Decarbonizing agriculture: The impact of trade and renewable energy on CO₂ emissions

Öz ü k, Nil Si el
A icle
Deca bonizing ag icul u e: The impac o ade and
enewable ene gy on CO₂ emissions
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: Öz ü k, Nil Si el (2025) : Deca bonizing ag icul u e: The impac o ade and
enewable ene gy on CO₂ emissions, Economies, ISSN 2227-7099, MDPI, Basel, Vol. 13, Iss. 6, pp.
1-17,
h ps://doi.o g/10.3390/economies13060162
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/329442
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 : Sanzidu Rahman
Recei ed: 23 Ap il 2025
Re ised: 27 May 2025
Accep ed: 2 June 2025
Published: 6 June 2025
Ci a ion: Si el Öz ü k, N. (2025).
Deca bonizing Ag icul u e: The
Impac o T ade and Renewable
Ene gy on CO2Emissions. Economies,
13(6), 162. h ps://doi.o g/10.3390/
economies13060162
Copy igh : © 2025 by he au ho .
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
Deca bonizing Ag icul u e: The Impac o T ade and Renewable
Ene gy on CO2Emissions
Nil Si el Öz ü k
Depa men o Cus oms Managemen , Ke¸san Yusu Çap az School o Applied Sciences, T akya Uni e si y, Ke¸san,
Edi ne 22880, Tu key; [email p o ec ed]
Abs ac : This s udy in es iga es he en i onmen al e ec s o ag icul u al ade, enew-
able ene gy use, and economic g ow h in a panel o 14 selec ed coun ies o he pe iod
2000–2021. Pe capi a CO
2
emissions a e modeled as he dependen a iable using a
second-gene a ion panel da a me hod, he Augmen ed Mean G oup (AMG) es ima o ,
which accoun s o c oss-sec ional dependence and slope he e ogenei y. The analysis
e eals ha he sha e o enewable ene gy in o al ene gy consump ion signi ican ly e-
duces ca bon emissions, emphasizing he ole o g een ene gy policies in en i onmen al
imp o emen . In con as , economic g ow h is ound o inc ease emissions, indica ing he
alidi y o only he ini ial phase o he En i onmen al Kuzne s Cu e (EKC) hypo hesis.
Addi ionally, ag icul u al impo s—and in ce ain cases, expo s—exe upwa d p essu e
on emissions, likely due o logis ics and p oduc ion- ela ed ex e nali ies embedded in he
ade p ocess. G oup-speci ic esul s highligh dis inc dynamics ac oss coun ies: while
enewable ene gy adop ion plays a s onge ole in emission mi iga ion in de eloping
economies, ade composi ion and p oduc ion echnology d i e en i onmen al ou comes
in de eloped ones. The indings unde sco e he need o edesign ade and ene gy s a e-
gies wi h explici conside a ion o en i onmen al ex e nali ies o align wi h long- e m
sus ainabili y objec i es.
Keywo ds: ag icul u al ade; enewable ene gy; CO2emissions
JEL Classi ica ion: Q56; F18; O44
1. In oduc ion and Li e a u e Re iew
Global economic g ow h has accele a ed since he second hal o he 20 h cen u y,
la gely d i en by he expansion o in e na ional ade. Howe e , his g ow h has also
imposed signi ican cos s on en i onmen al sus ainabili y. In pa icula , he ise in en i on-
men al awa eness since he 1990s has led o a subs an ial body o heo e ical and empi ical
esea ch examining he en i onmen al implica ions o economic ac i i y. Wi hin his con-
ex , he ela ionship be ween economic g ow h, in e na ional ade, and en i onmen al
quali y has become a cen al ocus in shaping sus ainable de elopmen policies (S e n,2004;
An weile e al.,2001).
A he co e o hese discussions lies he ecological economics pe spec i e, which
concep ualizes he economy as a subsys em o he na u al en i onmen and a gues ha
economic ac i i y mus ope a e wi hin ecological limi s. Acco ding o his app oach,
con en ional g ow h models exe i e e sible p essu es on na u e and, in he long e m,
pose isks o bo h en i onmen al and social well-being (Cos anza e al.,1997;Daly,1997).
Ecological economics ad oca es o a sys emic e alua ion o he en i onmen al cos s
associa ed wi h ene gy consump ion, p oduc ion s uc u es, and in e na ional ade.
Economies 2025,13, 162 h ps://doi.o g/10.3390/economies13060162
Economies 2025,13, 162 2 o 17
One o he mos widely ci ed app oaches in he en i onmen al economics li e a u e o
explaining he ela ionship be ween economic ac i i y and en i onmen al deg ada ion is
he En i onmen al Kuzne s Cu e (EKC) hypo hesis. The EKC posi s an in e ed U-shaped
ela ionship be ween pe capi a income and en i onmen al deg ada ion. A ea ly s ages
o economic g ow h, en i onmen al ha m inc eases; howe e , beyond a ce ain income
h eshold, his end is expec ed o e e se due o g ea e en i onmen al awa eness, he
adop ion o cleane echnologies, and he implemen a ion o egula o y policies (G ossman
& K uege ,1995;Dinda,2004). Ne e heless, he uni e sal alidi y o his hypo hesis ac oss
coun ies and sec o s emains subjec o deba e.
Ano he ac o ha adds complexi y o he ela ionship be ween ade, g ow h, and
he en i onmen is he di ec ion and na u e o ade’s en i onmen al impac s. While con-
en ional heo y sugges s ha ade can imp o e en i onmen al e iciency h ough he
op imal alloca ion o p oduc ion ac o s, mechanisms such as he “pollu ion ha en” and
“scale e ec ” imply po en ial nega i e consequences (An weile e al.,2001). Analyzing
he en i onmen al e ec s o ade wi hou sec o al dis inc ions may obscu e hese impac s.
In pa icula , ag icul u al ade can con ibu e signi ican ly o ca bon emissions du ing
bo h p oduc ion and anspo a ion phases (Ve bu g e al.,2011). Acco ding o IEA (2021),
emissions om p e- and pos -p oduc ion p ocesses in ag i ood sys ems—including ans-
po a ion and logis ics—amoun ed o 5.8 G CO
2
e in 2019, ep esen ing app oxima ely 35%
o o al ag i ood sys em emissions (FAO Global,2022).
Al hough ag icul u e is o en pe cei ed as an en i onmen ally iendly sec o , he
expansion o global supply chains and he in e na ional anspo a ion o ag icul u al
p oduc s ha e signi ican ly inc eased CO
2
emissions, pa icula ly om logis ics. The ca bon
oo p in o global ag icul u al ade highligh s he necessi y o de eloping coun ies
o align hei ade s a egies wi h sus ainable de elopmen goals (Kas ne e al.,2012).
The e o e, esea ch on he en i onmen al impac s o ag icul u al ade p o ides aluable
insigh s no only o economic analysis bu also o en i onmen al policymaking.
On he o he hand, he use o enewable ene gy has ecen ly eme ged as one o
he mos c i ical policy ins umen s o mi iga ing he en i onmen al cos s o economic
g ow h. Inc easing he sha e o enewables in o al ene gy consump ion plays a pi o al
ole in educing CO
2
emissions (IEA,2021). In his con ex , i is e iden ha coun ies’
g ow h s a egies and ade policies mus be conside ed in close in eg a ion wi h ene gy
ansi ion e o s.
This s udy analyzes he en i onmen al e ec s o ag icul u al ade (impo s and
expo s), enewable ene gy use, and economic g ow h using a panel da ase co e ing
14 coun ies om 2000 o 2021. I s main con ibu ion lies in simul aneously add essing
bo h he classical g ow h–en i onmen ela ionship and he ade dimension h ough he
ag icul u al sec o . To accoun o issues speci ic o panel da a, such as c oss-sec ional
dependence and slope he e ogenei y, he Augmen ed Mean G oup (AMG) es ima o is
employed, enabling a conside a ion o coun y-speci ic dynamics. Acco dingly, he s udy
aims o o e a mo e comp ehensi e pe spec i e on he heo e ical and empi ical links
be ween en i onmen al ou comes and in e na ional ade in ag icul u al p oduc s.
The ole o he ag icul u al sec o in ca bon emissions has a ac ed inc easing aca-
demic in e es in ecen yea s. In pa icula , he e ec s o dynamics such as in e na ional
ade, p oduc ion s uc u es, and ene gy use on ag icul u e- ela ed g eenhouse gas emis-
sions ha e been examined in a mul idimensional manne ac oss a ious egions. In his
con ex , he indings o his s udy align wi h he ecen li e a u e, which highligh s bo h
he mi iga ing impac o enewable ene gy use on emissions and he in luen ial ole o
ag icul u al ade in shaping en i onmen al ou comes.
Economies 2025,13, 162 3 o 17
The impac o ag icul u al ade on ca bon emissions has been empi ically e alua ed
in nume ous s udies. Fo ins ance, W. Wang e al. (2024) ind ha ade openness in ag i-
cul u e con ibu es o emission educ ions, shaped h ough channels such as scale e ec s,
echnological ad ancemen , and s uc u al ans o ma ion. Simila ly, G. Li e al. (2024)
show ha ade libe aliza ion educes emission in ensi y, suppo ed by echnology di u-
sion and shi s in indus ial composi ion. This line o e idence complemen s he p esen
s udy’s ocus on he en i onmen al e ec s o ag icul u al ade and enewable ene gy.
The en i onmen al impac s o ag icul u al ade ha e been examined om mul iple
pe spec i es. In a ecen e iew o he pas wo decades o esea ch, P. Wang e al. (2023)
a gue ha he en i onmen al e ec s o ade libe aliza ion a e shaped by channels such as
scale, s uc u al ans o ma ion, anspo a ion, and echnological change, o en esul ing
in nega i e ou comes. Sugges ed policy esponses include imp o emen s in ac o allo-
ca ion, policy e o ms, echnological inno a ion, and he de elopmen o compensa o y
mechanisms. This heo e ical pe spec i e echoes he empi ical obse a ions made in he
p esen s udy conce ning he en i onmen al impac o ag icul u al impo s.
The e is g owing e idence ha he en i onmen al impac s o ag icul u al ade a y
depending on ac o s such as egional cha ac e is ics and policy h esholds. Rong e al.
(2023) ind ha in China, ag icul u al ade openness can educe emissions only when
en i onmen al egula ion su passes a ce ain h eshold; below ha le el, he e ec s a e
e e sed. This pa e n is consis en wi h he indings o he p esen s udy, whe e ag icul-
u al impo s exhibi posi i e e ec s on emissions in some coun ies and nega i e e ec s
in o he s. Mo eo e , he h eshold-based panel model used by Rong e al. o e s a plau-
sible explana ion o he he e ogeneous coe icien s iden i ied in ou second-gene a ion
panel analysis.
The spa ial and sec o al dimensions o ca bon ans e s d i en by ag icul u al ade a e
ecei ing inc easing a en ion in he li e a u e. analyze how ag icul u al ca bon emissions
a e edis ibu ed ac oss Chinese p o inces h ough ade, and how his edis ibu ion
in luences policy accoun abili y. Simila ly, Liu e al. (2024) model in e na ional ca bon
lows in ag icul u al ade using ne wo k-based s uc u es, p oposing bo h p oduc ion- and
consump ion-o ien ed s a egies. In his con ex , he p esen s udy’s use o coun y-speci ic
coe icien s h ough he AMG es ima o aligns me hodologically wi h his s and o he
li e a u e by add essing c oss-coun y he e ogenei y in ca bon ou comes.
The in e naliza ion o en i onmen al ex e nali ies in o ade policies has gained p omi-
nence wi h he eme gence o ins umen s such as he Ca bon Bo de Adjus men Mech-
anism (CBAM), which a e ele an o bo h de eloped and de eloping coun ies. Bux
e al. (2024) and Bassi e al. (2024) discuss he po en ial o he CBAM o p e en ca bon
leakage, as well as i s une en impac s on de eloping economies. Fou nie Gabela e al.
(2024) p opose a CBAM amewo k speci ically ailo ed o he ag icul u al sec o , analyzing
how such a policy could be made mo e easible. In his ega d, he policy implica ions
o he p esen s udy esona e wi h cu en deba es on balancing he emission-inc easing
e ec s o ade in de eloping coun ies h ough mechanisms like he CBAM.
The ole o g een echnological inno a ion in educing ag icul u al emissions has been
emphasized in nume ous ecen s udies. Rong e al. (2023) ind ha g een inno a ion exe s
bo h di ec and spa ially media ed mi iga ing e ec s on ag icul u al ca bon emission in en-
si y. L. Zhang and Cai (2024) iden i y an in e ed U-shaped ela ionship, sugges ing ha
main aining echnology a an op imal le el can yield bo h en i onmen al and p oduc i i y
bene i s. Qayyum e al. (2023) and Huang and Ke (2024) highligh egional dispa i ies in
he adop ion o g een inno a ion and e eal he indi ec e ec s o digi aliza ion and o ga-
niza ional lea ning on ag icul u al emissions. The p esen s udy’s inding o a signi ican
Economies 2025,13, 162 4 o 17
nega i e ela ionship be ween enewable ene gy use and CO
2
emissions is in line wi h
hese conclusions.
Ca bon p icing, en i onmen al axa ion, and ca bon seques a ion a e also widely
discussed in he li e a u e as di ec mi iga ion s a egies. Iyke-O oedu e al. (2024) em-
phasize he impac o en i onmen al axes on ca bon seques a ion in Sou h A ica, while
Gong and Huo (2024) a gue ha ca bon axes alone a e insu icien o educing ag icul-
u al g eenhouse gas emissions and mus be complemen ed wi h seques a ion s a egies.
Kausa e al. (2024) iden i y a mul idimensional ela ionship be ween ag icul u al p oduc-
ion and en i onmen al axa ion, sugges ing ha ax-based solu ions a e mo e sus ainable
han di ec p oduc ion cons ain s. These s udies p o ide a concep ual ounda ion o he
policy ecommenda ions made in he conclusion o he p esen esea ch ega ding ene gy
ansi ion and en i onmen ally conscious ade s a egies.
The ela ionship be ween ag icul u al p oduc ion and ca bon emissions also aises he
issue o balancing p oduc i i y and en i onmen al sus ainabili y. (X. Zhang e al.,2024)
epo a simul aneous inc ease in p oduc i i y and emissions, highligh ing he need o
a balanced applica ion o g een echnologies Acco si e al. (2023) sugges ha inno a i e
p oduc ion echniques and digi al in as uc u e can suppo his balance in a sus ainable
way. In his con ex , he p esen s udy also inds ha while ag icul u al ade con ibu es o
ising emissions, he ene gy ansi ion may help o se his impac .
In e na ional ag icul u al ade is a key d i e o no only domes ic bu also c oss-
bo de g eenhouse gas ans e s. (Adenaue e al.,2025) show ha global ag icul u al
emissions sp ead h ough ne wo k s uc u es, making imp o emen s in p oduc ion ech-
nologies and consump ion s a egies e ec i e a bo h local and global scales. In his con ex ,
he coun y-speci ic e ec s iden i ied h ough he AMG model in his s udy may e lec such
s uc u al in e dependencies. Mo eo e , echnological ans o ma ion no only educes
emissions bu also gene a es ansna ional spillo e e ec s.
A limi ed bu g owing numbe o s udies ha e examined he ela ionship be ween
ag icul u al ade and ca bon emissions in na ional o egional con ex s. Fo ins ance,
Tu an (2025) analyzes he long- un impac o ag icul u al expo s on CO
2
emissions in
Tü kiye using ARDL models and inds ha expo -led g ow h can educe emissions, pa -
icula ly when suppo ed by enewable ene gy adop ion. In a mo e de ailed egional
s udy,
Q. Li and Zhang (2024)
in es iga e he edis ibu ion o ag icul u al ca bon emis-
sions ac oss Chinese p o inces and show ha bo h impo and expo ac i i ies can con-
ibu e o emission educ ion h ough s uc u al ans o ma ion and echnology di usion.
These indings highligh he signi icance o na ional condi ions, p oduc ion models, and
ade s uc u es in de e mining he en i onmen al ou comes o ag icul u al ade.
Se e al s udies sugges ha he en i onmen al e ec i eness o enewable ene gy in
India may be cons ained by s uc u al and policy limi a ions. Singh e al. (2023) highligh
he coun y’s hea y eliance on biomass and he une en de elopmen o enewable sec o s,
which may unde mine emission educ ions. Simila ly, Dubey e al. (2023) no e ha
challenges such as g id ins abili y, policy unce ain y, and luc ua ing sola a i s a ec he
en i onmen al ou comes o enewable ene gy deploymen . These insigh s a e ele an in
unde s anding why enewable ene gy use in India may no ye yield consis en emission-
educing e ec s.
Finally, a a ie y o me hodological app oaches ha e been employed in ecen s udies,
including panel ARDL, AMG, GMM, panel NARDL, spa ial Du bin models, and QARDL
echniques (e.g., W. Wang e al.,2024;Kausa e al.,2024;L. Zhang & Cai,2024). By using
he AMG es ima o , which accoun s o slope he e ogenei y and c oss-sec ional dependence,
his s udy also con ibu es me hodologically o he exis ing li e a u e.

Economies 2025,13, 162 5 o 17
In his s udy, ag icul u al ade is examined h ough he disagg ega ed e ec s o
impo s and expo s, measu ed in cu en USD, while enewable ene gy use is de ined as
he sha e o enewables in o al ene gy consump ion. De ailed de ini ions, measu emen
uni s, and da a sou ces o all a iables a e p o ided in Table 1.
Table 1. Va iables used in he s udy.
Va iable Code Desc ip ion Uni Sou ce
CO2Emission CO2emissions pe capi a
Me ic ons ( ons/capi a)
WB
ln_GDPCapi a Real GDP pe capi a USD
ln_Ag iIM Ag icul u al impo s Billion USD
ln_Ag iEX Ag icul u al expo s Billion USD
RenewableE Sha e o enewables in o al
ene gy ea ing Pe cen age (%)
No e: Va iables p eceded by “ln” indica e ha he na u al loga i hm o he o iginal alues has been aken.
In his con ex , ecen app oaches and empi ical indings in he li e a u e la gely align
wi h he esul s o his s udy, emphasizing he need o examine he complex ela ionship
be ween ag icul u al ade and en i onmen al sus ainabili y om a mul idimensional pe -
spec i e. The li e a u e sugges s ha he in e ac ions among ene gy ansi ion, echnology
policies, and ade egula ions a e pa icula ly impo an o ad ancing ca bon-neu al
ag icul u al s a egies.
2. Analysis and Findings
2.1. Coun y and Va iable Selec ion
This s udy analyzes a g oup o 14 coun ies ha includes bo h de eloped and de el-
oping economies: he Uni ed S a es, he Uni ed Kingdom, Aus alia, Canada, Ge many,
F ance, Denma k, China, India, Indonesia, Russia, B azil, Tü kiye, and Mexico. These
coun ies a e no only key playe s in global ag icul u al ade bu also exhibi di e se
cha ac e is ics in e ms o ene gy consump ion, g eenhouse gas emissions, and economic
g ow h dynamics. This di e si y enables a compa a i e analysis o he ela ionship be ween
ag icul u al ade and ca bon emissions ac oss coun ies a di e en s ages o de elopmen .
Coun y selec ion also conside ed da a a ailabili y and he consis ency o he obse a-
ion pe iod. The analysis co e s he yea s 2000 o 2021, a ime ame chosen o o e ing a
su icien ly long obse a ion window while also encompassing bo h he p e- and pos -Pa is
Ag eemen pe iods. This allows o he indi ec obse a ion o he e ec s o in e na ional
en i onmen al egimes and g een de elopmen policies.
The a iables used in his s udy a e selec ed om key economic and sec o al indica o s
commonly employed in he li e a u e o assess en i onmen al impac s. The dependen
a iable is pe capi a ca bon dioxide (CO
2
) emissions. Al hough no exclusi e o he
ag icul u al sec o , his a iable is widely used as a mac o-le el indica o o en i onmen al
impac and se es as a p oxy o o e all na ional g eenhouse gas ends.
The independen a iables a e de ined as ollows. Ag icul u al impo s e lec he
deg ee o ade openness in he ag icul u al sec o and he global logis ics ac i i y wi hin
he ood supply chain. Ag icul u al expo s ep esen he en i onmen al bu den associa ed
wi h p oduc ion-based ade and a e ele an in he con ex o po en ial ca bon leakage.
Pe capi a GDP is included o examine he en i onmen al impac o economic g ow h,
pa icula ly in ela ion o he En i onmen al Kuzne s Cu e (EKC) hypo hesis. Renewable
ene gy use e e s o he sha e o enewables in o al ene gy consump ion and se es as a
key indica o o assessing he en i onmen al e ec s o sus ainable ene gy ansi ions.
The selec ion o a iables in his s udy is b oadly in o med by he IPAT iden i y
(I = P
×
A
×
T) and i s s ochas ic ex ension, he STIRPAT model. These amewo ks
Economies 2025,13, 162 6 o 17
concep ualize en i onmen al impac (I) as a unc ion o popula ion (P), a luence (A), and
echnology (T). Acco dingly, pe capi a GDP is used as a p oxy o a luence, while enew-
able ene gy sha e ep esen s he echnological componen . T ade- ela ed a iables cap u e
he p oduc ion and s uc u al aspec s o ag icul u al sys ems ha in luence emissions. The
inclusion o hese indica o s allows o a mac o-le el in e p e a ion o d i e s o ca bon
emissions, consis en wi h en i onmen al economic heo y.
Rega ding da a ans o ma ion, loga i hmic o ms we e applied selec i ely based on
dis ibu ional cha ac e is ics and in e p e abili y. Speci ically, a iables wi h high skew-
ness and wide magni ude anges (such as GDP pe capi a and CO
2
emissions) we e
log- ans o med o ensu e linea i y and educe he e oskedas ici y. Va iables exp essed in
pe cen age o m o bounded be ween 0 and 100 (such as enewable ene gy sha e) we e
e ained in le el o m o p ese e hei scale in e p e abili y and a oid dis o ion. This
ans o ma ion s a egy aligns wi h common p ac ices in empi ical en i onmen al s udies.
De ailed in o ma ion on hese a iables is p esen ed in Table 1.
2.2. C oss-Sec ional Dependence (CD Tes )
The CD es de eloped by Pesa an (2005) was applied o examine c oss-sec ional
dependence in he panel da ase . This es is designed o de ec he exis ence o dependence
among c oss-sec ional uni s—such as coun ies in a panel—due o common shocks o
sha ed dynamics. I is pa icula ly use ul in da ase s wi h a la ge ime dimension and is
widely applied o examine whe he obse a ions a e co ela ed ac oss uni s. The hypo heses
o he es a e de ined as ollows (Pesa an,2015):
H0 (Null Hypo hesis): The e is weak c oss-sec ional dependence among he a iables; o example,
a coun y’s impo le els a e independen o hose in o he coun ies.
H1 (Al e na i e Hypo hesis): The e is s ong c oss-sec ional dependence; ha is, impo beha io
ac oss coun ies is mu ually in luenced.
The s a is ical o mula ion o he es is based on he a e age pai wise co ela ion
coe icien s among panel uni s and is calcula ed as ollows:
CD =√N
√2T
N−1
∑
i=1
N
∑
j=i+1
ˆ
Pij
In he equa ion, N ep esen s he numbe o coun ies in he panel, and Tdeno es he
ime dimension. The e m
ˆ
Pij
e e s o he co ela ion coe icien be ween he a iables o
coun y iand coun y j. The p- alue ob ained om he CD es de e mines whe he he null
hypo hesis (H
0
) can be ejec ed a con en ional signi icance le els (e.g., 1%, 5%, o 10%). I
he p- alue is below 0.10, H
0
is ejec ed, indica ing he p esence o s ong c oss-sec ional
dependence among he a iables (Pesa an,2005).
The esul s o he CD es a e p esen ed in Table 2.
As shown in Table 2, he CD and CDw
+
s a is ics a e s a is ically signi ican (p< 0.01)
o mos o he a iables. This indica es he p esence o c oss-sec ional dependence among
coun ies, pa icula ly in a iables such as ag icul u al ade (ln_Ag iIM, ln_Ag iEX), pe
capi a income (ln_GDPCapi a), and CO
2
emissions. Al hough he CDw and CD* es s a e
no signi ican o all a iables, he consis en signi icance o he CDw
+
es —which has
g ea e s a is ical powe —sugges s ha con en ional i s -gene a ion panel echniques may
be inadequa e. In his con ex , ex e nal shocks, global economic in eg a ion, and common
en i onmen al policies may con ibu e o he o ma ion o sha ed dependence s uc u es
ac oss coun ies.
Economies 2025,13, 162 7 o 17
Table 2. CD es esul s.
Va iable CD p-Value CDw p-Value CDw+p-Value CD * p-Value
ln_Ag iIM 36.69 (0.000) *** 0.20 (0.839) 350.22 (0.000) *** −2.20 (0.028) **
ln_GDPCapi a 39.31 (0.000) *** 0.17 (0.865) 375.17 (0.000) *** 1.51 (0.132)
ln_Ag iEX 34.66 (0.000) *** −0.20 (0.839) 330.43 (0.000) *** 0.41 (0.683)
CO2Emissions 1.66 (0.097) 1.02 (0.309) 312.34 (0.000) *** 5.65 (0.000) ***
RenewableE 3.09 (0.002) ** 7.09 (0.000) *** 260.30 (0.000) *** −0.33 (0.739)
No e: ***, **, and * deno e s a is ical signi icance a he 1%, 5%, and 10% le els, espec i ely. CDw: Weigh ed
c oss-sec ional dependence es , CDw
+
: Enhanced es wi h s onge s a is ical powe , CD: Bias-co ec ed s a is ic
based on (Pesa an,2015).
2.3. Slope Homogenei y (He e ogenei y) Tes
The slope homogenei y es de eloped by Pesa an and Yamaga a (2008), also known
as he del a es , is used o de e mine whe he he eg ession coe icien s ac oss panel uni s
a e simila . This me hod es s he alidi y o a common coe icien ac oss he en i e panel.
The co e idea is o s a is ically assess how much each uni ’s indi idual coe icien de ia es
om he o e all a e age. In doing so, i helps de e mine whe he pa ame e he e ogenei y
should be conside ed in panel da a modeling.
The del a es is exp essed h ough he ollowing o mulas:
1. S anda d Del a Tes : ∆=qN
21
N∑N
i=1ˆ
βi−β
2. Augmen ed Del a Tes (Del a Tilde): ∼
∆=√N1
N∑N
i=1
ˆ
βi−β
σi
In hese o mulas, N ep esen s he numbe o panel uni s,
ˆ
βi
, is he es ima ed slope
coe icien o uni
i
,
β
is he a e age slope coe icien ac oss all uni s, and
σi
deno es he
s anda d e o o he es ima ed coe icien o uni i.
The esul s o he slope homogenei y es a e p esen ed in Table 3.
Table 3. Slope homogenei y es esul s.
Tes Type S a is ic p-Value
Del a Tes 11.466 (0.000) ***
Adjus ed Del a Tes 13.445 (0.000) ***
No e: Unde he null hypo hesis o slope homogenei y, bo h he Del a and Adjus ed Del a es s a e asymp o ically
no mally dis ibu ed. The p- alues indica e signi icance a he 1% le el (*** p< 0.01).
The slope homogenei y es esul s p esen ed in Table 3(Pesa an & Yamaga a,2008)
a e highly signi ican based on bo h he Del a and Adjus ed Del a s a is ics (p< 0.01). This
leads o he ejec ion o he null hypo hesis, which assumes ha slope coe icien s a e
homogeneous ac oss all coun ies.
The indings sugges ha he impac o he independen a iables in he model—such
as ag icul u al impo s, expo s, pe capi a income, and he sha e o enewable ene gy—on
he dependen a iable (CO
2
emissions) di e s ac oss coun ies, indica ing he p esence o
slope he e ogenei y.
2.4. Uni Roo Tes
The Augmen ed Dickey–Fulle es de eloped by Pesa an (2007) o e s a uni oo
es ing app oach ha accoun s o c oss-sec ional dependence in panel da a se ings. Known
as he C oss-sec ionally Augmen ed Dickey–Fulle (CADF) es , his me hod imp o es
he eliabili y and ealism o uni oo es ing by inco po a ing in e dependencies among
coun ies o panel uni ac o s o en neglec ed in adi ional panel uni oo es s. As a
esul , i p o ides a mo e obus assessmen o s a iona i y, pa icula ly o ime se ies
in luenced by common shocks o simila ends ac oss coun ies.
Economies 2025,13, 162 8 o 17
The s anda d o m o he Augmen ed Dickey–Fulle (ADF) es o panel da a is
exp essed as ollows:
∆yi =αi+βiyi +∑pi
k=1γik∆yi −k+ϵi
In his o mula ion, Y
i
, ep esen s he obse a ion o uni ia ime ,
αi
, is he
indi idual ixed e ec ,
βi
, is he coe icien o he lagged le el e m ( he speed o adjus men ),
γik, a e he coe icien s o he lagged di e ences, and ϵi , is he e o e m.
Pesa an’s CADF es inco po a es c oss-sec ional dependence and is speci ied
as ollows:
∆yi =αi+βiyi +∑pi
k=1γik∆yi −k+ϵi +δy −1+ϵi
He e, y
−1
, deno es he c oss-sec ional mean a ime
−
1, which cap u es common
ac o s ac oss uni s in he panel.
The esul s o he uni oo es a e p esen ed in Table 4.
Table 4. Pesa an CADF uni oo es esul s.
Va iable S a iona i y Le el -Ba C i ical Value
(5%) Z[ -Ba ] p-Value Conclusion
ln_Ag iIM Le el [I(0)] −2.309 −2.250 −2.079 0.019 ** S a iona y
ln_GDPCapi a Fi s Di e ence [I(1)] −2.606 −2.250 −3.225 0.001 *** S a iona y
ln_Ag iEX Fi s Di e ence [I(1)] −2.972 −2.250 −4.635 0.000 *** S a iona y
CO2Emissions Fi s Di e ence [I(1)] −2.474 −2.250 −2.715 0.003 *** S a iona y
Renewable Ene gy Fi s Di e ence [I(1)] −3.250 −2.250 −5.709 0.000 ** S a iona y
No e: ***, and ** deno e s a is ical signi icance a he 1%, and 5% le els, espec i ely.
Acco ding o he esul s o he panel CADF es de eloped by Pesa an (2007), only
he a iable ln_Ag iIM is s a iona y a le el [I(0)], while he emaining a iables become
s a iona y a hei i s di e ences [I(1)]. This indica es ha he a iables exhibi di e en
o de s o in eg a ion, which limi s he applicabili y o con en ional panel models assuming
ixed coe icien s.
The e o e, second-gene a ion panel es ima ion me hods—such as Panel ARDL s uc-
u es, AMG, o CCE es ima o s—a e mo e app op ia e, as hey allow o a combina ion o
I(0) and I(1) a iables.
2.5. Augmen ed Mean G oup (AMG) Es ima ion Resul s
The Augmen ed Mean G oup (AMG) es ima o , de eloped by Ebe ha d and
Bond (2009) and Ebe ha d and Teal (2010), is designed o es ima e long- un ela ion-
ships in he e ogeneous panel da ase s. This app oach ex ends he con en ional Mean
G oup (MG) es ima o by inco po a ing a common dynamic p ocess ha accoun s o
c oss-sec ional dependence.
In a panel da a model, whe e y
i
deno es he dependen a iable and x
i
is a ec o o
explana o y a iables, he gene al speci ica ion is gi en as ollows:
yi =αi+β′
ixi +ui
He e, i= 1,
. . .
,Ndeno es he coun ies, and = 1,
. . .
,T ep esen s he ime pe iods.
αi
cap u es he coun y-speci ic ixed e ec s, while
βi
is a k
×
1 ec o o coe icien s o each
coun y. ui deno es he e o e m.
To accoun o c oss-sec ional dependence, he e o e m ui is modeled as ollows:
ui =λi +εi
Economies 2025,13, 162 15 o 17
Ag icul u al expo s, by con as , do no ha e a signi ican e ec in he o e all model,
bu coun y-speci ic esul s o e use ul insigh s. In B azil and Russia, expo s a e associa ed
wi h highe emissions, possibly due o la ge-scale, indus ialized a ming. The Uni ed
S a es p esen s a con as ing case, whe e expo s appea o educe emissions, po en ially
due o ad anced echnologies and esou ce-e icien p ac ices.
Coun y-le el esul s sugges ha he impac o g ow h, ade, and ene gy a iables
di e s meaning ully ac oss na ional con ex s. Fo ins ance, in Russia, pe capi a income
is nega i ely associa ed wi h emissions—hin ing a he possibili y o decoupling be ween
economic g ow h and en i onmen al ha m. Likewise, coun ies such as Tü kiye, B azil,
and Mexico exhibi s a is ically signi ican emission- educing e ec s om enewable ene gy
use, indica ing he e ec i eness o cu en ene gy policies.
These a ia ions ein o ce he conclusion ha uni e sal policy ecommenda ions
may no be e ec i e. Ins ead, coun y-speci ic s a egies ha conside s uc u al and
sec o al dynamics a e likely o yield be e ou comes. The s udy emphasizes he alue o
disagg ega ed en i onmen al analysis o bo h academic esea ch and policy design.
The analysis also demons a es ha ag icul u al impo s and enewable ene gy use
ha e p edic i e powe o e emission le els, as con i med by G ange causali y es s. This
adds obus ness o he empi ical indings and s eng hens he policy ele ance o he model.
Add essing he emission in ensi y o ag icul u al impo s will equi e s a egies such
as ca bon ce i ica ion schemes, sus ainable sou cing s anda ds, and g eene supply chains.
In coun ies whe e enewable ene gy con ibu es o emission educ ion, ene gy ansi ion
should be deepened h ough be e in as uc u e and e iciency gains. Whe e enewables
a e no ye deli e ing emission educ ions, s uc u al ba ie s mus be add essed o unlock
hei po en ial.
In conclusion, his s udy in eg a es ade, ene gy, and g ow h in o a comp ehensi e
amewo k o explain c oss-coun y a ia ion in ca bon emissions. The esul s con i m
ha sus ainable de elopmen equi es mo e han economic expansion—i necessi a es a
undamen al e hinking o how ade and ene gy sys ems in e ac wi h he en i onmen .
Fu u e wo k may bene i om es ing nonlinea g ow h e ec s, analyzing g een p oduc
ade, and inco po a ing mo e de ailed sec o al da a.
Funding: This esea ch ecei ed no ex e nal unding.
Ins i u ional Re iew Boa d S a emen : No applicable.
In o med Consen S a emen : No applicable.
Da a A ailabili y S a emen : The da a used in his s udy a e publicly a ailable om he Wo ld Bank
Open Da a pla o m: h ps://da a.wo ldbank.o g, accessed on 27 May 2025, as also s a ed in Table 1.
No new da a we e c ea ed.
Con lic s o In e es : The au ho decla es no con lic o in e es .
Appendix A
Table A1. Robus ness check: CCEMG es ima ion esul s.
Va iable Coe icien S d. E . p-Value
ln_GDPCapi a 387.82 233.14 0.096 *
ln_Ag iIM 227.65 94.89 0.016 **
ln_Ag iEX −20.02 55.44 0.718
RenewableE −30.64 17.46 0.079 *
**, and * indica e s a is ical signi icance a he 5%, and 10% le els, espec i ely.

Economies 2025,13, 162 16 o 17
Re e ences
Acco si, R., Akke man, R., Pawa , K. S., & Sasso, M. (2023). Sus ainable ood supply chains—Eme ging echnologies and s a egies
owa d low-ca bon ecosys ems. Jou nal o Cleane P oduc ion,413, 137393. [C ossRe ]
Adenaue , M., Adjemian, M. K., A i a, S., B o sen, W., Coope , J., Goh, G., Ka ali, B., Mallo y, M. L., Thompson, W., & Yu, J. (2025).
Impac s o unila e al U.S. ca bon policies on ag icul u al sec o g eenhouse gas emissions and commodi y ma ke s. En i onmen al
Resea ch Le e s,20(2), 024022. [C ossRe ]
An weile , W., Copeland, B. R., & Taylo , M. S. (2001). Is ee ade good o he en i onmen ? Ame ican Economic Re iew,91(4), 877–908.
[C ossRe ]
Bassi, A. M., Calciola i, F., Cos an ini, V., D’Angeli, M., & Paglialunga, E. (2024). Ca bon bo de adjus men s o clima e clubs: Impac s
on A ican ag icul u al sec o s unde di e en coope a i e scena ios. The Wo ld Economy,48(1), 119–152. [C ossRe ]
Bux, C., Rana, R. L., T icase, C., Gea i, P., & Lomba di, M. (2024). Ca bon bo de adjus men mechanism (CBAM) o ackle ca bon
leakage in he in e na ional e ilize ade. Sus ainabili y,16(23), 10661. [C ossRe ]
Cos anza, R., d’A ge, R., De G oo , R., Fa be , S., G asso, M., Hannon, B., Limbu g, K., Naeem, S., O’neill, R. V., & Pa uelo, J. (1997).
The alue o he wo ld’s ecosys em se ices and na u al capi al. Na u e,387(6630), 253–260. [C ossRe ]
Daly, H. E. (1997). Beyond g ow h: The economics o sus ainable de elopmen . Beacon P ess.
Dinda, S. (2004). En i onmen al kuzne s cu e hypo hesis: A su ey. Ecological Economics,49(4), 431–455. [C ossRe ]
Dubey, B., Ag awal, S., & Sha ma, A. K. (2023). India’s enewable ene gy po olio: An in es iga ion o he un apped po en ial o RE,
policies, and incen i es a o ing ene gy secu i y in he coun y. Ene gies,16(14), 5491. [C ossRe ]
Dumi escu, E.-I., & Hu lin, C. (2012). Tes ing o G ange non-causali y in he e ogeneous panels. Economic Modelling,29(4), 1450–1460.
[C ossRe ]
Ebe ha d , M., & Bond, S. (2009). C oss-sec ion dependence in nons a iona y panel models: A no el es ima o . MPRA.
Ebe ha d , M., & Teal, F. (2010). P oduc i i y analysis in global manu ac u ing p oduc ion. Uni e si y o Ox o d.
FAO Global. (2022). Regional and coun y ends, 2000–2020. FAO.
Fou nie Gabela, J. G., Spiegel, A., S epanyan, D., F eund, F., Banse, M., Goch , A., Söde , M., Heidecke, C., Os e bu g, B., & Ma hews,
A. (2024). Ca bon leakage in ag icul u e: When can a ca bon bo de adjus men mechanism help? Clima e Policy,24(10), 1410–1425.
[C ossRe ]
Gong, Z., & Huo, X. (2024). Assessing he impac o ca bon mi iga ion s a egies on ag icul u al GHG emissions: Insigh s om a
dynamic CGE model analysis. F on ie s in En i onmen al Science,12, 1424076. [C ossRe ]
G ossman, G. M., & K uege , A. B. (1995). Economic g ow h and he en i onmen . The Qua e ly Jou nal o Economics,110(2), 353–377.
[C ossRe ]
Huang, S., & Ke, C. (2024). Resea ch on dis up i e g een echnological inno a ion in ag icul u e d i en by low-ca bon ini ia i es.
Sus ainabili y,16(24), 11230. [C ossRe ]
IEA. (2021). Wo ld ene gy ou look 2021. IEA. A ailable online: h ps://www.iea.o g/ epo s/wo ld-ene gy-ou look-2021 (accessed on
5 Ap il 2025).
Iyke-O oedu, M. I., Takon, S. M., Ugwun a, D. O., Ezeaku, H. C., Nso o , E. S., & Egbo, O. P. (2024). Impac o CO
2
emissions embodied
in he ag icul u al sec o on ca bon seques a ion in Sou h A ica: The ole o en i onmen al axes and echnological inno a ion.
Jou nal o Cleane P oduc ion,444, 141210. [C ossRe ]
Kas ne , T., Ri as, M. J. I., Koch, W., & Nonhebel, S. (2012). Global changes in die s and he consequences o land equi emen s o
ood. P oceedings o he Na ional Academy o Sciences o he Uni ed S a es o Ame ica,109(18), 6868–6872. [C ossRe ]
Kausa , A., Rasul, F., & Asgha , N. (2024). How o ge g een wi h ag icul u al oo p in : A global analysis o ca bon emissions,
en i onmen al axes, and ag ochemical use. Jou nal o En i onmen al Managemen ,370, 122665. [C ossRe ]
Li, G., Huang, Y., Peng, L., You, J., & Meng, A. (2024). Ag icul u al ca bon educ ion in China: The syne gy e ec o ade and
echnology on sus ainable de elopmen . En i onmen al Resea ch,252, 119025. [C ossRe ] [PubMed]
Li, Q., & Zhang, X. (2024). E ec s o ag icul u al ade on educing ca bon emissions unde he “Dual Ca bon” Ta ge : E idence om
China. Ag icul u e,14(8), 1274. [C ossRe ]
Liu, K., Wang, P., Zhang, B., Yan, K., & Shao, L. (2024). Ne wo k s uc u es and mi iga ion po en ial o ade linked global ag icul u al
g eenhouse gas emissions. Scien i ic Repo s,14(1), 30973. [C ossRe ]
Pesa an, M. H. (2005). Gene al diagnos ic es s o c oss sec ion dependence in panels. Camb idge Wo king Pape s. Economics, 1240(1), 1.
[C ossRe ]
Pesa an, M. H. (2007). A simple panel uni oo es in he p esence o c oss-sec ion dependence. Jou nal o Applied Econome ics,
22(2), 265–312. [C ossRe ]
Pesa an, M. H. (2015). Tes ing weak c oss-sec ional dependence in la ge panels. Econome ic Re iews,34(6–10), 1089–1117. [C ossRe ]
Pesa an, M. H., & Yamaga a, T. (2008). Tes ing slope homogenei y in la ge panels. Jou nal o Econome ics,142(1), 50–93. [C ossRe ]
Economies 2025,13, 162 17 o 17
Qayyum, M., Zhang, Y., Wang, M., Yu, Y., Li, S., Ahmad, W., Maodaa, S. N., Sayed, S. R. M., & Gan, J. (2023). Ad ancemen s in
echnology and inno a ion o sus ainable ag icul u e: Unde s anding and mi iga ing g eenhouse gas emissions om ag icul u al
soils. Jou nal o En i onmen al Managemen ,347, 119147. [C ossRe ]
Rong, J., Hong, J., Guo, Q., Fang, Z., & Chen, S. (2023). Pa h mechanism and spa ial spillo e e ec o g een echnology inno a ion on
ag icul u al CO2emission in ensi y: A case s udy in Jiangsu P o ince, China. Ecological Indica o s,157, 111147. [C ossRe ]
Singh, K., Meena, R. S., Kuma , S., Dhyani, S., Sheo an, S., Singh, H. M., Pa hak, V. V., Khalid, Z., Singh, A., & Chop a, K. (2023). India’s
enewable ene gy esea ch and policies o phase down coal: Success a e Pa is ag eemen and possibili ies pos -Glasgow Clima e
Pac . Biomass and Bioene gy,177, 106944. [C ossRe ]
S e n, D. I. (2004). The ise and all o he en i onmen al kuzne s cu e. Wo ld De elopmen ,32(8), 1419–1439. [C ossRe ]
Tu an, G. (2025). Do ag icul u al expo s mi iga e he impac s o ca bon emissions in Tü kiye? Eu opean Re iew,1, 1–24. [C ossRe ]
Ve bu g, P. H., Neumann, K., & Nol, L. (2011). Challenges in using land use and land co e da a o global change s udies. Global
Change Biology,17(2), 974–989. [C ossRe ]
Wang, P., Ren, Z., & Qiao, G. (2023). How does ag icul u al ade libe aliza ion ha e en i onmen al impac s? E idence om a Li e a u e
Re iew. Sus ainabili y,15(12), 9379. [C ossRe ]
Wang, W., Chen, Y., & Pei, X. (2024). Can ag icul u al ade openness acili a e ag icul u al ca bon educ ion? E idence om Chinese
p o incial da a. Jou nal o Cleane P oduc ion,441, 140877. [C ossRe ]
Zhang, L., & Cai, C. (2024). Inno a i e measu emen , ade-o -syne gy ela ionship and in luencing ac o s o ag icul u al ne ca bon
emissions and e ec i e supply o ag icul u al p oduc s in China. Heliyon,10(3), e24621. [C ossRe ] [PubMed]
Zhang, X., Hou, Y., & Geng, K. (2024). En i onmen al egula ion, g een echnology inno a ion, and indus ial g een ans o ma ion:
Empi ical e idence om a de eloping economy. Sus ainabili y,16(16), 6833. [C ossRe ]
Disclaime /Publishe ’s No e: The s a emen s, opinions and da a con ained in all publica ions a e solely hose o he indi idual
au ho (s) and con ibu o (s) and no o MDPI and/o he edi o (s). MDPI and/o he edi o (s) disclaim esponsibili y o any inju y o
people o p ope y esul ing om any ideas, me hods, ins uc ions o p oduc s e e ed o in he con en .