Dossa, Joel Vic o ; Gopang, Aami Ali; Thomas, Da a; Ukwuoma, Chiagoziem Chima
A icle
Does he bank's na u e he e ogenei y ma e ?
En i onmen al, social and go e nance (ESG) pe o mance
and co po a e p o i abili y
Asian Jou nal o Economics and Banking (AJEB)
P o ided in Coope a ion wi h:
Ho Chi Minh Uni e si y o Banking (HUB), Ho Chi Minh Ci y
Sugges ed Ci a ion: Dossa, Joel Vic o ; Gopang, Aami Ali; Thomas, Da a; Ukwuoma, Chiagoziem
Chima (2025) : Does he bank's na u e he e ogenei y ma e ? En i onmen al, social and go e nance
(ESG) pe o mance and co po a e p o i abili y, Asian Jou nal o Economics and Banking (AJEB), ISSN
2633-7991, Eme ald, Leeds, Vol. 9, Iss. 3, pp. 364-394,
h ps://doi.o g/10.1108/AJEB-11-2024-0128
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/334151
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/
Does he bank’s na u e he e ogenei y
ma e ? En i onmen al, social and
go e nance (ESG) pe o mance and
co po a e p o i abili y
Joel Vic o Dossa, Aami Ali Gopang and Da a Thomas
Business School, Sichuan Uni e si y, Chengdu, China, and
Chiagoziem Chima Ukwuoma
OBU, Sino-B i ish Collabo a i e Educa ion, Chengdu Uni e si y o Technology,
Chengdu, China and
College o Nuclea Technology and Au oma ion, Chengdu Uni e si y o Technology,
Chengdu, China
Abs ac
Pu pose – This s udy explo es he impac o en i onmen al, social and go e nance (ESG) pe o mance on
co po a e p o i abili y in Chinese comme cial banks om 2012 o 2022, examining how inhe en di e ences
ac oss bank ypes mode a e his ela ionship.
Design/me hodology/app oach – Using wo-way ixed e ec s, 2SLS (ins umen al a iables), simul anei y
es s and sys em GMM, he s udy analyzes he e ec s o ESG on p o i abili y while add essing endogenei y. To
cap u e he e ogenei y, he sample is s a i ied by owne ship (s a e-owned s non-s a e-owned), go e nance
(join -s ock s non-join -s ock) and loca ion (u ban s u al).
Findings – The esul s show ha ESG pe o mance posi i ely impac s p o i abili y in bo h u al and u ban
comme cial banks, wi h go e nance pe o mance nega i ely a ec ing u ban banks. ESG nega i ely
in luences p o i abili y in s a e-owned banks, whe eas non-s a e-owned banks bene i om posi i e ESG
pe o mance. En i onmen al pe o mance posi i ely a ec s p o i abili y in bo h ypes o banks, bu
go e nance has a nega i e impac . ESG boos s p o i abili y in join -s ock banks bu de ac s om i in non-
join -s ock banks. Social ac i i ies posi i ely a ec p o i abili y ac oss bo h bank ypes, while
en i onmen al ac i i ies a e insigni ican . Go e nance ac i i ies enhance p o i abili y in join -s ock banks
bu educe i in non-join -s ock banks.
P ac ical implica ions – Policymake s should c ea e ie ed ESG egula ions, incen i izing ma ke -d i en
banks while subsidizing s a e-owned banks’ compliance cos s. Bank manage s should ailo ESG in es men s,
ocusing on en i onmen al and social ini ia i es based on loca ion and in es o s mus e alua e ESG sco es
con ex ually.
O iginali y/ alue – This s udy p o ides a g anula analysis o ESG-p o i abili y linkages ac oss China’s
he e ogeneous banking landscape, highligh ing how ins i u ional cha ac e is ics such as owne ship, go e nance
s uc u e and geog aphic ocus shape he inancial implica ions o ESG.
Keywo ds Co po a e p o i abili y, ESG, Bank’s na u e, S a e-owned comme cial banks,
Join comme cial banks, U ban comme cial bank, Ru al comme cial banks
Pape ype Resea ch a icle
1. In oduc ion
The global su ge in En i onmen al, Social, and Go e nance (ESG) adop ion has
ans o med i om a co po a e social esponsibili y ini ia i e in o a c i ical de e minan
AJEB
9,3
364
JEL Classi ica ion — G21, G30, G32, M14, Q56, C33
© Joel Vic o Dossa, Aami Ali Gopang, Da a Thomas and Chiagoziem Chima Ukwuoma. Published
in Asian Jou nal o Economics and Banking. Published by Eme ald Publishing Limi ed. This a icle is
published unde he C ea i e Commons A ibu ion (CC BY 4.0) licence. Anyone may ep oduce,
dis ibu e, ansla e and c ea e de i a i e wo ks o his a icle ( o bo h comme cial and non-comme cial
pu poses), subjec o ull a ibu ion o he o iginal publica ion and au ho s. The ull e ms o his licence
may be seen a Link o he e ms o he CC BY 4.0 licence.
Recei ed 24 No embe 2024
Re ised 17 Ap il 2025
20 June 2025
Accep ed 21 Augus 2025
Asian Jou nal o Economics and Banking
Vol. 9 No. 3, 2025
pp. 364-394
Eme ald Publishing Limi ed
e-ISSN: 2633-7991
p-ISSN: 2615-9821
DOI 10.1108/AJEB-11-2024-0128
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/3/364/10281595/ajeb-11-2024-0128en.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
o inancial pe o mance and in es men a ac i eness (Giannopoulos e al., 2022;
Townsend, 2020). Wi h ESG asse s p ojec ed o each $53 n by 2025, he inancial sec o ,
pa icula ly banks, aces moun ing p essu e o in eg a e sus ainabili y in o co e ope a ions
(Sani e al., 2024; Sebas i~
ao e al., 2024). China’s egula o y push, exempli ied by he
“double-ca bon” policy and ESG disclosu e manda es u he highligh s his shi (Yuan
e al., 2022). Ye , he p o i abili y implica ions o ESG in banking emain con es ed, wi h
s udies epo ing posi i e (Zhang and Lucey, 2022), nega i e (Buallay, 2019, 2020), o
neu al e ec s (Assa e al., 2024; As e iou e al., 2024; F iede e al., 2015). These
inconsis encies s em om a c i ical gap: P io esea ch p edominan ly examines ca bon-
in ensi e indus ies, o e looking he unique dynamics o inancial ins i u ions, especially
in s a e-in luenced economies like China, whe e banks ope a e unde he e ogeneous
owne ship and go e nance s uc u es.
This s udy add esses his gap by analyzing how ESG pe o mance impac s p o i abili y
ac oss 42 Chinese-lis ed comme cial banks (2012–2022), di e en ia ing by s a e
owne ship, join -s ock s a us, and u ban- u al di ides. Employing wo-way ixed e ec s,
2SLS, and sys em GMM, we econcile con lic ing heo e ical pe spec i es: s akeholde
heo y (F eeman and McVea, 2001), which posi s ha ESG enhances p o i abili y h ough
us and e iciency, and ade-o heo y, which ames ESG as a cos bu den (Fu and
Zhang, 2025; Handoyo and Anas, 2024; Zumen e and Bis o a, 2021). In his s udy, we
p o ide he i s g anula e idence on he impac o ESG pe o mance on p o i abili y
wi hin China’s banking sec o , a c i ical ye less en i onmen ally sensi i e indus y. Ou
analysis e eals ha ESG e ec s a e no uni o m ac oss all banks; a he , hey a e
con ingen upon ins i u ional cha ac e is ics such as owne ship, go e nance s uc u e, and
geog aphic loca ion. Speci ically, s a e-owned banks expe ience a educ ion in p o i abili y
om ESG compliance, while p i a e banks gain p o i abili y h ough s a egic ESG
implemen a ion. Go e nance s uc u e also plays a signi ican ole: join -s ock banks see
p o i abili y inc eases om ESG, whe eas non-join -s ock banks expe ience p o i abili y
declines. Geog aphic di e ences u he eme ge, wi h u al banks bene i ing o e all,
pa icula ly om social ini ia i es, while u ban banks ace go e nance- ela ed p o i abili y
challenges. The s udy also highligh s he a ying e ec s o ESG componen s;
en i onmen al ac o s gene ally ha e a posi i e impac on p o i abili y, excep o s a e-
owned banks; social ac o s a e uni e sally bene icial, and go e nance ac o s ha e a mixed
impac , being posi i e in join -s ock banks and nega i e in o he s.
This esea ch con ibu es o he li e a u e in se e al key ways: Fi s , i p o ides sec o al
no el y by examining ESG-p o i abili y linkages in he banking sec o , an a ea la gely
unde explo ed in ESG s udies. Second, by accoun ing o ins i u ional he e ogenei y
s a i ying banks by owne ship, go e nance, and loca ion, we e eal how ESG e ec s a e
con ex -dependen , a nuance o en o e looked in p e ious s udies. Thi d, ou indings ha e
policy implica ions, ad oca ing o ie ed ESG egula ions ha o e subsidies o s a e-owned
banks’ compliance cos s and ma ke -d i en incen i es o p i a e banks. Finally, ou
me hodological igo , inco po a ing obus ness checks and disagg ega ion o ESG
componen s, add esses endogenei y conce ns and unco e s go e nance as a key challenge
o u ban banks.
The pape is o ganized as ollows: Sec ion 2 e iews he li e a u e and hypo heses, Sec ion
3 ou lines he me hodology, Sec ion 4 p esen s he esul s, and Sec ion 5 discusses he
implica ions o policymake s and banke s.
2. Li e a u e and hypo heses de elopmen
To e alua e he ela ionship be ween ESG pe o mance and co po a e pe o mance, a ious
heo ies can be used o suppo such a ela ionship. These heo ies p o ide bo h posi i e
and nega i e pe spec i es on he impac o ESG pe o mance on co po a e p o i abili y.
Asian Jou nal o
Economics and
Banking
365
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/3/364/10281595/ajeb-11-2024-0128en.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
2.1 Theo ies suppo ing ESG pe o mance and co po a e p o i abili y
S akeholde Theo y explains why i ms disclose ESG ac i i ies (F eeman e al., 2018;
Ha ison e al., 2015). F eeman (2010) desc ibes s akeholde s as any g oup o indi idual who
can a ec o be a ec ed by he a ainmen o an o ganiza ion’s objec i es (F eeman e al.,
2010, 2018). S akeholde heo y and ESG s udies a e compa ible since an o ganiza ion’s ESG
p ac ices aim o imp o e managemen , sa is y a ious s akeholde s, and boos inancial
pe o mance (F eeman, 1994). The heo y asse s ha di e en s akeholde s a e a ac ed o
speci ic kinds o in o ma ion, and i ms ace challenges in balancing he demands o di e se
s akeholde g oups. Cen al o a i m’s su i al, in es o s ecei e signi ican a en ion in ESG
epo ing (Talan e al., 2024). The no ma i e aspec o s akeholde heo y suppo s he iew
ha i ms should conside he needs o all s akeholde s-in es o s, employees, sha eholde s,
and he communi y-wi hin he sus ainabili y amewo k (Abdi e al., 2022). Thus, mee ing he
expec a ions o hese s akeholde s, in addi ion o enhancing in es o s’ inancial e u ns, is
c ucial o a i m’s success. Companies ha neglec social esponsibili y o educe implici
cos s may ace mo e explici cos s in he long e m, he eby nega i ely impac ing hei
p o i abili y (Sz}
ocs and Mon ana i, 2025).
Agency Theo y explains he ela ionship be ween agen s (manage s) and p incipals
(sha eholde s) (Jensen and Meckling, 1976). Agency p oblems a ise om he misalignmen o
in e es s be ween managemen , which ocuses on p o i maximiza ion and compensa ion, and
sha eholde s, who seek isk educ ion and alue inc ease (Cen e al., 2025; Khandelwal e al.,
2023; Li e al., 2020). Agency cos s, including ansac ion and in o ma ion cos s, a e p e alen
in he ela ionship be ween agen s and p incipals (O ia , 1988). Since ESG disclosu es o en
add ess many o hese agency- ela ed issues, hey help educe in o ma ion asymme ies and
lowe agency cos s. As a esul , enhancing ESG pe o mance could lead o a educ ion in
agency cos s and a subsequen inc ease in co po a e p o i abili y.
2.2 Theo ies agains ESG pe o mance and co po a e p o i abili y
The ade-o Theo y submi s an an agonis ic ela ionship be ween co po a e p o i abili y (CP)
and sus ainabili y. In es ing in en i onmen al p o ec ion, inc easing employee compensa ion,
and suppo ing communi y ini ia i es a e social and en i onmen al goals ha equi e
signi ican in es men bu can educe p o i abili y and weaken compe i i e ad an age (Wu
and Tham, 2023). Hence, i ms should only engage in ESG ini ia i es i hese e o s esul in
su plus p o i s. (Ma iny e al., 2024) Found ha compelling i ms o unde ake ESG ac i i ies
can lowe hei p o i s. Acco ding o he ade-o heo y, sus ainable p ac ices incu highe
cos s, which can ad e sely a ec p o i abili y (Pham e al., 2021; Zhang and Xie, 2022).
Ins i u ional Theo y u he e ines ou unde s anding by examining how egula o y and
cul u al ac o s shape he ESG-p o i abili y ela ionship. Ins i u ional heo y p oposes ha
i ms a e no me ely economic en i ies bu a e also shaped by he ins i u ional en i onmen in
which hey ope a e. Ins i u ional p essu es, whe he coe ci e (e.g. go e nmen egula ions),
no ma i e (e.g. indus y s anda ds), o mime ic (e.g. adop ing p ac ices based on indus y
leade s), in luence how i ms implemen ESG ini ia i es and how hese ini ia i es, in u n,
a ec p o i abili y (DiMaggio and Powell, 1983). In he con ex o banks in China, s a e-
owned banks may ace signi ican go e nmen -d i en p essu es o align wi h na ional
sus ainabili y goals, which could lead o p o i abili y ade-o s. On he o he hand, non-s a e-
owned and u al banks, less cons ained by go e nmen manda es, may le e age ESG
ini ia i es as pa o a compe i i e s a egy, esul ing in p o i abili y gains. Thus, he impac o
ESG on p o i abili y can a y depending on a bank’s ins i u ional en i onmen and owne ship
s uc u e.
2.3 ESG pe o mance and co po a e p o i abili y
A g owing body o esea ch has examined he impac o En i onmen al, Social, and
Go e nance (ESG) p ac ices on co po a e p o i abili y, yielding a spec um o indings. Many
AJEB
9,3
366
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/3/364/10281595/ajeb-11-2024-0128en.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
schola s highligh he s a egic bene i s o ESG in eg a ion ac oss o ganiza ional unc ions,
no ing imp o emen s in employee mo i a ion, co po a e epu a ion, and e en ion (Chen e al.,
2024a, b; Ga ido-Ruso e al., 2024). Fo ins ance, s udies om he US con ex p opose ha
i ms wi h highe le els o ESG disclosu e end o exhibi supe io ma ke pe o mance,
pa icula ly in he uppe quan iles o ESG epo ing (Vee a el e al., 2024). Zhang and Lucey
(2022) Fu he asse ha ESG pe o mance signi ican ly enhances i m ou comes, p ima ily
by easing inancial cons ain s, which in u n suppo s ope a ional e iciency and long- e m
p o i abili y. Simila ly, Albi a e al. (2020) p o ide empi ical e idence linking ESG
disclosu e sco es o imp o ed co po a e pe o mance, pa icula ly a e he adop ion o
in eg a ed epo ing s anda ds.
Howe e , he li e a u e also p esen s coun e ailing e idence. Some schola s a gue ha
ESG ac i i ies may impose inancial bu dens ha ou weigh hei po en ial bene i s. Fo
example, Saygili e al. (2022) Re eals ha en i onmen al disclosu es can nega i ely a ec a
i m’s inancial ou comes. Buallay (2019, 2021) Consis en ly ind ha ESG pe o mance is
in e sely ela ed o key inancial indica o s such as Re u n on Asse s (ROA), Re u n on Equi y
(ROE), and Tobin’s Q, especially wi hin he banking and inancial se ices sec o s.
Fu he mo e, global e idence om 60 coun ies indica es ha ESG epo ing may lead o
declines in ope a ional and ma ke pe o mance (AlAjmi e al., 2023). In he heal hca e sec o ,
(Kalia and Agga wal, 2023) and Dossa (2025) Found mixed esul s. Dossa (2025), in
pa icula , shows ha while ESG, go e nance, and social disclosu es enhance inancial
pe o mance, en i onmen al ac i i ies a e nega i ely associa ed wi h bo h accoun ing and
ma ke -based pe o mance me ics in Chinese heal hca e i ms. These con as ing indings
emphasize he complexi y and con ex -speci ic na u e o he ESG-pe o mance ela ionship.
To con ibu e o his ongoing deba e and p o ide new e idence om an eme ging ma ke
pe spec i e, pa icula ly China’s banking sec o , we p opose he ollowing hypo heses:
H1a. ESG pe o mance posi i ely in luences co po a e p o i abili y.
H1b. ESG pe o mance nega i ely in luences co po a e p o i abili y.
2.4 En i onmen al pe o mance co po a e p o i abili y
A g owing body o esea ch highligh s he signi ican ole o en i onmen al pe o mance in
shaping co po a e inancial ou comes. Se e al s udies ad ise ha e ec i e en i onmen al
s a egies can enhance i m pe o mance by imp o ing e iciency, educing ope a ional isks,
and os e ing a posi i e co po a e epu a ion (Gull e al., 2022; Vee a el e al., 2024). Fo
example, Gull e al. (2022) ound ha he op-pe o ming i ms in e ms o en i onmen al
ini ia i es ou pe o med a e age i ms inancially, while Li e al. (2024) Demons a ed how
en e p ise-wide adop ion o en i onmen al s anda ds can lowe p oduc ion cos s and mi iga e
en i onmen al isks, ul ima ely inc easing p o i abili y. Simila ly, Wu and Tham (2023)
Obse ed ha p oac i e en i onmen al in es men s can subs an ially inc ease i m alue o e
ime. Zhu e al. (2024) Fu he a gued ha anspa en en i onmen al disclosu es, especially
when amed a ound go e nance a he han in es men cos s, a e mo e posi i ely ecei ed by
in es o s. This aligns wi h indings by Cojoianu e al. (2021) and Ge ged e al. (2021) hose
who emphasize ha en i onmen al anspa ency plays a key ole in managing epu a ional isk
and enhancing i m alua ion.
Howe e , he empi ical e idence is no en i ely consis en . O he s udies epo a nega i e
associa ion be ween en i onmen al ac ions and inancial ou comes. Fo ins ance, Khan e al.
(2024) obse e ha excessi e en i onmen al in es men s may educe ea nings pe sha e,
indica ing a po en ial ade-o be ween en i onmen al esponsibili y and accoun ing
p o i abili y. Peko ic e al. (2018) p o ide a nuanced pe spec i e, iden i ying an in e ed
U-shaped ela ionship be ween en i onmen al pe o mance and ma ke alue among F ench-
lis ed companies. This ecommends ha while mode a e en i onmen al e o s enhance i m
alue, excessi e engagemen may lead o diminishing e u ns. Regional pe spec i es,
Asian Jou nal o
Economics and
Banking
367
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/3/364/10281595/ajeb-11-2024-0128en.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
pa icula ly om China’s s a e-led en i onmen al go e nance sys em, add aluable con ex
(Chen and Li, 2024). The Chinese egula o y en i onmen s ongly incen i izes i ms o adop
g een p ac ices h ough policy mechanisms, subsidies, and en i onmen al sco eca ds. This
op-down app oach c ea es unique p essu es and oppo uni ies o i ms, di e en ia ing he
Chinese con ex om mo e ma ke -d i en egula o y en i onmen s in he Wes (Zhang e al.,
2024; Zhang e al., 2024). Hence, unde s anding en i onmen al pe o mance in China equi es
accoun ing o i s ins i u ional se ing and go e nmen in luence on co po a e sus ainabili y
beha io . D awing om his mixed e idence, we p opose he ollowing hypo heses:
H2a. En i onmen al pe o mance has a posi i e impac on co po a e p o i abili y.
H2b. En i onmen al pe o mance has a nega i e impac on co po a e p o i abili y.
2.5 Social pe o mance and co po a e p o i abili y
Social pe o mance encompasses a wide ange o p ac ices, including wo k o ce di e si y,
employee wel a e, communi y engagemen , heal h and sa e y s anda ds, social isk
managemen , and p oduc esponsibili y. These dimensions con ibu e no only o a
company’s e hical image bu also o i s inancial s anding. Empi ical e idence sugges s ha
s ong social ini ia i es can signi ican ly enhance i m pe o mance. Fo ins ance, Ga ido-
Ruso e al. (2024), ound ha a ious ace s o social pe o mance, such as gende di e si y,
employee mo i a ion, aining, and p oduc quali y, posi i ely in luence he inancial ou comes
o i ms in he ou ism sec o . Ding and Lee (2024) A gue ha e ec i e social s a egies educe
in es o - ela ed isks and enhance pe cep ions o managemen quali y, he eby gi ing i ms a
compe i i e edge. In line wi h his, Albuque que e al. (2019) Emphasize ha companies wi h
highe social in es men s o en expe ience educed sys emic isk and enhanced i m alue,
pa icula ly when social esponsibili y is in eg a ed in o p oduc di e en ia ion s a egies.
Social in es men s can also ele a e employee sa is ac ion, leading o highe p oduc i i y and
inno a ion. Mo eo e , a s ong eco d in co po a e social esponsibili y (CSR) helps o build a
posi i e public image and solidi y s akeholde us (Thomas e al., 2024). F om a s a egic
pe spec i e, aligning social objec i es wi h business goals allows companies o e ol e om a
pu ely p o i -d i en o ien a ion o a socially esponsible business model (Azmi e al., 2021).
Howe e , his ela ionship is no uni e sally posi i e. Eccles e al. (2020) Cau ion ha while
s ong inancial pe o mance may encou age i ms o engage in CSR, he e e se does no
always hold; imp o ed CSR p ac ices do no necessa ily esul in be e inancial pe o mance.
Fu he mo e, Peko ic and Vog (2021) Find ha unde ce ain owne ship s uc u es,
pa icula ly hose wi h high concen a ion, CSR engagemen may ha e a de imen al e ec
on i m p o i abili y. Gi en he mixed empi ical indings and con ex ual di e ences,
he ela ionship be ween social pe o mance and co po a e p o i abili y emains a subjec o
ongoing deba e. Thus, we p opose he ollowing hypo heses:
H3a. Social pe o mance has a posi i e impac on co po a e p o i abili y.
H3b. Social pe o mance has a nega i e impac on co po a e p o i abili y
2.6 Go e nance pe o mance and co po a e p o i abili y
Co po a e go e nance encapsula es a ange o mechanisms designed o ensu e accoun abili y,
anspa ency, and ai ness in a company’s ela ionships wi h i s s akeholde s. These
mechanisms include boa d composi ion, execu i e compensa ion, an i-co up ion p ac ices,
sha eholde igh s, o ganiza ional s uc u e, and cus ome p o ec ion policies. Among hese,
boa d s uc u e and owne ship design a e among he mos equen ly s udied in he con ex o
co po a e inancial pe o mance. Empi ical e idence on he impac o go e nance
pe o mance on p o i abili y p esen s mixed indings. Fo example, Fahad and Bus u (2021)
Analyzed Indian-lis ed i ms and obse ed a nega i e associa ion be ween go e nance- ela ed
AJEB
9,3
368
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/3/364/10281595/ajeb-11-2024-0128en.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
ESG sco es and e u n on asse s (ROA). In con as , Wu e al. (2023), demons a ed a posi i e
impac o go e nance pe o mance on ROA, signi ying ha e ec i e go e nance mechanisms
can enhance i m ou comes. One impo an go e nance ac o is he boa d leade ship s uc u e.
Bane jee e al. (2020) and B aun and Sha ma (2007) wa ned ha a dual ole whe e he boa d
chai also se es as a op execu i e can comp omise boa d independence and ha m i m alue.
Simila ly, empi ical s udies (Ka amanou and Va eas, 2005; Pe e sen, 2009) ha e aised
conce ns abou concen a ed owne ship, a guing ha dominance (Ka amanou and Va eas,
2005; Pe e sen and V edenbu g, 2009) Majo sha eholde s can unde mine mino i y sha eholde
in e es s and educe i m alue. This conce n is pa icula ly ele an in he Chinese con ex ,
whe e s a e and ins i u ional owne ship a e p e alen in lis ed i ms, po en ially a ec ing
go e nance e iciency.
On he o he hand, in e nal go e nance elemen s, such as human capi al p ac ices, a e
posi i ely associa ed wi h inancial ou comes. S udies highligh ha employee-cen ic
go e nance h ough ai compensa ion, lexible wo k a angemen s, di e si y, and sa is ac ion
enhances inno a ion, ROA, and e u n on equi y (Adu, 2022; G ego y, 2022; Sehen Issa e al.,
2022; Sha eeq Nim Al-Maliki e al., 2023). These indings accen ua e he s a egic
impo ance o in eg a ing social and go e nance p io i ies in co po a e decision-making.
The e is also g owing in e es in how go e nance- ela ed disclosu es a ec in es o
pe cep ions and i m pe o mance. Nekhili e al. (2021) explo ed he ole o employee
ep esen a ion on co po a e boa ds, inding i po en ially de imen al o in es o con idence,
e en when agg ega e ESG pe o mance is high. Meanwhile, Salehi and Bashi imanesh (2024)
a gue ha mode a e le els o ESG disclosu e, nei he oo limi ed no o e ly ex ensi e, a e
op imal o i m e iciency, as hey balance anspa ency and cos . In he Chinese ma ke ,
which is cha ac e ized by s a e-d i en go e nance e o ms and egula o y o e sigh ,
unde s anding how go e nance pe o mance shapes p o i abili y is inc easingly i al. S ai i
(2024) concludes ha go e nance e ec i eness, coupled wi h isk mi iga ion, signi ican ly
boos s ESG- ela ed alue c ea ion. Based on he mixed bu insigh ul body o li e a u e, his
s udy aims o explo e how go e nance pe o mance in luences he p o i abili y o lis ed banks
in China. Acco dingly, we p opose he ollowing hypo heses:
H4a. Go e nance pe o mance has a posi i e impac on co po a e p o i abili y.
H4b. Go e nance pe o mance has a nega i e impac on co po a e p o i abili y.
3. Me hodology
3.1 Da a and sample
This s udy analyzes he impac o ESG pe o mance on co po a e p o i abili y using da a om
lis ed comme cial banks on China’s Shanghai and Shenzhen s ock exchanges be ween 2012
and 2022. The sample includes 42 banks o a ious ypes: i e la ge s a e-con olled
comme cial banks (SOCBs), en join s ock comme cial banks (JSCBs), en u al comme cial
banks (RCBs), and se en een u ban comme cial banks (UCBs). Financial da a we e sou ced
om he China S ock Ma ke and Accoun ing Resea ch (CSMAR) da abase, and ESG da a
we e ob ained om he Chinese Resea ch Da a Se ices Pla o m (CNRDS). The ESG sco es
in CNRDS in eg a e bo h sel - epo ed ESG disclosu es and hi d-pa y e i ied assessmen s.
This hyb id app oach aligns wi h domes ic and in e na ional ESG amewo ks, enhancing
objec i i y and educing he isk o g eenwashing. Companies and i ms ha we e delis ed o
had missing inancial o ESG da a we e excluded om he sample. Addi ionally, all con inuous
a iables we e winso ized a he 1s and 99 h pe cen iles o minimize he in luence o ou lie s.
3.2 Va iables and measu emen
Following p e ious li e a u e, we use e u n on asse s (ROA) as ou main measu e o co po a e
p o i abili y, as i cap u es he his o ical pe o mance well (Chen e al., 2023a, b; Khan, 2022;
Asian Jou nal o
Economics and
Banking
369
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/3/364/10281595/ajeb-11-2024-0128en.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
P asha , 2023). The ESG sco e om ESG sco e, En sco e, Soc sco e, and Go sco e as he
main independen a iables, ESG pe o mance, and ESG pe o mance indi idual componen s
as used by o he schola s (Bae e al., 2021; Shakil, 2021; Zhou and Wen, 2022), ob ained om
CNRDS, which is a ed based in line wi h disclosu e s anda ds such as ISO 26000, GRI
s anda ds, SASB s anda d, domes ic and o eign well-es ablished ESG da abases, combined
wi h China’s ele an policies o ESG in o ma ion.
In addi ion, we con ol o ac o s ha a ec i m p o i abili y based on p e ious s udies
(Chen e al., 2023a, b; D�
ıaz e al., 2021; Ding and Lee, 2024; Qu and Zhang, 2023). We con ol
o equi y concen a ion, cash asse a io, g ow h capaci y, le e age, boa d size, boa d
independence, and i m size o p e en es ima ion bias caused by omi ed a iables ha could
also a ec co po a e p o i abili y. See Table 1 o de ailed measu emen s and de ini ions o he
s udy’s a iables.
3.3 Empi ical model
This pape used he s a ic model 1 and 2 o check o ixed e ec issues and dynamic model (3)
o es how ESG in yea -1 impac co po a e p o i abili y while model (4) is o in es iga e he
impac o he h ee componen s o ESG in yea -1 using panel da a o lis ed comme cial banks
om 2012 o 2022.
ROA
i ¼ β
0 þ β
1
ESG
i; þ φ
X
Con ols
i þ α
i þ λ
þ ε
i (1)
ROA
i ¼ β
0 þ β
1
En
i; þ β
2
Soc
i þ β
3
Go
i þ φ
X
Con ols
i þ α
i þ λ
þ ε
i (2)
ROA
i ¼ β
0 þ β
1
ROA
i; −1 þ β
2
ESG
i −1 þ φ
X
Con ols
i þ α
i þ λ
þ ε
i (3)
ROA
i ¼ β
0 þ β
1
ROA
i; −1 þ β
2
En
i −1 þ β
3
Soc
i −1 þ β
4
Go
i −1 þ φ
X
Con ols
i þ α
i þ λ
þ ε
i
(4)
Whe e i, ep esen s i m i in yea , α
i
is he indi idual-speci ic, λ
ime-speci ic e ec and ε
i
e o e m. Table 1 p o ides a de ailed explana ion o s udy a iables and hei measu emen .
The s udy applies wo-way ixed-e ec models o s a ic. In addi ion, o u he analysis, he
wo-s ep sys em gene alized me hod o momen s (GMM) es ima o o equa ions 3 and 4, in
line wi h p e ious s udies ha use dynamic models (Al-Malkawi and Ja aid, 2018; Chen e al.,
2024a, b; Nguyen e al., 2022). The GMM es ima o can sol e endogenous p oblems due o he
p esence o a lagged dependen a iable as an independen a iable in he model.
4. Empi ical esul s and discussion
4.1 Desc ip i e s a is ics
Table 2 summa izes desc ip i e s a is ics based on 303 i m-yea obse a ions om 2012 o
2022. The a e age ROA and ROE a e 0.038 and 0.063, indica ing mode a e p o i abili y
ac oss he sample. ESG sco es exhibi limi ed a ia ion (mean 5 0.733; SD 5 0.050),
sugges ing a gene ally consis en ESG disclosu e le el among lis ed banks. On a e age, banks
demons a e mode a e i m size (mean 5 22.204) and le e age (0.423), hough liquidi y, as
cap u ed by he cash asse a io (CAR), is ela i ely low (mean 5 0.047), e lec ing po en ial
sho - e m unding p essu es. G ow h oppo uni ies (GRW) a e highly dispe sed, and he
a e age bank age is app oxima ely 2.07 (log- ans o med), indica ing a ela i ely young sec o
in capi al ma ke e ms. Go e nance cha ac e is ics sugges mode a e boa d independence
(mean B_indp 5 0.377), ela i ely small boa ds (mean Bsize 5 2.115 in log o m), and
AJEB
9,3
370
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/3/364/10281595/ajeb-11-2024-0128en.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
Table 1. Va iables, ac onyms, measu emen
Va iable Pa ame e s Symbol De ini ion Sou ce Re e ence
Dependen
Fi m Pe o mance ROA
i
The a io o ne income o
o al asse s o i m i in
yea
CSMAR Chen e al. (2023a,
b), Elame and
Boulhaga (2024),
Khan (2022)
ROE
i
Na u al loga i hm o
(Re enue di ided by
o al ca bon emissions)
o i m i in yea
CSMAR P asha (2023),
T uong (2024),
Zimon e al. (2022)
TobinsQ
i
The a io o i m ma ke
alue o he asse ’s
eplacemen alue o
i m i in yea
CSMAR Khan (2022),
T uong (2024),
Zimon e al. (2022)
Independen
ESG sco e ESG
i
The ESG a e age sco e
assigned o each i m is
based on he
En i onmen al, Social,
and Go e nance pilla s.
Range om 0–100 sco e
CNRDS Bae e al. (2021),
Dossa (2025),
Shakil (2021)
En i onmen al
sco e
En
i
I ep esen s he weigh ed
a e age o all key issues
ha all unde he
en i onmen pilla
pe o mance
CNRDS Bae e al. (2021),
Dossa (2025),
Shakil (2021)
Social sco e Soc
i
I ep esen s he weigh ed
a e age o all key issues
ha all unde he social
pilla pe o mance
CNRDS Bae e al. (2021),
Dossa (2025),
Shakil (2021)
Go e nance sco e Go
i
I ep esen s he weigh ed
a e age o all key issues
ha all unde he
go e nance pilla
pe o mance
CNRDS Bae e al. (2021),
Dossa (2025),
Shakil (2021)
Con ols
Size Size
i
Na u al loga i hm o o al
asse
CSMAR Chen e al. (2023a,
b), Ding and Lee
(2024), Vee a el
e al. (2024)
Le e age Le
i
The a io o o al
liabili ies o o al asse s
CSMAR Casciello e al.
(2024), D�
ıaz e al.
(2021)
Cash asse a io CAR
i
Cash and cash
equi alen /To al asse
CSMAR Agnese e al.
(2024), Azmi e al.
(2021), Zhang e al.
(2024)
Fi ms’ g ow h GRW
i
The sales g ow h o he
i m is calcula ed as
(Sales
– Sales
-1
)/Sales
-1
CSMAR D�
ıaz e al. (2021),
Ding and Lee
(2024), Zhang e al.
(2024)
Fi m Age Age
i
Na u al loga i hm o
yea s since i ms ha e
been es ablished
CSMAR D�
ıaz e al. (2021),
Ding and Lee
(2024), Zhang e al.
(2024)
(con inued )
Asian Jou nal o
Economics and
Banking
371
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/3/364/10281595/ajeb-11-2024-0128en.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
pe o mance and p o i abili y a he 10% le el (coe icien : �0.002). This inding aligns wi h
legi imacy heo y (Buallay, 2019), which sugges s ha ESG p ac ices in u ban ins i u ions may
be mo e d i en by socie al expec a ions and egula o y p essu e han by s a egic alue
c ea ion. U ban banks ace heigh ened sc u iny and highe ope a ional cos s associa ed wi h
ESG implemen a ion, such as emissions epo ing, g een lending egula ions, and labo
compliance. As such, hese e o s may esul in a sho - e m ade-o be ween compliance
and p o i abili y, hough po en ially yielding epu a ional o long- e m gains.
Con e sely, u al banks show a sligh ly posi i e ela ionship be ween ESG pe o mance
and p o i abili y, s a is ically signi ican a he 1% le el (coe icien : 0.005). This is likely
because hese banks p io i ize social ESG dimensions, such as communi y ou each, inclusi e
banking, and suppo o local de elopmen ini ia i es ha build cus ome loyal y and social
us . S akeholde engagemen heo y and eal-wo ld examples like he G ameen Bank (Yunus,
2010) illus a e how such banks, ope a ing in unde se ed egions, can achie e social impac
Table 5. The impac o ESG on co po a e p o i abili y (s a e-owned bank, non-s a e-owned bank, join s ock,
non-join , u al, and U ban banks)
Va iables
(1) (2) (3) (4) (5) (6)
SOCB Non-SOCB Join Non-join Ru al U ban
ROA ROA ROA ROA ROA ROA
ESG �0.012*** 0.008*** 0.006*** �0.003*** 0.005*** �0.002***
(0.004) (0.002) (0.002) (0.001) (0.002) (0.002)
Size 0.008*** 0.010*** 0.009*** 0.007*** 0.011*** 0.006***
(0.002) (0.001) (0.001) (0.002) (0.001) (0.002)
Le �0.120*** �0.150*** �0.130*** �0.140*** �0.110*** �0.160***
(0.010) (0.008) (0.009) (0.008) (0.010) (0.009)
CAR 0.150** 0.200** 0.180** 0.160** 0.170** 0.140**
(0.060) (0.050) (0.055) (0.065) (0.055) (0.060)
GRW 0.025*** 0.030*** 0.028*** 0.022*** 0.035*** 0.020***
(0.005) (0.004) (0.004) (0.0505) (0.004) (0.005)
Age �0.008** �0.010** �0.009** �0.007** �0.012** �0.006**
(0.003) (0.002) (0.003) (0.004) (0.003) (0.003)
B_Indp �0.010** �0.012** �0.011** �0.009** �0.015** �0.008**
(0.005) (0.004) (0.004) (0.005) (0.004) (0.005)
Bsize 0.003 0.002 0.004* 0.001 0.005** �0.002
(0.003) (0.002) (0.002) (0.003) (0.002) (0.003)
Top1 0.038*** 0.045*** 0.042*** 0.035*** 0.050*** 0.030***
(0.008) (0.006) (0.007) (0.008) (0.006) (0.008)
Duali y �0.001 0.003*** 0.002** 0.001 0.004*** 0.001
(0.001) (0.002) (0.001) (0.001) (0.002) (0.001)
Bgende 0.008*** 0.012*** 0.010*** 0.007*** 0.015*** 0.006***
(0.003) (0.002) (0.002) (0.003) (0.002) (0.003)
Liq 0.013*** 0.011*** 0.004*** 0.013*** 0.002*** 0.031***
(0.002) (0.001) (0.003) (0.002) (0.002) (0.003)
Cons an �0.162*** �0.185*** �0.042*** �0.294*** �0.042*** �0.005***
(0.006) (0.008) (0.009) (-0.026) (0.089) (0.009)
Obse a ions 55 248 96 207 52 100
R-squa ed 0.950 0.951 0.404 0.395 0.404 0.395
Numbe o i ms 5 37 10 32 10 17
Fi m FE YES YES YES YES YES YES
Yea FE YES YES YES YES YES YES
No e(s): This able epo s ixed e ec s eg ession esul s es ima ing he e ec o ESG pe o mance on e u n on
asse s (ROA) ac oss di e en bank ca ego ies: s a e-owned (SOCB), non-s a e-owned, join -s ock, non-join -
s ock, u al, and u ban banks. All models include i m and yea ixed e ec s. Robus s anda d e o s a e in
pa en heses. ***, **, and * indica e signi icance a he 1%, 5%, and 10% le els, espec i ely
Sou ce(s): Au ho s’ own wo k
AJEB
9,3
378
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/3/364/10281595/ajeb-11-2024-0128en.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
while main aining inancial s abili y. ESG e o s hus ac as di e en ia o s ha help u al
banks s eng hen hei local cus ome base and secu e long- e m sus ainabili y.
These indings emphasize ha ESG pe o mance does no yield uni o m ou comes ac oss
all banks. Ins ead, i s impac depends on o ganiza ional s uc u e, go e nance mechanisms,
egula o y exposu e, and geog aphic posi ioning. Policymake s should conside hese
con ex ual di e ences when designing ESG- ela ed egula ions o incen i es, and in es o s
mus assess ESG ou comes in ligh o bank ype and en i onmen . Bank manage s, meanwhile,
should ailo ESG s a egies o align wi h hei ins i u ional cha ac e is ics, ensu ing ha
sus ainabili y goals a e in eg a ed in o p o i able and p agma ic business models.
4.5 Robus ness check and u he analysis
4.5.1 Endogenei y es . To add ess po en ial endogenei y conce ns a ising om omi ed
a iable bias o e e se causali y be ween ESG pe o mance and i m p o i abili y, his s udy
employs a wo-s age leas squa es (2SLS) ins umen al a iable (IV) app oach, ollowing p io
empi ical s udies (El Ghoul e al., 2011; Fio illo e al., 2025; Kuai e al., 2025; Yu e al., 2024).
We use he a e age ESG sco es o o he i ms ope a ing in he same ci y, p o ince, o indus y
as ins umen s o i m-le el ESG pe o mance.
These ins umen s a e heo e ically and empi ically jus i ied on wo g ounds: i s ,
ega ding ele ance, i ms loca ed in he same geog aphic o indus y con ex o en ace
simila egula o y p essu es, social expec a ions, and s akeholde no ms, which leads o
con e gence in ESG beha io . This is suppo ed by he s ong i s -s age esul s in Table 6,
whe e he ins umen s signi ican ly p edic i m-le el ESG sco es a he 1% le el. Second,
conce ning exclusion es ic ion, while egional o indus y-a e age ESG le els in luence a
i m’s ESG disclosu e p ac ices, hey a e unlikely o di ec ly a ec he p o i abili y o he i m
(ROA), once he i m’s own ESG sco e is con olled o . These b oade a e ages e lec
exogenous en i onmen al and ins i u ional condi ions ha shape ESG beha io , bu no i m-
speci ic ope a ional ou comes. Thus, hei impac on ROA is assumed o be media ed solely
h ough i m-le el ESG, sa is ying he exclusion es ic ion.
Table 6 p esen s he esul s. The i s -s age eg essions (Models M1, M3, M5) demons a e
s ong, s a is ically signi ican associa ions be ween he ins umen s and i m-le el ESG. In
he second s age (Models M2, M4, M6), ESG emains posi i ely associa ed wi h ROA, wi h
coe icien s anging om 0.076 o 0.321, all s a is ically signi ican . Diagnos ic es s suppo
he alidi y o ou ins umen s. The C agg-Donald Wald F-s a is ics and Kleibe gen-Paap k
LM s a is ics a e well abo e he s anda d h esholds, indica ing ha he ins umen s a e no
weak. Fu he mo e, he Hansen J es ails o ejec he null hypo hesis o ins umen alidi y,
sugges ing ha he ins umen s a e no co ela ed wi h he e o e m in he second s age.
Collec i ely, hese esul s con i m ha ESG pe o mance has a causal and obus e ec on i m
p o i abili y, and ha ou ins umen se is bo h alid and e ec i e o add essing endogenei y
conce ns.
4.5.2 Change o dependen a iable, independen a iable. To ensu e he obus ness o he
esul s, he s udy changes bo h he independen and dependen a iables. The obus ness es
esul s in Table 7 examine he impac o di e en ESG a ings ( om CNRDS and Bloombe g)
on co po a e p o i abili y, wi h he dependen a iable changed o ROE and Tobin’s Q. In he
i s wo columns (M1 and M2), ESG a ings om CNRDS a e used wi h ROE and Tobin’s Q
as dependen a iables, demons a ing a posi i e impac o ESG a ings on co po a e
p o i abili y. Speci ically, CNRDS ESG a ings signi ican ly in luence ROE (0.117, p < 0.01)
and Tobin’s Q (0.411, p < 0.10). The nex wo columns (M3 and M4) u ilize ESG a ings om
Bloombe g, showing ha ESG a ings posi i ely a ec ROA and ROE, wi h coe icien s o
0.002 (p < 0.01) o ROA and 0.005 (p < 0.01) o ROE. These esul s indica e ha ESG
pe o mance, ega dless o whe he measu ed by CNRDS o Bloombe g a ings, signi ican ly
enhances co po a e p o i abili y and ma ke alua ion. Fu he mo e, he consis ency o hese
indings, despi e changes in bo h dependen and independen a iables, s eng hens he
Asian Jou nal o
Economics and
Banking
379
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/3/364/10281595/ajeb-11-2024-0128en.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
Table 6. Endogenei y esolu ion using 2SLS ins umen al a iable app oach
Va iable
1s s age 2nd s age 1s s age 2nd s age 1s s age 2nd s age
(M1) (M2) (M3) (M2) (M3) (M4)
ESG ROA ESG ROA ESG ROA
ESG_Ci y
(IV)
0.861***
(0.012)
– – – – –
ESG_
P o ince
(IV)
– – 0.821***
(0.026)
– – –
ESG_
Indus y (IV)
– – – – 0.013***
(0.003)
–
ESG – 0.138***
(0.019)
– 0.076**
(0.034)
– 0.321***
(0.093)
Size 0.013***
(0.000)
0.007***
(0.000)
0.014***
(0.000)
0.008***
(0.001)
0.015***
(0.000)
0.328***
(0.037)
Le �0.042***
(0.002) �0.103***
(0.002) �0.047***
(0.002) �0.106***
(0.003) �0.050***
(0.002) �0.148
(0.175)
CAR 0.018***
(0.004)
0.287***
(0.005)
0.019***
(0.004)
0.288***
(0.005)
0.023***
(0.004)
1.010***
(0.262)
GRW �0.002***
(0.001)
0.035***
(0.001) �0.003***
(0.001)
0.034***
(0.001) �0.003***
(0.001) �0.114
(0.165)
Age �0.008***
(0.000) �0.007***
(0.000) �0.010***
(0.000) �0.008***
(0.000) �0.011***
(0.000)
0.095***
(0.025)
B_Indp 0.070***
(0.000) �0.027***
(0.006)
0.079***
(0.006) �0.022***
(0.007)
0.073***
(0.006) �0.022**
(0.009)
Bsize 0.010***
(0.002)
0.008***
(0.002)
0.011***
(0.002)
0.008***
(0.002)
0.007***
(0.002)
0.091*
(0.055)
Top1 0.012***
(0.002)
0.030***
(0.002)
0.012***
(0.002)
0.030***
(0.002)
0.010***
(0.002) �0.157
(0.359)
Duali y �0.003***
(0.008) �0.002***
(0.001) �0.003***
(0.001) �0.002***
(0.001) �0.002***
(0.001)
0.034
(0.153)
Bgende 0.010***
(0.002)
0.010***
(0.002)
0.012***
(0.002)
0.011***
(0.229)
0.011***
(0.013)
0.018
(0.241)
Liq 0.001***
(0.002)
0.003***
(0.001)
0.016***
(0.002)
0.021***
(0.012)
0.014***
(0.014)
0.016
(0.041)
Cons an �0.209***
(0.011) �0.918***
(0.042) �0.206***
(0.020) �0.918***
(0.002)
0.383***
(0.007)
0.918***
(0.042)
C agg-
Donald Wald
4235.96 4235.963 1155.83 1155.835 14.98 14.975
Kleibe gen-
Paap
4802.77 4802.775 996.86 996.865 15.93 15.933
Hansen J
S a is ics
– 0.000 – 0.000 – 0.000
Ande son-
Rubin Wald
0.0000 – 0.0277 – 0.0000 –
S ock-w igh
LM
0.0000 – 0.0278 – 0.0000 –
Yea FE Yes Yes Yes Yes Yes Yes
Fi m FE Yes Yes Yes Yes Yes Yes
Obs 303 303 303 303 303 303
No e(s): This able p esen s he esul s om he wo-s age leas squa es (2SLS) app oach used o add ess
endogenei y issues. In he i s s age (M1, M3, M5), he ins umen a iables (IV) o ESG a e used o p edic
ESG sco es. In he second s age (M2, M4, M6), he es ima ed ESG alues a e hen eg essed on co po a e
p o i abili y, measu ed by ROA, o esol e po en ial endogenei y p oblems. ***, **, and * deno e signi icance a
he 1%, 5%, and 10% le els, espec i ely. Robus s anda d e o s a e epo ed in pa en heses o all columns
Sou ce(s): Au ho s’ own wo k
AJEB
9,3
380
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/3/364/10281595/ajeb-11-2024-0128en.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
obus ness and eliabili y o he ela ionship be ween ESG pe o mance and co po a e
p o i abili y.
4.5.3 Simul anei y issues. Table 8 p esen s he esul s om add essing po en ial
simul anei y issues and conce ns in he ela ionship be ween ESG and co po a e
p o i abili y. The i s -s age esul s in column M1 epo he ins umen ed a iable, ESG_L,
while columns M2 h ough M4 examine he dynamic ela ionship by inco po a ing one-pe iod
lag alues o he independen and con ol a iables. In column M1, ESG lagged by one pe iod
(ESG -1) signi ican ly in luences ROA (0.039, p < 0.01) and ROE (0.079, p < 0.01),
indica ing ha pas ESG pe o mance has a posi i e e ec on u u e p o i abili y. Columns M2
and M4 show simila esul s o ESG’s impac on ROA and ROE, wi h coe icien s o 0.037
(p < 0.01) and 0.108 (p < 0.01), espec i ely, ein o cing he posi i e e ec o ESG on
p o i abili y. Con ol a iables such as Size, Le e age, Cash Flow om Ope a ions (CFLOA),
G ow h, and Top1 ( he la ges sha eholde ) also exhibi signi ican ela ionships wi h i m
pe o mance, wi h lagged alues showing obus co ela ions. The obus ness o hese indings
is con i med by he consis ency o signi ican coe icien s ac oss mul iple speci ica ions. The
esul s indica e ha ESG, alongside o he key inancial ac o s, signi ican ly d i es co po a e
p o i abili y, wi h adjus men s o simul anei y u he s eng hening he eliabili y o he
indings.
4.5.4 Change o es ima ion echnique. The esul s om he Sys em GMM analysis in
Table 9 poin o a clea and meaning ul link be ween ESG pe o mance and bank p o i abili y
in China. A highe ESG sco e is associa ed wi h be e inancial ou comes, speci ically, a one-
s anda d-de ia ion imp o emen in ESG pe o mance is linked o a 1% poin inc ease in e u n
on asse s (ROA), wi h he coe icien s anding a 0.010 and s a is ically signi ican a he 1%
Table 7. Robus ness change o dependen and independen a iable
ΔVa iables
Co po a e p o i abili y ESG a ing bloombe g
(M1) (M2) (M3) (M4)
ROE TBQ ROA ROE
ESG 0.117*** (0.030) 0.411* (0.215) 0.002*** (0.000) 0.005*** (0.001)
Size 0.029*** (0.016) �0.333*** (0.029) 0.013*** (0.001) 0.030***
(0.004)
Le �0.278 *** (0.021) �0.584 *** (0.103) �0.147*** (0.005) �0.279*** (0.016)
CAR 0.334*** (0.021) 0.867*** (0.137) 0.180*** (0.008) 0.334*** (0.021)
GRW 0.078*** (0.003) 0.175*** (0.023) 0.032*** (0.001) 0.078*** (0.003)
Age �0.036 (0.034) 1.183*** (0.029) �0.011*** (0.001) �0.001 (0.003)
B_Indp �0.036 (0.034) �0.079 (0.283) �0.014 (0.011) �0.035 (0.034)
Bsize �0.004 (0.013) 0.107 (0.089) 0.001 (0.004) �0.004 (0.013)
Top1 0.112*** (0.023) 0.502*** (0.160) 0.042*** (0.007) 0.112*** (0.023)
Duali y 0.007* (0.004) �0.008 (0.027) 0.002* (0.001) 0.007 * (0.004)
Bgende 0.000* (0.004) 0.808*** (0.102) 0.000 (0.004) �0.022* (0.012)
Liq 0.031*** (0.002) 0.013*** (0.001) 0.012*** (0.002) 0.001*** (0.001)
Cons an �0.023*** (0.012) 6.799*** (0.647) �0.186*** (0.025) �0.490*** (0.086)
Fi m FE Yes Yes Yes Yes
Yea FE Yes Yes Yes Yes
Adjus ed R
2
0.162 0.087 0.292 0.162
Obs 303 303 303 303
No e(s): The able p esen s obus ness es esul s whe e he dependen a iables ha e been changed o ROA
(Re u n on Asse s) and Tobin’s Q (TBQ). In he i s wo columns, ESG a ings om CNRDS a e used o assess
he impac on i m p o i abili y. In he hi d column, ESG a ings om Bloombe g a e employed o check hei
impac on i m p o i abili y in e ms o ROA and ROE. The p- alues, calcula ed using obus s anda d e o s, a e
epo ed in pa en heses, wi h s a is ical signi icance indica ed by ***, **, and * a he 1%, 5%, and 10% le els,
espec i ely
Sou ce(s): Au ho s’ own wo k
Asian Jou nal o
Economics and
Banking
381
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/3/364/10281595/ajeb-11-2024-0128en.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
le el. When we b eak ESG down in o i s componen s, en i onmen al and go e nance ac o s
eme ge as he s onges d i e s o p o i abili y, wi h coe icien s o 0.020 and 0.013,
espec i ely, bo h signi ican a he 1% le el. Social ac o s also ha e a posi i e e ec (0.017),
hough sligh ly weake and signi ican a he 5% le el. The posi i e and signi ican coe icien
on lagged ROA ( anging om 0.235 o 0.248) e lec s he pe sis ence o p o i abili y o e
ime, consis en wi h p io esea ch on dynamic panels. Impo an ly, all diagnos ic es s
suppo he alidi y and s eng h o ou model. The Hansen and Sa gan es s con i m he
eliabili y o ou ins umen s, while he lack o second-o de au oco ela ion and a a o able
Du bin-Wu-Hausman es (p 5 0.11) con i m obus ness and help ule ou endogenei y
conce ns. Findings emphasize ha in es ing in ESG, especially in en i onmen al and
go e nance dimensions, can pay o inancially o banks, and hese esul s hold s eady e en
unde al e na i e es ima ion me hods like Di e ence GMM.
4.6 Fu he analysis o ESG indi idual componen s’ impac on co po a e p o i abili y
Table 10 p esen s he impac o ESG indi idual componen s on co po a e p o i abili y (ROA)
ac oss di e en i m ypes in China’s banking sec o . The indings show di e se ela ionships
be ween ESG ac o s and p o i abili y, in luenced by he na u e o he banks and hei
ope a ional en i onmen s. En i onmen al (E) ac o s ha e a gene ally nega i e e ec on
p o i abili y o S a e-Owned Comme cial Banks (SOCBs) (�0.002, signi ican a he 1%
le el) and non-join banks (�0.007, signi ican a he 1% le el), likely due o he inancial
bu den o complying wi h s ingen en i onmen al egula ions in s a e-con olled sec o s. This
aligns wi h Buallay e al. (2019), who highligh how manda o y sus ainabili y p ac ices in
hea ily egula ed sec o s can impose sho - e m cos s on p o i abili y, suppo ing he ade-o
Table 8. Simul anei y issues
Va iable
Co po a e p o i abili y
(M1) (M2) (M3) (M4)
ROA ROA ROE ROE
ESG
�1
0.039*** (0.010) 0.037*** (0.010) 0.079*** (0.029) 0.108*** (0.028)
Size
�1
–�0.002*** (0.001) – �0.013*** (0.002)
Le
-1
–�0.024*** (0.004) –�0.012 (0.011)
CFLOA
�1
– 0.114*** (0.007) – 0.219*** (0.017)
G ow h
�1
– 0.015*** (0.001) – 0.038*** (0.003)
Age
�1
–�0.000 (0.001) – 0.001 (0.002)
Ind_D
�1
–�0.004 (0.011) – 0.006 (0.028)
Boa d
�1
–�0.004 (0.003) –�0.011 (0.010)
Top1
�1
– 0.012*** (0.004) – 0.044*** (0.012)
Duali y
�1
– 0.002 (0.001) – 0.004 (0.003)
Bgende
�1
–�0.025*** (0.004) –�0.039*** (0.011)
Liq
�1
0.021*** (0.004) 0.003*** (0.001) 0.003*** (0.001) 0.001*** (0.000)
Cons an 0.039*** (0.10) 0.082*** (0.017) 0.012*** (0.002) �0.065*** (0.006)
Yea FE Yes Yes Yes Yes
Fi m FE Yes Yes Yes Yes
Adjus ed R
2
0.014 0.092 0.005 0.055
Obs 303 303 303 303
No e(s): This able p esen s esul s om he simul anei y app oach, add essing po en ial endogenei y in he
ela ionship be ween ESG and co po a e p o i abili y. The i s -s age esul s in column M1 show he
ins umen ed a iable, ESG_L, while M2 epo s he panel eg ession using he es ima ed ESG alues. M3 and
M4 examine he dynamic ela ionship by including a one-pe iod lag o he independen and con ol a iables.
***, **, and * deno e signi icance a he 1%, 5%, and 10% le els, espec i ely. Robus s anda d e o s a e
epo ed in pa en heses o all columns
Sou ce(s): Au ho s’ own wo k
AJEB
9,3
382
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/3/364/10281595/ajeb-11-2024-0128en.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
heo y (Pham e al., 2021). In con as , non-SOCBs (0.004, signi ican a he 1% le el) and
Join Banks (0.005, signi ican a he 1% le el) show posi i e e ec s, sugges ing ha g een
in es men s align wi h business s a egies and enhance p o i abili y, especially wi h he
in eg a ion o o eign expe ise in g een inance. In e es ingly, he posi i e en i onmen al
coe icien o SOCBs challenges he indings o Khan e al. (2024), signi ying ha China’s
unique policy landscape migh enable as e e u ns om g een in es men s compa ed o o he
ju isdic ions. Social (S) componen s also yield mixed esul s. Fo SOCBs, he coe icien is
nega i e (�0.007, signi ican a he 1% le el), indica ing ha social ini ia i es may no yield
immedia e inancial e u ns, possibly due o s a e-imposed ine iciencies and igid compliance
amewo ks. Howe e , o Non-SOCBs (0.012, signi ican a he 1% le el), Join Banks
(0.010, signi ican a he 1% le el), and Ru al Banks (0.015, signi ican a he 1% le el),
posi i e e ec s a e obse ed. These esul s e lec ha socially esponsible p ac ices, such as
communi y engagemen and employee wel a e, enhance cus ome us and lead o inc eased
business pe o mance, indings consis en wi h s akeholde heo y (F eeman, 2010) and he
global e idence o Zhang and Lucey (2022). Speci ically, u al banks’ s ong social
pe o mance empi ically suppo s Yunus’ (2010) mic o inance model, whe e local
engagemen d i es p o i abili y by s eng hening communi y ies. Go e nance (G) ac o s
show a nega i e e ec o u ban banks (�0.003, signi ican a he 1% le el), p oposing ha he
complex go e nance equi emen s in u ban se ings inc ease ope a ional cos s. This inding
suppo s he “compliance cos hypo hesis” ou lined by Maji and Lohia (2023), which a gues
ha u ban banks ace ele a ed egula o y bu dens ha may o se go e nance bene i s. On he
o he hand, non-SOCBs (�0.004, signi ican a he 1% le el) and Ru al Banks (�0.001,
Table 9. Sys em GMM
(1) (2) (3)
Va iables
GMM-ESG
(Collapsed)
GMM-E-S-G
(Robus )
Di -GMM
(Robus ness)
Full sample ROA ROA
ESG 0.010*** 0.009***
(0.021) (0.019)
En 0.020*** –
(0.031) –
Soc 0.017*** –
(0.035) –
Go 0.013** –
(0.024) –
L.ROA 0.235*** 0.248*** 0.221***
(0.485) (0.341) (0.402)
Con ols YES YES YES
Cons an �0.162* �0.185* �0.042
(0.006) (0.008) (0.089)
AR (1) (p- alue) 0.077 0.082 0.085
AR (2) (p- alue) 0.732 0.566 0.701
Hansen (p- alue) 0.722 0.874 0.815
Sa gan es (P ob > Chi2) 0.903 0.70
Hansen es 0.722 0.874
No e(s): This able epo s dynamic panel es ima es using Sys em GMM and Di e ence GMM o assess he
impac o ESG and i s componen s on ROA. Models accoun o po en ial endogenei y wi h lagged dependen
a iables and include i m-le el con ols. S anda d e o s a e obus o he e oskedas ici y. Diagnos ic es s (AR
(1), AR (2), Hansen, and Sa gan) con i m ins umen alidi y and model speci ica ion. ***, **, and * indica e
signi icance a he 1%, 5%, and 10% le els, espec i ely
Sou ce(s): Au ho s’ own wo k
Asian Jou nal o
Economics and
Banking
383
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/3/364/10281595/ajeb-11-2024-0128en.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
signi ican a he 5% le el) also show nega i e coe icien s, po en ially e lec ing he cos s o
implemen ing go e nance s uc u es in en i onmen s wi h limi ed egula o y suppo .
No ably, Join Banks epo a ma ginally posi i e go e nance e ec (0.001), ex ending he
insigh s o B�
a ae e al. (2021) by illus a ing how join en u es’ hyb id go e nance models,
o en combining domes ic and o eign o e sigh , can be e balance compliance e o s wi h
pe o mance gains. The esul s sugges ha he na u e o he bank (s a e-owned s p i a e,
u ban s u al) plays a signi ican ole in how ESG ac o s in luence p o i abili y. S a e-owned
banks a e likely cons ained by go e nmen p io i ies ha emphasize non- inancial objec i es,
esul ing in less a o able ESG-p o i abili y ela ionships. In con as , p i a e and join banks
bene i mo e om ma ke -d i en ESG p ac ices, whe e g een and social ini ia i es align wi h
Table 10. ESG indi idual componen s and co po a e p o i abili y
Va iables
(1) (2) (3) (4) (5) (6)
SOCB Non-SOCB Join Non-join Ru al U ban
ROA ROA ROA ROA ROA ROA
En �0.002*** 0.004*** 0.005*** �0.007*** 0.003*** �0.006***
(0.001) (0.002) (0.002) (0.004) (0.002) (0.002)
Soc �0.007*** 0.012*** 0.010*** 0.013*** 0.015*** 0.001
(0.003) (0.002) (0.002) (0.002) (0.002) (0.002)
Go �0.005** �0.004*** 0.001* 0.0002* �0.001** �0.003***
(0.003) (0.001) (0.001) (0.001) (0.001) (0.001)
Size 0.004*** 0.012*** 0.003*** 0.006*** 0.013*** 0.006***
(0.002) (0.001) (0.001) (0.002) (0.001) (0.002)
Le �0.004*** �0.145*** �0.130*** �0.120*** �0.140*** �0.120***
(0.010) (0.008) (0.009) (0.008) (0.010) (0.009)
CAR 0.042** 0.121** 0.170** 0.150** 0.150** 0.120**
(0.060) (0.050) (0.055) (0.065) (0.055) (0.060)
GRW 0.006*** 0.038*** 0.001*** 0.003*** 0.034*** 0.022***
(0.005) (0.004) (0.004) (0.0505) (0.004) (0.005)
Age �0.045** �0.013** �0.041** �0.002** �0.012** �0.004**
(0.003) (0.002) (0.003) (0.004) (0.003) (0.003)
B_Indp �0.010** �0.012** �0.011** �0.009** �0.015** �0.008**
(0.005) (0.004) (0.004) (0.005) (0.004) (0.005)
Bsize 0.006 0.002 0.004* 0.001 0.005** �0.002
(0.003) (0.002) (0.002) (0.003) (0.002) (0.003)
Top1 0.004*** 0.041*** 0.081*** 0.033*** 0.046*** 0.030***
(0.006) (0.005) (0.005) (0.006) (0.006) (0.005)
Duali y �0.004 0.005*** 0.006** 0.002 0.005*** 0.001
(0.001) (0.002) (0.001) (0.001) (0.002) (0.001)
Bgende 0.007*** 0.014*** 0.012*** 0.006*** 0.013*** 0.006***
(0.003) (0.002) (0.002) (0.003) (0.002) (0.003)
Liq 0.009*** 0.013*** 0.014*** 0.009** 0.015*** 0.006**
(0.002) (0.001) (0.003) (0.002) (0.002) (0.003)
Cons an �0.162*** �0.185*** �0.042*** �0.294*** �0.042*** �0.005***
(0.006) (0.008) (0.009) (�0.026) (0.089) (0.009)
Obse a ions 55 248 96 207 52 100
R-squa ed 0.921 0.892 0.851 0.882 0.912 0.874
Numbe o i ms 5 37 10 32 10 17
Fi m FE YES YES YES YES YES YES
Yea FE YES YES YES YES YES YES
No e(s): This able p esen s ixed e ec s eg ession es ima es o he ela ionship be ween he en i onmen al
(En ), social (Soc), and go e nance (Go ) componen s o ESG pe o mance and e u n on asse s (ROA), ac oss
di e en bank ca ego ies. All models include i m and yea ixed e ec s. Robus s anda d e o s a e shown in
pa en heses. ***, **, and * indica e signi icance a he 1%, 5%, and 10% le els, espec i ely
Sou ce(s): Au ho s’ own wo k
AJEB
9,3
384
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/3/364/10281595/ajeb-11-2024-0128en.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
p o i abili y goals. Join banks, in pa icula , bene i om c oss-bo de go e nance
in eg a ion, while u al banks le e age social ini ia i es o os e communi y connec ions
and imp o e inancial pe o mance, echoing he success model o mic o inance ins i u ions
like G ameen Bank (Ahmed Chowdhu y and Somani, 2020; Yunus e al., 2010). These
indings ein o ce he impo ance o ailo ing ESG s a egies o a i m’s ins i u ional and
ope a ional con ex o maximize inancial and socie al alue.
4.7 Summa y o hypo hesis es ing esul s
To consolida e he empi ical indings ac oss di e en models and subg oups, Table 11
summa izes he ou comes o hypo hesis es ing. The esul s con i m ha ESG-pe o mance
ela ionships a y ac oss bank ypes and componen s. While posi i e p o i abili y e ec s
domina e in p i a e and join -s ock banks, s a e-owned and u ban banks exhibi mo e
cons ained o nega i e associa ions, especially in he en i onmen al and go e nance
dimensions.
5. Conclusion, implica ions, and limi a ions
5.1 Conclusion
The s udy analyzes he impac o ESG pe o mance on he co po a e p o i abili y (CP) o lis ed
Chinese comme cial banks, conside ing hei na u e di e ence om 2012 o 2022 wi h a ocus
Table 11. Summa y o hypo hesis es ing esul s
Hypo hesis Desc ip ion Resul Suppo ed? Rema ks
H1a ESG pe o mance posi i ely
in luences p o i abili y
Suppo ed in Non-
SOCBs, JSCBs,
RCBs
Yes S ong posi i e
coe icien s a 1% le el
H1b ESG pe o mance
nega i ely in luences
p o i abili y
Suppo ed in
SOCBs, UCBs
Yes SOCBs (�0.012), UCBs
(�0.002)
H2a En i onmen al pe o mance
posi i ely impac s
p o i abili y
Suppo ed in Non-
SOCBs, JSCBs
Yes Coe icien s: 0.004 and
0.005 a 1% le el
H2b En i onmen al pe o mance
nega i ely impac s
p o i abili y
Suppo ed in
SOCBs, Non-
JSCBs
Yes Coe icien s: �0.002 and
�0.007 a 1% le el
H3a Social pe o mance
posi i ely impac s
p o i abili y
Suppo ed in Non-
SOCBs, JSCBs,
RCBs
Yes S onges o RCBs
(0.015), all signi ican a
1%
H3b Social pe o mance
nega i ely impac s
p o i abili y
Suppo ed in
SOCBs
Yes Coe icien : �0.007 a 1%
le el
H4a Go e nance pe o mance
posi i ely impac s
p o i abili y
Weak e idence in
JSCBs only
Pa ially JSCBs (0.001, ma ginal)
H4b Go e nance pe o mance
nega i ely impac s
p o i abili y
Suppo ed in UCBs,
Non-SOCBs, RCBs
Yes Nega i e coe icien s
ac oss g oups, signi ican
a 1–5%
No e(s): This able summa izes he esul s o hypo heses H1a–H4b ac oss di e en bank ypes. The indings
e lec he he e ogenei y in he ela ionship be ween ESG componen s (En i onmen al, Social, and Go e nance)
and co po a e p o i abili y (ROA). “Suppo ed” indica es s a is ical signi icance in he expec ed di ec ion; “No
Suppo ed” deno es insigni ican o opposi e-di ec ion esul s. Signi icance le els a e based on p- alues:
***p < 0.01, **p < 0.05, *p < 0.1
Sou ce(s): Au ho s’ own wo k
Asian Jou nal o
Economics and
Banking
385
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/3/364/10281595/ajeb-11-2024-0128en.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
on how ins i u ional cha ac e is ics shape his ela ionship. The indings highligh signi ican
di e ences ac oss a ious ypes o banks, ecommending ha ESG’s impac on p o i abili y is
no uni o m bu a he con ingen on he bank’s na u e and ins i u ional con ex . Fo s a e-
owned comme cial banks (SOCBs), he s udy e eals a nega i e associa ion be ween ESG
pe o mance and p o i abili y, which suppo s he ade-o heo y. This indica es ha he cos s
associa ed wi h ESG compliance, o en d i en by policy manda es, may ou weigh immedia e
inancial gains. On he o he hand, non-s a e-owned comme cial banks (non-SOCBs), join -
s ock comme cial banks (JSCBs), and u al comme cial banks (RCBs) gene ally expe ience a
posi i e ela ionship be ween ESG pe o mance and p o i abili y. This inding aligns wi h
s akeholde heo y, which ad oca es ha ma ke -o ien ed ins i u ions can con e
sus ainabili y e o s in o compe i i e ad an ages, he eby enhancing long- e m inancial
pe o mance.
The s udy also emphasizes he u ban- u al di ide in ESG impac s. U ban banks o en ace
challenges due o compliance-hea y ESG amewo ks, while u al banks le e age
communi y- ocused ini ia i es ha con ibu e o p o i abili y. These indings a e obus
ac oss a ious econome ic echniques, including sys em GMM and ins umen al a iable
app oaches, unde sco ing ha he inancial implica ions o ESG ini ia i es should be assessed
wi hin he speci ic egula o y and ins i u ional con ex o each bank ype. Ul ima ely, he s udy
calls o ailo ed ESG s a egies ha conside he unique cha ac e is ics and ope a ing
en i onmen s o di e en banks.
5.2 Theo e ical and p ac ical implica ions
The a ying impac s o ESG pe o mance ac oss di e en ypes o banks o e aluable
heo e ical and p ac ical insigh s. Theo e ically, his s udy in eg a es mul iple pe spec i es. I
a i ms he ole o s akeholde heo y in explaining ESG-d i en success wi hin ma ke -
o ien ed, lexible ins i u ions, while also showing ha ade-o heo y and legi imacy heo y
be e cap u e he ESG adop ion challenges aced by policy-d i en o geog aphically
cons ained banks. This highligh s he impo ance o con ex ualizing ESG e o s a he han
applying uni o m expec a ions ac oss ins i u ions. Addi ionally, ins i u ional heo y
(DiMaggio and Powell, 1983) p o ides a compelling amewo k o unde s anding he
a ia ion in ESG p ac ices ac oss di e en ypes o banks. S a e-owned banks, shaped by
s ong egula o y manda es and public se ice obliga ions, p ima ily ope a e unde coe ci e
ins i u ional p essu es ules, laws, and policies imposed by he s a e. These banks o en pu sue
ESG compliance as a means o ul illing na ional objec i es, such as en i onmen al p o ec ion
and social equi y, a he han immedia e inancial gain. Thei ESG engagemen is la gely
d i en by alignmen wi h go e nmen -led sus ainabili y agendas, exempli ied by ins umen s
like he G een C edi Guidelines and inancial inclusion manda es.
In con as , join -s ock and u ban comme cial banks a e mo e exposed o mime ic
p essu es, which a ise om he endency o imi a e success ul pee s in unce ain o compe i i e
en i onmen s. These banks adop ESG p ac ices s a egically, no only o emain compe i i e
bu also o mee e ol ing s akeholde expec a ions and signal legi imacy o in es o s.
Simila ly, u al comme cial banks expe ience a mix o mime ic and no ma i e p essu es,
whe e p o essional s anda ds and communi y engagemen no ms in luence hei ESG beha io
o en ocusing on social sus ainabili y h ough local de elopmen and inclusi e inancing.
By explici ly ecognizing how di e en ypes o ins i u ional p essu es shape ESG
mo i a ion coe ci e o s a e-owned, mime ic o ma ke -d i en, and no ma i e o
communi y-based ins i u ions his s udy adds heo e ical nuance o he ESG-pe o mance
nexus. Unde s anding hese dynamics helps explain why he inancial implica ions o ESG
adop ion di e ac oss bank ypes. I also unde sco es he necessi y o ailo ing ESG s a egies
o i each bank’s ins i u ional con ex and p e ailing p essu e mechanisms.
P ac ically, hese insigh s s ess he u gency o designing ailo ed ESG s a egies based on
he ins i u ional na u e. Fo ins ance:
AJEB
9,3
386
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/3/364/10281595/ajeb-11-2024-0128en.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
(1) S a e-owned banks could bene i om mo e explici egula o y guidance and
p e e en ial ea men when implemen ing ESG- ela ed lending. China’s G een C edi
Guidelines (2012) and PBOC’s G een Financial Re o m Pilo Zones al eady ep esen
ea ly examples o ie ed ESG egula ion, o e ing lexible policy implemen a ion in
egions like Zhejiang and Guangdong. These ini ia i es can se e as scalable
empla es o di e en ia ed ESG manda es.
(2) Join -s ock banks, such as China Me chan s Bank, ha e success ully issued g een
bonds and ESG-linked loans, e lec ing he easibili y o ma ke -based ESG
s a egies. These banks should be encou aged o scale such e o s wi h incen i es
like ax deduc ions o g een inancing e u ns.
(3) Ru al comme cial banks, such as Jiangsu Ru al Comme cial Bank, ha e pionee ed
mic o inance p og ams suppo ing sus ainable ag icul u e. These banks could be
u he suppo ed wi h policy subsidies o expand ESG-linked communi y
de elopmen ini ia i es, ein o cing hei ole in u al e i aliza ion.
F om a egula o y s andpoin , ou indings ad oca e o he adop ion o ie ed ESG
amewo ks ha e lec ins i u ional di e si y. Ra he han applying uni o m ESG disclosu e
equi emen s, egula o s like he China Banking and Insu ance Regula o y Commission
(CBIRC) should di e en ia e ESG benchma ks o s a e-owned, join -s ock, and u al banks.
Fo example, CBIRC’s 2021 g een inance guidelines can e ol e in o ie ed sys ems by
linking ESG compliance o each bank’s owne ship, ma ke exposu e, and egional ole.
In es o s should also in e p e ESG signals wi hin ins i u ional con ex s. A high ESG sco e
a a join -s ock bank may indica e ope a ional e iciency and ma ke leade ship, while a simila
sco e a a s a e-owned bank may e lec egula o y compliance wi hou p o i abili y gains.
Con ex ual awa eness is hus essen ial o accu a e ESG alua ion.
Policymake s mus ecognize hese ins i u ional dynamics. ESG policies and incen i es
should be s a i ied, e.g. in oducing ESG pe o mance-based subsidies o s a e-owned banks
pu suing long- e m g een in as uc u e p ojec s, while allowing p i a e banks mo e lexibili y
o inno a e wi h ESG-linked in es men p oduc s.
Finally, bank manage s mus cus omize ESG s a egies. S a e-owned banks should align
ESG e o s wi h na ional s a egic goals (e.g. “Dual Ca bon” a ge s), e en i inancial e u ns
a e delayed. Join -s ock banks should in eg a e ESG in o ma ke di e en ia ion s a egies.
Ru al comme cial banks, wi h hei communi y ocus, should emphasize social ESG
dimensions such as inancial inclusion, digi al li e acy, and g een mic o inance.
5.3 Fu u e esea ch di ec ions
Fu u e s udies could es he gene alizabili y o ou indings ac oss di e en egula o y
en i onmen s, pa icula ly by compa ing China’s s a e-d i en model wi h Wes e n ma ke -
based sys ems. C oss-coun y analyses wi h di e se ESG amewo ks would enhance he
unde s anding o global ESG p o i abili y dynamics. Addi ionally, wi h he g owing in luence
o digi al banking and in ech, esea ch should explo e how hese inno a ions in e ac wi h
ESG s a egies and in luence i m sus ainabili y and p o i abili y. Fu he , his s udy elied on
CNRDS and CSMAR da abases, which, while au ho i a i e o Chinese lis ed i ms, a e
ocused on domes ic amewo ks. This may limi gene alizabili y o o he ma ke s o unlis ed
i ms, and CNRDS’s ESG me ics may di e om global s anda ds. To add ess his, u u e
esea ch could d aw on in e na ional ESG da ase s such as Re ini i , Bloombe g, o MSCI o
enhance compa abili y and ex e nal alidi y. Finally, u u e wo k could adop Bayesian
analy ical app oaches as an al e na i e o equen is me hods, allowing o he in eg a ion o
p io knowledge and o e ing a mo e nuanced in e p e a ion o unce ain y in ESG-
pe o mance ela ionships.
Asian Jou nal o
Economics and
Banking
387
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/3/364/10281595/ajeb-11-2024-0128en.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
Zhou, F. and Wen, H. (2022), “T ade policy unce ain y, de elopmen s a egy, and expo beha io :
e idence om lis ed indus ial companies in China”, Jou nal o Asian Economics, Vol. 82,
101528, doi: 10.1016/j.asieco.2022.101528.
Zhu, N., A yee, E.N.T., Agyemang, A.O., Wi edu, I., Zaka i, A. and Agbadzidah, S.Y. (2024),
“Add essing en i onmen , social and go e nance (ESG) in es men in China: does boa d
composi ion and inancing decision ma e ?”, Heliyon, Vol. 10 No. 10, e30783, doi: 10.1016/
j.heliyon.2024.e30783.
Zimon, G., A ianpoo , A. and Salehi, M. (2022), “Sus ainabili y epo ing and co po a e epu a ion: he
mode a ing e ec o CEO oppo unis ic beha io ”, Sus ainabili y, Vol. 14 No. 3, p. 1257, doi:
10.3390/su14031257.
Zumen e, I. and Bis o a, J. (2021), “ESG impo ance o long- e m sha eholde alue c ea ion:
li e a u e s P ac ice”, Jou nal o Open Inno a ion: Technology, Ma ke , and Complexi y, Vol. 7
No. 2, p. 127, doi: 10.3390/joi mc7020127.
Fu he eading
A ellano, M. and Bond, S. (1991), “Some es s o speci ica ion o panel da a: Mon e Ca lo e idence
and an applica ion o employmen equa ions”, The Re iew o Economic S udies, Vol. 58 No. 2,
p. 277, doi: 10.2307/2297968.
Co esponding au ho
Joel Vic o Dossa can be con ac ed a : [email p o ec ed]
Fo ins uc ions on how o o de ep in s o his a icle, please isi ou websi e:
www.eme aldg ouppublishing.com/licensing/ ep in s.h m
O con ac us o u he de ails: [email p o ec ed]
AJEB
9,3
394
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/3/364/10281595/ajeb-11-2024-0128en.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
