe e , di e en egions exhibi dis inc ends. Addi ionally, he s udy
e ealed he speci ic impac o en i onmen al ac o s on e iciency. The ollowing sec ions
will u he explo e he unde lying causes o hese phenomena.
5.1. Discussion on he E iciency Dis ibu ion o he Pha maceu ical Indus y Ac oss
China’s Regions
The esul s p esen ed in he esul s sec ion e eal ha he e a e di e ences in e iciency
ac oss egions in bo h he i s and hi d s ages. To u he explo e he o e all di e ences
in he ope a ional e iciency o he PI ac oss egions and he unde lying causes, his s udy
decomposed TE in o PTE and SE. By using he a e age alues o TE and SE as h esholds,
he e iciency o each egion was classi ied in o ou quad an s: high PTE–high SE, high
PTE–low SE, low PTE–high SE, and low PTE–low SE. Figu e 3illus a es he ela i e
posi ions o he egions based on hei PTE and SE pe o mance.
(1)
E iciency analysis o pha maceu ical en e p ises in i s - ie egions.
Pha maceu ical en e p ises in i s - ie egions, including Tianjin, Liaoning, and
Jiangxi, exhibi high PTE and SE. These egions demons a e s ong esou ce al-
loca ion capabili ies and signi ican economies o scale, leading o highe o e all
e iciency. While he economic s eng h o hese egions is no he highes , hey e ec-
i ely le e age abundan human esou ces, low labo cos s, and s ong go e nmen
suppo o op imize esou ce u iliza ion. Th ough he syne gis ic e ec s o echnology,
Economies 2025,13, 90 20 o 35
esou ces, and policies, hese egions ha e de eloped hei unique high-e iciency
models. Addi ionally, he well-es ablished indus ial suppo ounda ion in hese
egions p omo es in e -indus y coo dina ion, which u he enhances bo h echnical
and scale e iciencies.
Figu e 3. A e age p e and se dis ibu ion o PI ac oss Chinese p o inces.
(2)
E iciency analysis o pha maceu ical en e p ises in second- ie egions.
Pha maceu ical en e p ises in second- ie egions also show high PTE bu ela i ely
low SE. These egions include Jiangsu, Shandong, Guangdong, and Ningxia. Excep
o Ningxia, he o he p o inces (Jiangsu, Shandong, Guangdong) a e among he
economically s onges in China. Acco ding o he SFA analysis, he high e iciency in
Ningxia is p ima ily due o a o able en i onmen al ac o s such as s ong na ional
suppo , a la ge en i onmen al capaci y, and he ela i ely low income and li ing
s anda ds o he local popula ion, which oge he c ea e a high le el o esou ce allo-
ca ion.
Howe e , en e p ises in Jiangsu, Shandong, and Guangdong, despi e hei s ong
pe o mance in echnical inno a ion and esou ce alloca ion, e icien ly u ilize ech-
nology and managemen me hods o achie e high echnical e iciency. These egions’
en e p ises possess ad anced p oduc ion echnologies and obus R&D capabili ies.
Howe e , despi e hei s eng hs in echnological e iciency, hey ha e s uggled o
achie e economies o scale du ing expansion, esul ing in a pe sis en si ua ion o
dec easing e u ns o scale (DRS) o e he pas decade. This phenomenon can be
a ibu ed o ac o s such as ma ke sa u a ion, wi h en e p ises inding i di icul
o ma ch inc eased p oduc ion capaci y wi h su icien ma ke demand, ine icien
esou ce alloca ion, despi e s ong echnological pe o mance, wi h issues in he
dis ibu ion o esou ces such as unds, alen , and equipmen , and managemen
bo lenecks, h ough which he o iginal managemen sys em is unable o adap o he
complexi y b ough abou ia scale expansion, he eby limi ing he imp o emen o
scale e iciency.

Economies 2025,13, 90 21 o 35
(3)
E iciency analysis o pha maceu ical en e p ises in hi d- ie egions.
Pha maceu ical en e p ises in hi d- ie egions show high SE bu ela i ely low PTE.
These egions include Beijing, Shanghai, Chongqing, Hunan, Henan, and Yunnan.
The po en ial causes o he low PTE in hese egions a y. Fo ins ance, in Chongqing,
Beijing, and Shanghai, al hough hese egions possess s ong economic and echno-
logical ounda ions, hey a e hinde ed by s ic en i onmen al and sa e y policies.
High en i onmen al p o ec ion equi emen s in places like Beijing and Shanghai ha e
es ic ed in es men in echnological upg ades and inno a ion, he eby limi ing im-
p o emen s in echnical e iciency. Addi ionally, high p oduc ion ac o cos s, such as
labo , land, and ene gy p ices, ha e become signi ican obs acles o enhancing echno-
logical pe o mance. Despi e hese cons ain s, en e p ises in hese egions can achie e
economies o scale h ough ma ke size ad an ages and indus ial agglome a ion
e ec s, which allow hem o lowe uni cos s h ough inc eased p oduc ion.
In con as , en e p ises in egions like Hunan and Yunnan, despi e acing challenges in
echnical e iciency, ha e pe o med well in scale e iciency. These egions’ en e p ises
ha e ela i ely limi ed in es men s in echnological inno a ion and R&D, which
esul s in s agna ion in echnological p og ess and a ec s hei p oduc ion e iciency.
Howe e , due o lowe p oduc ion ac o cos s and go e nmen policy suppo , such
as ax educ ions and indus ial subsidies, en e p ises in hese egions can achie e
economies o scale. The local go e nmen ’s policies u he enhance he e iciency o
en e p ises when expanding p oduc ion o mee he g owing ma ke demand, he eby
boos ing scale e iciency.
(4)
E iciency analysis o pha maceu ical en e p ises in ou h- ie egions.
Pha maceu ical en e p ises in ou h- ie egions, such as Gansu and Shanxi, gene ally
exhibi low TE and PTE. These egions’ en e p ises mainly ely on adi ional ech-
nologies and ela i ely ex ensi e managemen models, lacking ad anced echnical
suppo and p ocess op imiza ion. As a esul o insu icien echnological inno a ion,
he p oduc ion e iciency in hese egions is low. Mo eo e , he geog aphical loca ions
and economic ounda ions o Gansu and Shanxi make hem less a ac i e o highly
skilled alen and echnological capi al, p e en ing en e p ises om abso bing ex e nal
ad anced echnologies and managemen expe ience. The se e e sho age and ou low
o local alen ha e u he cons ained echnological imp o emen s. Addi ionally, he
weak economic ounda ion and limi ed go e nmen suppo o echnological inno-
a ion and esea ch es ic he abili y o en e p ises o in es in necessa y upg ades,
leading o a lack o imp o emen in echnical and pu e echnical e iciency.
5.2. Discussion on he Changes in Regional E iciency and Thei Causes
O e he pas decade, he o e all e iciency o he PI has shown an upwa d end.
Howe e , i is impo an o no e ha , in some egions, e iciency has declined yea by yea .
The e o e, his s udy di ides he 31 egions o China in o se en la ge a eas and decomposes
he TE o hese a eas in o PTE and SE. PTE e lec s he e iciency imp o emen s achie ed
h ough echnological inno a ion o esou ce alloca ion op imiza ion in a egion, excluding
he impac o economies o scale, while SE measu es whe he a egion can e ec i ely
achie e economies o scale as p oduc ion scale inc eases. Figu e 4a–c show he changes in
pha maceu ical-indus y e iciency in he se en la ge egions o China o e he decade.
Economies 2025,13, 90 22 o 35
(a) T ends in TE o he PI in di e en egions o China.
(b) T ends in PTE o he PI in di e en egions o
China.
(c) T ends in SE o he PI in di e en egions o China.
Figu e 4. Decomposi ion o he ope a ion e iciency in S age 3.
As shown in Figu e 4a, he o e all e iciency o China’s PI has s eadily inc eased o e
he pas decade, wi h mos egions expe iencing ela i ely s able changes and minimal
luc ua ions. This indica es ha he o e all de elopmen end o he PI is s able, and
e iciency is g adually becoming balanced. Howe e , ce ain egions ha e shown pe sis en
inc eases o dec eases in e iciency. This sugges s ha , al hough he o e all end is posi i e,
some a eas s ill ace di e en challenges and oppo uni ies, necessi a ing a mo e in-dep h
analysis o he unde lying causes.
The ope a ion e iciency o Cen al China (including Henan, Hubei, and Hunan)
expe ienced a no iceable decline om 2013 o 2022. A decomposi ion o TE e eals ha he
p ima y eason o his decline is low PTE. The possible causes include ine icien esou ce
alloca ion and insu icien inno a ion ou pu in hese egions, which ha e cons ained
he imp o emen o PTE. Speci ically, he lack o inno a ion capaci y in he cen al egion
(including Henan, Hubei, and Hunan) has esul ed in poo pe o mance in echnological
ad ancemen and p oduc ion e iciency op imiza ion, he eby limi ing he enhancemen
o PTE. These h ee p o inces lag behind in echnological inno a ion, po en ially due o
insu icien go e nmen emphasis on pha maceu ical esea ch and de elopmen (R&D),
Economies 2025,13, 90 23 o 35
limi ed in es men in R&D, and a con inued eliance on adi ional p oduc ion me hods.
The lack o in oduc ion and applica ion o new echnologies has u he cons ained
he imp o emen o p oduc ion e iciency. Mo eo e , he pha maceu ical indus y in he
cen al egion emains p edominan ly ocused on manu ac u ing, leading o slow p og ess
in echnological upg ading and indus ial ans o ma ion. The issue o b ain d ain is also
p ominen in hese p o inces, wi h a sho age o high-end echnical and inno a i e alen
u he es ic ing he de elopmen o echnological inno a ion. Inadequa e policy suppo
and inancial in es men om he go e nmen ha e weakened he mo i a ion o en e p ises
o engage in R&D and echnological ans o ma ion, making i di icul o es ablish an
e ec i e inno a ion ecosys em. The e o e, he cen al egion aces mul iple challenges in
imp o ing ope a ional e iciency and echnical e iciency. I is impe a i e o add ess hese
issues by inc easing R&D in es men , op imizing esou ce alloca ion, p omo ing indus ial
upg ading, and a ac ing high-end alen o imp o e he cu en si ua ion.
In con as , he TE in Sou he n and No hwes e n China has signi ican ly imp o ed
om 2013 o 2022. Gi en he economic cha ac e is ics and scale di e ences be ween hese
wo egions, u he decomposi ion o he TE in o PTE and SE e eals di e en unde lying
causes o he imp o emen . As is shown in Figu e 4b, he inc ease in TE in Sou he n
China is p ima ily due o imp o emen s in PTE, while he e iciency imp o emen in
No hwes e n China is mainly a ibu ed o SE. This sugges s ha he d i ing o ces behind
g ow h in hese egions a e dis inc , and a ge ed s a egies a e needed o p omo e hei
con inued de elopmen .
The annual inc ease in TE in Sou he n China is p ima ily due o imp o emen s in pu e
echnical e iciency. Sou he n China, being one o he mos economically de eloped egions
in he coun y, has seen subs an ial suppo om bo h he go e nmen and en e p ises
o high- ech indus ies, pa icula ly in sec o s such as pha maceu icals, elec onics, and
ad anced manu ac u ing, wi h esea ch and de elopmen in es men s inc easing annu-
ally. Addi ionally, he sou he n egion bene i s om well-de eloped in as uc u e and
a s a egic geog aphical loca ion, a ac ing a concen a ion o high-end echnical alen
and inno a i e en e p ises and he eby u he op imizing esou ce alloca ion e iciency.
Imp o emen s in managemen p ac ices and p oduc ion o ganiza ion me hods ha e also
con ibu ed o highe p oduc ion e iciency. These ac o s collec i ely unde pin he egion’s
ou s anding pe o mance in PTE, d i ing he o e all imp o emen in echnical e iciency.
Mo ing o wa d, i he sou he n egion con inues o p io i ize echnological inno a ion
and alen acquisi ion while u he op imizing indus ial s uc u e and esou ce alloca ion,
i s echnical e iciency is expec ed o achie e e en g ea e p og ess.
The imp o emen in TE in No hwes e n China is mainly d i en by SE. The likely
cause is ha PI in his egion has expanded p oduc ion scale, achie ing economies o scale,
which, in u n, has enhanced TE. An analysis o e u ns o scale (RTS) e eals ha , o e he
pas decade, in he o al 50 samples om he i e p o inces in No hwes e n China, only
one yea exhibi ed dec easing e u ns o scale (DRS), while he emaining yea s we e unde
inc easing e u ns o scale (IRS). This indica es ha inc easing scale in es men s in he
egion can u he imp o e e iciency. Wi h he ans e o he PI o No hwes e n China,
SE in his egion has imp o ed, leading o an inc ease in TE. This indus ial eloca ion is
p ima ily a ibu ed o na ional policy suppo , ela i ely lenien en i onmen al egula ions,
and abundan en i onmen al capaci y. In ecen yea s, he Chinese go e nmen has imple-
men ed s a egies such as he “Wes e n De elopmen ” and he “Bel and Road” ini ia i es,
p o iding ax incen i es, iscal subsidies, and in as uc u e suppo o he no hwes e n e-
gion, he eby a ac ing a signi ican numbe o pha maceu ical en e p ises. Simul aneously,
he ela i ely elaxed en i onmen al s anda ds in he no hwes ha e educed ope a ional
cos s o en e p ises, while i s as land, spa se popula ion, and abundan na u al esou ces
Economies 2025,13, 90 24 o 35
ha e p o ided a o able condi ions o la ge-scale p oduc ion. These ac o s collec i ely
con ibu ed o he imp o emen o scale e iciency and, h ough economies o scale, p o-
mo ed he enhancemen o echnical e iciency, injec ing new momen um in o egional
economic de elopmen . In he u u e, i he no hwes e n egion can u he s eng hen
in as uc u e cons uc ion, op imize he business en i onmen , and enhance echnological
inno a ion capabili ies, he de elopmen po en ial o i s pha maceu ical indus y will
become e en mo e signi ican .
5.3. The Discussion o Regional Dispa i ies in China’s Pha maceu ical Indus y
To u he discuss he egional dispa i ies o PI ac oss di e en egions in China, his
s udy employed he coe icien o a ia ion (CV) as an indica o . A la ge CV indica es
egional dispa i ies be ween egions, sugges ing an une en dis ibu ion o esou ce alloca-
ion and ope a ional e iciency. In con as , a smalle CV e lec s mo e balanced ope a ional
e iciency, implying ela i ely a ional esou ce dis ibu ion. Figu e 5p esen s he CV end
o TE in PI ac oss 31 egions om 2013 o 2022, gi ing deepe insigh s in o de elopmen al
dispa i ies be ween egions.
As shown in Figu e 5, bo h in S age 1 and S age 3, despi e a sligh inc ease in he
coe icien o a ia ion (CV) be ween 2017 and 2019, he o e all end shows a signi ican
decline. This sugges s ha he egional dispa i ies in China’s pha maceu ical indus y a e
g adually na owing.
Figu e 5. T end o CV changes in China’s PI om 2013 o 2022.
The po en ial easons o his include go e nmen mac o-con ol measu es and he
changes in scale e iciency b ough abou ia indus ial ans e . Fi s ly, he go e nmen has
implemen ed s ingen sa e y, en i onmen al, and ax policies in de eloped egions, which
ha e placed signi ican p essu e on he pha maceu ical indus y, pa icula ly on pollu ion-
in ensi e sec o s such as ac i e pha maceu ical ing edien s (APIs) and pha maceu ical
in e media es. This has p omp ed hese indus ies o g adually shi o unde de eloped
egions, con ibu ing o a mo e balanced egional dis ibu ion. Addi ionally, pha maceu-
ical companies end o choose egions wi h ewe egula ions, ax incen i es, and la ge
en i onmen al capaci ies o hei de elopmen .
Howe e , he inc eased dispa i y be ween 2017 and 2019 could be a ibu ed o majo
policy adjus men s, such as cen alized p ocu emen and a consis ency e alua ion o he
gene ic d ug in PI. Indus ies such as APIs and in e media es, which could no pa icipa e
in cen alized p ocu emen and consis ency e alua ion, expe ienced a decline in o e all
ou pu , he eby inc easing egional dispa i ies.
Economies 2025,13, 90 31 o 35
Da a A ailabili y S a emen : Da a in his s udy is a ailable in h ps://doi.o g/10.5281/zenodo.147
55469.
Acknowledgmen s: The au ho s hank he edi o ial eam and anonymous e iewe s o hei insigh -
ul commen s on he a icle.
Con lic s o In e es : Au ho Jiaqiang Sun was employed by he company Chengdu Maidison
Pha maceu ical Technology Co., L d. The emaining au ho s decla e ha he esea ch was conduc ed
in he absence o any comme cial o inancial ela ionships ha could be cons ued as a po en ial
con lic o in e es .
No e
1
No e: This s udy’s da a sou ces do no include Hong Kong, Macau, o Taiwan, as hei s a is ical da a a e p ocessed sepa a ely in
o icial Chinese epo s.
Re e ences
Ahmed, M. H., & Melesse, K. A. (2018). Impac o o - a m ac i i ies on echnical e iciency: E idence om maize p oduce s o eas e n
E hiopia. Ag icul u al and Food Economics,6(1), 1–15. [C ossRe ]
Aigne , D., Lo ell, C. K., & Schmid , P. (1977). Fo mula ion and es ima ion o s ochas ic on ie p oduc ion unc ion models. Jou nal o
Econome ics,6(1), 21–37. [C ossRe ]
And é, F. J., Buendía, A., & San os-A eaga, F. J. (2024). E icien wa e use and eusing p ocesses ac oss Spanish egions: A ci cula
da a en elopmen analysis wi h undesi able inpu s. Jou nal o Cleane P oduc ion,434, 139929. [C ossRe ]
A ki an, N. K., & Rowlands, T. (2008). How o be e iden i y he ue manage ial pe o mance: S a e o he a using DEA. Omega,
36(2), 317–324. [C ossRe ]
Bai, H., I an, M., & Hao, Y. (2022). How does indus ial ans e a ec en i onmen al quali y? E idence om China. Jou nal o Asian
Economics,82, 101530. [C ossRe ]
Bai d, G. L., & Biebe , S. L. (2016). The Goldilocks dilemma: Impac s o mul icollinea i y—A compa ison o simple linea eg ession,
mul iple eg ession, and o de ed a iable eg ession models. Jou nal o Mode n Applied S a is ical Me hods,15(1), 18. [C ossRe ]
Banke , R. D., Cha nes, A., & Coope , W. W. (1984). Some models o es ima ing echnical and scale ine iciencies in da a en elopmen
analysis. Managemen Science,30(9), 1078–1092. [C ossRe ]
Ba ney, J. (1991). Fi m esou ces and sus ained compe i i e ad an age. Jou nal o Managemen ,17(1), 99–120. [C ossRe ]
Ba ney, J. B., & A ikan, A. M. (2005). The esou ce-based iew: O igins and implica ions. In The Blackwell handbook o s a egic
managemen (pp. 123–182). Blackwell Publishe s L d.
Bha dwaj, R. (2024). Economics o he pha maceu ical and medical de ice indus y: Supply chain, ade and inno a ion. Taylo & F ancis.
[C ossRe ]
Bo ja Reis, C. F. d., & Pin o, J. P. G. (2022). Cen e –pe iphe y ela ionships o pha maceu ical alue chains: A c i ical analysis based on
goods and knowledge ade lows. Re iew o Poli ical Economy,34(1), 124–145. [C ossRe ]
Cai, L., & Sun, Y. (2013). Resea ch on he e iciency o biopha maceu ical en e p ises based on DEA and SFA. Science Technology and
Managemen Resea ch,33(2), 24.
Camanho, A. S., Sil a, M. C., Pi an, F. S., & Lace da, D. P. (2024). A li e a u e e iew o economic e iciency assessmen s using Da a
En elopmen Analysis. Eu opean Jou nal o Ope a ional Resea ch,315(1), 1–18. [C ossRe ]
Cha nes, A., Coope , W. W., & Rhodes, E. (1978). Measu ing he e iciency o decision making uni s. Eu opean Jou nal o Ope a ional
Resea ch,2(6), 429–444. [C ossRe ]
Cha u edi, P., Shukla, P., Gi i, B. S., Chowdha y, P., Chand a, R., Gup a, P., & Pandey, A. (2021). P e alence and haza dous impac o
pha maceu ical and pe sonal ca e p oduc s and an ibio ics in en i onmen : A e iew on eme ging con aminan s. En i onmen al
Resea ch,194, 110664. [C ossRe ] [PubMed]
Chen, G., & Chen, F. (2024). Changes in echnological inno a ion e iciency and in luencing ac o s o lis ed ex ile and appa el
companies esea ch—Based on h ee-s age DEA wi h Tobi modeling. PLoS ONE,19(8), e0307820. [C ossRe ]
Chen, P., Ding, X., Chen, M., Song, H., & Im an, M. (2025). The impac o esou ce spa ial misma ch on he con igu a ion analysis o
ag icul u al g een o al ac o p oduc i i y. Ag icul u e,15(1), 23. [C ossRe ]
Coelli, T. J., Rao, D. S. P., O’donnell, C. J., & Ba ese, G. E. (2005). An in oduc ion o e iciency and p oduc i i y analysis. Sp inge Science
& Business Media. [C ossRe ]
Colm, G. (1960). The heo y o public inance: A s udy in public economy. JSTOR.
Coope , W. W., Sei o d, L. M., & Tone, K. (2007). Da a en elopmen analysis: A comp ehensi e ex wi h models, applica ions, e e ences and
dea-sol e so wa e (Vol. 2). Sp inge . [C ossRe ]

Economies 2025,13, 90 32 o 35
Deo e, A. B., Dhumane, J. R., Wagh, R., & Sonawane, R. (2019). The s ages o d ug disco e y and de elopmen p ocess. Asian Jou nal o
Pha maceu ical Resea ch and De elopmen ,7(6), 62–67. [C ossRe ]
Dou, J., & Han, X. (2019). How does he indus y mobili y a ec pollu ion indus y ans e in China: Empi ical es on Pollu ion
Ha en Hypo hesis and Po e Hypo hesis. Jou nal o Cleane P oduc ion,217, 105–115. [C ossRe ]
Dyson, R. G., Allen, R., Camanho, A. S., Podino ski, V. V., Sa ico, C. S., & Shale, E. A. (2001). Pi alls and p o ocols in DEA. Eu opean
Jou nal o Ope a ional Resea ch,132(2), 245–259. [C ossRe ]
Fa ell, M. J. (1957). The measu emen o p oduc i e e iciency. Jou nal o he Royal S a is ical Socie y Se ies A: S a is ics in Socie y,120(3),
253–281.
Feng, M., & Li, X. (2020). E alua ing he e iciency o indus ial en i onmen al egula ion in China: A h ee-s age da a en elopmen
analysis app oach. Jou nal o Cleane P oduc ion,242, 118535. [C ossRe ]
F ied, H. O., Lo ell, C. K., Schmid , S. S., & Yaisawa ng, S. (2002). Accoun ing o en i onmen al e ec s and s a is ical noise in da a
en elopmen analysis. Jou nal o P oduc i i y Analysis,17, 157–174. [C ossRe ]
F iedman, M. (2007). The social esponsibili y o business is o inc ease i s p o i s. In Co po a e e hics and co po a e go e nance (pp.
173–178). Sp inge .
Gascón, F., Lozano, J., Pon e, B., & de la Fuen e, D. (2017). Measu ing he e iciency o la ge pha maceu ical companies: An indus y
analysis. The Eu opean Jou nal o Heal h Economics,18, 587–608. [C ossRe ]
Ge, X., Im an, M., & Ali, K. (2024). Na u al esou ce-d i en p ospe i y: Un eiling he ca alys s o sus ainable economic de elopmen
in he Uni ed S a es. In Na u al esou ces o um. Blackwell Publishing L d.
G eenac e, M., G oenen, P. J., Has ie, T., d’Enza, A. I., Ma kos, A., & Tuzhilina, E. (2022). P incipal componen analysis. Na u e Re iews
Me hods P ime s,2(1), 100. [C ossRe ]
G eene, W. H. (2008). The econome ic app oach o e iciency analysis. The Measu emen o P oduc i e E iciency and P oduc i i y G ow h,
1(1), 92–250.
G ossman, S. J., & S igli z, J. E. (1977). On alue maximiza ion and al e na i e objec i es o he i m. The Jou nal o Finance,32(2),
389–402.
Guanglan, Z., & Zhening, Z. (2024). E alua ion and in luencing ac o s o egional en i onmen al e iciency in China based on
h ee-s age DEA model. SAGE Open,14(4), 21582440241280825. [C ossRe ]
Guo, C., Song, Q., Yu, M.-M., & Zhang, J. (2024). A digi al economy de elopmen index based on an imp o ed hie a chical da a
en elopmen analysis app oach. Eu opean Jou nal o Ope a ional Resea ch,316(3), 1146–1157. [C ossRe ]
Guo, H., & Shi, K. (2021). E alua ion o he in e na ional compe i i eness o China’s pha maceu ical indus y and high-quali y
de elopmen s a egies du ing he 14 h Fi e-Yea Plan. Jou nal o Beijing Uni e si y o Technology (Social Sciences Edi ion),21(3),
65–79. [C ossRe ]
Guo, Y., Yu, M.-M., & See, K. F. (2024). De eloping a sus ainable de elopmen goals index o OECD coun ies: An e ec i eness-based
hie a chical da a en elopmen analysis. En i onmen al Science & Policy,160, 103836. [C ossRe ]
Hai , J., Black, W., Babin, B., & Ande son, R. (2010). Mul i a ia e da a analysis (7 h ed.). P en ice Hall.
Hai o sky, Y. (1969). Mul icollinea i y in eg ession analysis: Commen . The Re iew o Economics and S a is ics,51, 486–489.
Hamad, M., & Ta noczi, T. (2021). E iciency analysis o companies ope a ing in he pha maceu ical indus y in he Viseg ad coun ies.
In ellec ual Economics,15(2), 131–155. [C ossRe ]
Hao, B., & Ruan, X. (2022). Resea ch on he echnological inno a ion e iciency o Chinese lis ed biopha maceu ical en e p ises based
on he wo-s age DEA model. China Pha macy,12, 7–12. [C ossRe ]
Hjalma sson, L., Kumbhaka , S. C., & Heshma i, A. (1996). DEA, DFA and SFA: A compa ison. Jou nal o P oduc i i y Analysis,7,
303–327. [C ossRe ]
Im an, M., Alam, M. S., Jijian, Z., Oz u k, I., Wahab, S., & Do˘gan, M. (2024a). F om esou ce cu se o g een g ow h: Explo ing he ole
o ene gy u iliza ion and na u al esou ce abundance in economic de elopmen . In Na u al esou ces o um. Blackwell Publishing
L d. [C ossRe ]
Im an, M., & Jijian, Z. (2023). Explo ing he ela ionship be ween inancial inclusion and na u al esou ce u iliza ion in QUAD
economies. En i onmen al Science and Pollu ion Resea ch,30(58), 122958–122971. [C ossRe ] [PubMed]
Im an, M., Khan, M. K., Alam, S., Wahab, S., Tu ail, M., & Jijian, Z. (2024b). The implica ions o he ecological oo p in and enewable
ene gy usage on he inancial s abili y o Sou h Asian coun ies. Financial Inno a ion,10(1), 102. [C ossRe ]
Im an, M., Khan, M. K., Wahab, S., Ahmed, B., & Jijian, Z. (2025a). The pa adox o esou ce- ichness: Un a eling he e ec s on inancial
ma ke s in na u al esou ce abundan economies. Financial Inno a ion,11(1), 63. [C ossRe ]
Im an, M., Sa a , A., & Alam, M. S. (2024c). He e ogeneous analysis o ee ade ag eemen be ween Pakis an and China: A policy
guideline o CPEC. Jou nal o Economic and Adminis a i e Sciences,40(1), 76–94. [C ossRe ]
Im an, M., Tu ail, M., Mo, C., Wahab, S., Khan, M. K., Hoo, W. C., & Ling, Z. (2025b). F om esou ces o esilience: Unde s anding
he impac o s anda d o li ing and ene gy consump ion on na u al esou ce en in Asia. Ene gy S a egy Re iews,57, 101590.
[C ossRe ]
Economies 2025,13, 90 33 o 35
Im an, M., Zhang, G., & An, H. (2017). Impac o ma ke access and compa a i e ad an age on egional dis ibu ion o manu ac u ing
sec o . China Finance and Economic Re iew,5, 1–14. [C ossRe ]
Jia, F. (2022). Resea ch on he in e na ional compe i i eness o China’s pha maceu ical indus y om he pe spec i e o he global alue chain
[Unpublished Mas e ’s hesis, Kunming Uni e si y o Science and Technology].
Joe, Z., & Wu, G. (1996). Dea heo y, me hods and applica ions. Science P ess.
Kali ajan, K. P., & Shand, R. (1994). S ochas ic on ie p oduc ion unc ions and echnical e iciency measu emen s: A e iew. Palg a e
Macmillan UK. [C ossRe ]
Kau , N., & Singh, B. (2024). Dynamics o esou ce alloca ion– i m pe o mance ela ionship: Panel au o eg essi e dis ibu ed lag
app oach. Manage ial and Decision Economics,45(6), 3663–3676.
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 ]
Kumbhaka , S. C., & Tsionas, E. G. (2011). Some ecen de elopmen s in e iciency measu emen in s ochas ic on ie models. Jou nal
o P obabili y and S a is ics,2011(1), 603512. [C ossRe ]
Lai, H., Shi, H., & Zhou, Y. (2020). Regional echnology gap and inno a ion e iciency ap in Chinese pha maceu ical manu ac u ing
indus y. PLoS ONE,15(5), e0233093. [C ossRe ] [PubMed]
Lampe, H. W., & Hilge s, D. (2015). T ajec o ies o e iciency measu emen : A bibliome ic analysis o DEA and SFA. Eu opean Jou nal
o Ope a ional Resea ch,240(1), 1–21. [C ossRe ]
Lee, G. H., & Jeon, H. W. (2023). A s ochas ic on ie analysis (SFA)-based me hod o de ec ing changes in manu ac u ing ene gy
e iciency by sec o and ime. In IFIP in e na ional con e ence on ad ances in p oduc ion managemen sys ems (pp. 29–42). Sp inge .
Li, S., & Fan, C. (2009). A e iew and compa ison o SFA and DEA me hods. S a is ics and Decision,7(283), 25–28.
Lin, T.-X., Wu, Z.-h., Ji, X.-x., & Yang, J.-j. (2021). Resea ch on he ope a ing e iciency o Chinese lis ed pha maceu ical companies
based on wo-s age ne wo k DEA and Malmquis . Ma hema ical P oblems in Enginee ing,2021(1), 1475781. [C ossRe ]
Liu, T., Pan, S., Hou, H., & Xu, H. (2020). Analyzing he en i onmen al and economic impac o indus ial ans e based on an
imp o ed CGE model: Taking he Beijing–Tianjin–Hebei egion as an example. En i onmen al Impac Assessmen Re iew,83,
106386. [C ossRe ]
Liu, Y., Zhang, N., Xie, C., Jiang, Y., Qin, Y., Zhou, L., Fan, Y., Ren, L., Yin, C., Yang, H., Xie, W., Zhai, Q., Li, G., Chen, H., & Chen, X.
(2022). E olu ion o d ug egula ions and egula o y inno a ion o an icance d ugs in China. Ac a Pha maceu ica Sinica B,12(12),
4365–4377. [C ossRe ]
Luo, D. (2012). A no e on es ima ing manage ial ine iciency o h ee-s age DEA model. S a is ical Resea ch,29(4), 104–107.
Ma, Z. X. (2010). Da a en elopmen analysis models and me hods. Science P ess.
Madabushi, R., Seo, P., Zhao, L., Tegenge, M., & Zhu, H. (2022). Role o model-in o med d ug de elopmen app oaches in he li ecycle
o d ug de elopmen and egula o y decision-making. Pha maceu ical Resea ch,39(8), 1669–1680. [C ossRe ]
Madaleno, M., & Mou inho, V. (2023). S ochas ic on ie analysis: A e iew and syn hesis. Ad anced Ma hema ical Me hods o Economic
E iciency Analysis,692, 55–78. [C ossRe ]
Ma in, R., & Sunley, P. (2003). Decons uc ing clus e s: Chao ic concep o policy panacea? Jou nal o Economic Geog aphy,3(1), 5–35.
[C ossRe ]
Me goni, A., Em ouznejad, A., & De Wi e, K. (2024). Fi y yea s o da a en elopmen analysis. Eu opean Jou nal o Ope a ional Resea ch.
[C ossRe ]
Mohs, R. C., & G eig, N. H. (2017). D ug disco e y and de elopmen : Role o basic biological esea ch. Alzh Demen -T ci,3(4), 651–657.
[C ossRe ]
Moulay Ali, H., Guellil, M. S., Mokh a i, F., & Tsabe , A. (2024). The e ec o subsidies on ecnical e iciency o Alge ian ag icul u al
sec o : Using s ochas ic on ie model (SFA). Disco e Sus ainabili y,5(1), 98. [C ossRe ]
Na, H., Du, T., Sun, W., He, J., Sun, J., Yuan, Y., & Qiu, Z. (2019). Re iew o e alua ion me hodologies and in luencing ac o s o
ene gy e iciency o he i on and s eel indus y. In e na ional Jou nal o Ene gy Resea ch,43(11), 5659–5677. [C ossRe ]
Na ional Bu eau o S a is ics o China. (2023). China s a is ical yea book 2023. China S a is ics P ess. A ailable online: h ps://
www.s a s.go .cn (accessed on 27 Oc obe 2024).
Okeke, E. S., Ezeo ba, T. P. C., Okoye, C. O., Chen, Y., Mao, G., Feng, W., & Wu, X. (2022). En i onmen al and heal h impac o
un eco e ed API om pha maceu ical manu ac u ing was es: A e iew o con empo a y ea men , ecycling and managemen
s a egies. Sus ainable Chemis y and Pha macy,30, 100865. [C ossRe ]
Ðoki´c, D., No ako i´c, T., Teki´c, D., Ma ko ski, B., Zeki´c, S., & Mili´c, D. (2022). Technical e iciency o ag icul u e in he Eu opean
Union and Wes e n Balkans: SFA me hod. Ag icul u e,12(12), 1992. [C ossRe ]
Po e , M. E. (1990a). New global s a egies o compe i i e ad an age. Planning Re iew,18(3), 4–14. [C ossRe ]
Po e , M. E. (1990b). The compe i i e ad an age o na ions. Macmillan P ess LTD.
Po e , M. E. (1998). Clus e s and he new economics o compe i ion (Vol. 76, No. 6). Ha a d Business Re iew Bos on.
Qin, Y., Zhang, P., Deng, X., & Liao, G. (2023). Inno a ion e iciency e alua ion o indus ial echnology esea ch ins i u e based on
h ee-s age DEA. Expe Sys ems wi h Applica ions,224, 120004. [C ossRe ]
Economies 2025,13, 90 34 o 35
Qiu, L., Yu, R., Hu, F., Zhou, H., & Hu, H. (2023). How can China’s medical manu ac u ing lis ed i ms imp o e hei echnological
inno a ion e iciency? An analysis based on a h ee-s age DEA model and co po a e go e nance con igu a ions. Technological
Fo ecas ing and Social Change,194, 122684. [C ossRe ]
Rashid, A. A., See, K. F., & Yu, M.-M. (2024). Measu ing ai line e iciency using a dynamic ne wo k da a en elopmen analysis in he
p esence o inno a ion capi al. Technological Fo ecas ing and Social Change,206, 123457. [C ossRe ]
Res i, A. (2000). E iciency measu emen o mul i-p oduc indus ies: A compa ison o classic and ecen echniques based on
simula ed da a. Eu opean Jou nal o Ope a ional Resea ch,121(3), 559–578. [C ossRe ]
Riaz, M., Kazmi, S. M. A., Iqbal, M. S., & Hussain, A. (2023). E iciency analysis o pha maceu ical companies in Pakis an: A case s udy
o en amous companies. Pakis an Jou nal o Humani ies and Social Sciences,11(3), 3823–3830. [C ossRe ]
Schuhmache , A., Hinde , M., on S egmann und S ein, A., Ha l, D., & Gassmann, O. (2023). Analysis o pha ma R&D p oduc i i y—A
new pe spec i e needed. D ug Disco e y Today,28(10), 103726. [C ossRe ]
Schuhmache , A., Wilisch, L., Kuss, M., Kandelbaue , A., Hinde , M., & Gassmann, O. (2021). R&D e iciency o leading pha maceu ical
companies—A 20-yea analysis. D ug Disco e y Today,26(8), 1784–1789. [C ossRe ]
Shi, Y., Cha les, V., & Zhu, J. (2025). Bank inancial sus ainabili y e alua ion: Da a en elopmen analysis wi h andom o es and
Shapley addi i e explana ions. Eu opean Jou nal o Ope a ional Resea ch,321(2), 614–630. [C ossRe ]
Shin, K., Lee, D., Shin, K., & Kim, E. (2018). Measu ing he e iciency o US pha maceu ical companies based on open inno a ion ypes.
Jou nal o Open Inno a ion: Technology, Ma ke , and Complexi y,4(3), 34. [C ossRe ]
Sh es ha, N. (2020). De ec ing mul icollinea i y in eg ession analysis. Ame ican Jou nal o Applied Ma hema ics and S a is ics,8(2), 39–42.
[C ossRe ]
Singh, P. K., Bha , R. K., & Ki an, R. (2020). In es iga ing alloca i e e iciency, cos e iciency, echnical e iciency using s ochas ic
on ie analysis (SFA) o g oundnu a m households: E idence om a case in India. Cus os e Ag onegocio on Line,16(4), 105–129.
Solanki, N., Ve ma, J., Kuma , S., Seema, Mundlia, J., Saini, R., & Saini, S. (2024). Impac o con ac esea ch o ganiza ions on
pha maceu ical indus ies: A e iew. Applied D ug Resea ch, Clinical T ials and Regula o y A ai s, e26673371332371. [C ossRe ]
Song, H., & Ma, Y. (2024). Measu emen and spa ial and empo al cha ac e iza ion o logis ics e iciency–Based on a h ee-s age DEA
model. Heliyon,10(19), e38455. [C ossRe ] [PubMed]
S e i´c, Ž., Miški´c, S., Vojino i´c, D., Huskano i´c, E., S anko i´c, M., & Pamuˇca , D. (2022). De elopmen o a model o e alua ing he
e iciency o anspo companies: PCA–DEA–MCDM model. Axioms,11(3), 140. [C ossRe ]
Sun, J., Rosli, A. B., & Daud, A. (2024). Ope a ional e iciency o pha maceu ical companies in China: Based on h ee-s age DEA wi h
undesi able ou pu s. Sus ainabili y,17(1), 207. [C ossRe ]
Tang, Y., & Im an, M. (2024). T ade-o s and syne gies: Examining he impac o na u al esou ce en s and ene gy e iciency on
inancial de elopmen in he RCEP con ex . In Na u al esou ces o um. Blackwell Publishing L d. [C ossRe ]
Tang, Y., Yin, M., Yang, W., Li, H., Zhong, Y., Mo, L., Liang, Y., Ma, X., & Sun, X. (2019). Eme ging pollu an s in wa e en i onmen :
Occu ence, moni o ing, a e, and isk assessmen . Wa e En i onmen Resea ch,91(10), 984–991. [C ossRe ]
Teece, D. J., Pisano, G., & Shuen, A. (1997). Dynamic capabili ies and s a egic managemen . S a egic Managemen Jou nal, 18(7), 509–533.
Tone, K. (2001). A slacks-based measu e o e iciency in da a en elopmen analysis. Eu opean Jou nal o Ope a ional Resea ch,130(3),
498–509. [C ossRe ]
Wang, G., & Zhang, S. (2023). E alua ion o he in e na ional compe i i eness and coun y dependence o China’s pha maceu ical
manu ac u ing indus y based on he global alue chain. Science Technology and Indus y,23(13), 16–22.
Wang, J., Han, D., & Wang, Y. (2020). Empi ical esea ch on inno a ion e iciency in China based on SFA model. In Iop con e ence se ies:
Ea h and en i onmen al science (Vol. 474, p. 072055). IOP.
Wang, R., & Duan, Y. (2023). Dynamic compa ison on he echnical e iciency be ween China’s EPEs and PEs: Two-dimensional
measu emen based on SFA. Jou nal o Cleane P oduc ion,406, 136986. [C ossRe ]
Wang, W., Im an, M., Ali, K., & Sa a , A. (2024). G een policies and inancial de elopmen in G7 economies: An in-dep h analysis o
en i onmen al egula ions and g een economic g ow h. In Na u al esou ces o um. Blackwell Publishing L d. [C ossRe ]
Wang, Z., Li, Y., Wang, K., & Huang, Z. (2017). En i onmen -adjus ed ope a ional pe o mance e alua ion o sola pho o ol aic powe
plan s: A h ee s age e iciency analysis. Renewable and Sus ainable Ene gy Re iews,76, 1153–1162. [C ossRe ]
Wang, Z., Zhou, Y., Wang, T., & Zhao, N. (2024). E iciency o cons uc ion was e and ca bon educ ion in he cons uc ion indus y:
Based on imp o ed h ee s age SBM-DEA model in China. Enginee ing, Cons uc ion and A chi ec u al Managemen . [C ossRe ]
Wei, X., & Zhao, R. (2024). E alua ion and spa ial con e gence o ca bon emission educ ion e iciency in China’s powe indus y:
Based on a h ee-s age DEA model wi h game c oss-e iciency. Science o he To al En i onmen ,906, 167851. [C ossRe ] [PubMed]
Wu, B., Li, J., Yao, Z., Li, X., Wang, W., Wu, Z., & Zhou, Q. (2023). Cha ac e is ics and educ ion assessmen o GHG emissions om
c op esidue open bu ning in China unde he a ge s o ca bon peak and ca bon neu ali y. Science o he To al En i onmen ,905,
167235. [C ossRe ]
Xia, F., Cui, Y. Y., Liu, J. P., & He, Y. F. (2022). Analysis o ope a ional e iciency o lis ed pha maceu ical companies in China om 2013
o 2019. Chinese Pha maceu ical Jou nal,31(11), 1–5.
Economies 2025,13, 90 35 o 35
Xiong, A., & Meng, G. (2019). Resea ch on he echnological inno a ion e iciency o pha maceu ical en e p ises based on DEA me hod:
A case s udy o he op 15 lis ed companies in Shenzhen and Shanghai. Chinese Jou nal o New D ugs, 6. [C ossRe ]
Xu, T., Kang, C., & Zhang, H. (2022). China’s e o s owa ds ca bon neu ali y: Does ene gy-sa ing and emission- educ ion policy
mi iga e ca bon emissions? Jou nal o En i onmen al Managemen ,316, 115286. [C ossRe ]
Yan, L. (2012). Empi ical s udy on he egional inno a ion e iciency e alua ion index sys em in China. Managemen Wo ld, 174–175.
[C ossRe ]
Yang, R. (2024). Ope a ional e iciency analysis o lis ed biomedical companies in China: Th ee-s age DEA and Malmquis index.
Ad ances in Economics and Managemen Resea ch,10(1), 264.
Zhang, L., & Cui, J. (2024). Resea ch on o al ac o ene gy e iciency in wes e n China based on he h ee-s age DEA-Tobi model.
PLoS ONE,19(4), e0294329. [C ossRe ]
Zhang, M., Im an, M., & Juana as, R. A. (2024). Inno a e, conse e, g ow: A comp ehensi e analysis o echnological inno a ion,
ene gy u iliza ion, and ca bon emission in BRICS. In Na u al esou ces o um. Blackwell Publishing L d.
Zhang, Y., Wang, Y., Zhang, J., Liu, J., Ruan, J., Jin, X., Liu, D., Lu, Z., & Xu, Z. (2024). Resea ch on was e gas ea men echnology
and comp ehensi e en i onmen al pe o mance e alua ion o collabo a i e managemen o pollu ion and ca bon in China’s
pha maceu ical indus y based on li e cycle assessmen (LCA). Science o The To al En i onmen ,919, 170555. [C ossRe ]
Zhao, H., Guo, S., & Zhao, H. (2019). P o incial ene gy e iciency o China quan i ied by h ee-s age da a en elopmen analysis.
Ene gy,166, 96–107. [C ossRe ]
Zhong, S., Liang, S., Zhong, Y., Zheng, Y., & Wang, F. (2022). Measu e on inno a ion e iciency o China’s pha maceu ical manu ac u ing
indus y. F on ie s in Public Heal h,10, 1024997. [C ossRe ] [PubMed]
Zhu, J. (2009). Quan i a i e models o pe o mance e alua ion and benchma king: Da a en elopmen analysis wi h sp eadshee s (Vol. 2).
Sp inge .
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 .