Do environmental provisions in preferential trade agreements reduce emissions traded in global value chains?

Fai uz, Maisha; Fos e -McG ego , Neil
Wo king Pape
Do en i onmen al p o isions in p e e en ial ade
ag eemen s educe emissions aded in global alue
chains?
ADB Economics Wo king Pape Se ies, No. 755
P o ided in Coope a ion wi h:
Asian De elopmen Bank (ADB), Manila
Sugges ed Ci a ion: Fai uz, Maisha; Fos e -McG ego , Neil (2024) : Do en i onmen al p o isions in
p e e en ial ade ag eemen s educe emissions aded in global alue chains?, ADB Economics
Wo king Pape Se ies, No. 755, Asian De elopmen Bank (ADB), Manila,
h ps://doi.o g/10.22617/WPS240541-2
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/310378
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/3.0/igo/
ASIAN DEVELOPMENT BANK
ASIAN DEVELOPMENT BANK
6 ADB A enue, Mandaluyong Ci y
1550 Me o Manila, Philippines
www.adb.o g
DO ENVIRONMENTAL
PROVISIONS IN PREFERENTIAL
TRADE AGREEMENTS REDUCE
EMISSIONS TRADED IN
GLOBAL VALUE CHAINS?
Maisha Fai uz and Neil Fos e -McG ego
ADB ECONOMICS
WORKING PAPER SERIES
NO. 755
Decembe 2024
Do En i onmen al P o isions in P e e en ial T ade Ag eemen s Reduce Emissions
T aded in Global Value Chains?
This pape examines whe he p e e en ial ade ag eemen s (PTAs) inc ease emissions embodied
in ade, u he conside ing whe he en i onmen al p o isions in PTAs a ec his ela ionship.
The esul s sugges PTAs a e associa ed wi h inc eased emissions embodied in ade, wi h his e ec
mainly because o inc eased ade be ween PTA pa ne s. The p esence o an en i onmen al p o ision
in PTAs, howe e , mi iga es his e ec , wi h en i onmen al p o isions educing bo h he scale o ade
and he emissions in ensi y o ha ade.
Abou he Asian De elopmen Bank
ADB is commi ed o achie ing a p ospe ous, inclusi e, esilien , and sus ainable Asia and he Paci ic,
while sus aining i s e o s o e adica e ex eme po e y. Es ablished in 1966, i is owned by 69 membe s
—49 om he egion. I s main ins umen s o helping i s de eloping membe coun ies a e policy dialogue,
loans, equi y in es men s, gua an ees, g an s, and echnical assis ance.
ASIAN DEVELOPMENT BANK
The ADB Economics Wo king Pape Se ies
p esen s esea ch in p og ess o elici commen s
and encou age deba e on de elopmen issues
in Asia and he Paci ic. The iews exp essed
a e hose o he au ho s and do no necessa ily
e lec he iews and policies o ADB o
i s Boa d o Go e no s o he go e nmen s
hey ep esen .
ADB Economics Wo king Pape Se ies
Do En i onmen al P o isions in P e e en ial
T ade Ag eemen s Reduce Emissions T aded
in Global Value Chains?
Maisha Fai uz and Neil Fos e -McG ego
No. 755 | Decembe 2024
Maisha Fai uz (m. ai uz@maas ich uni e si y.nl)
is a PhD ellow a Maas ich Uni e si y.
Neil Fos e -McG ego (n os e mcg ego @adb.o g)
is a senio economis a he Economic Resea ch
and De elopmen Impac Depa men ,
Asian De elopmen Bank.
C ea i e Commons A ibu ion 3.0 IGO license (CC BY 3.0 IGO)
© 2024 Asian De elopmen Bank
6 ADB A enue, Mandaluyong Ci y, 1550 Me o Manila, Philippines
Tel +63 2 8632 4444; Fax +63 2 8636 2444
www.adb.o g
Some igh s ese ed. Published in 2024.
ISSN 2313-6537 (p in ), 2313-6545 (PDF)
Publica ion S ock No. WPS240541-2
DOI: h p://dx.doi.o g/10.22617/WPS240541-2
The iews exp essed in his publica ion a e hose o he au ho s and do no necessa ily e lec he iews and policies
o  he Asian De elopmen Bank (ADB) o i s Boa d o Go e no s o he go e nmen s hey ep esen .
ADB does no gua an ee he accu acy o he da a included in his publica ion and accep s no esponsibili y o any
consequence o hei use. The men ion o speci ic companies o p oduc s o manu ac u e s does no imply ha hey
a e endo sed o ecommended by ADB in p e e ence o o he s o a simila na u e ha a e no men ioned.
By making any designa ion o o e e ence o a pa icula e i o y o geog aphic a ea in his documen , ADB does no
in end o make any judgmen s as o he legal o o he s a us o any e i o y o a ea.
This publica ion is a ailable unde he C ea i e Commons A ibu ion 3.0 IGO license (CC BY 3.0 IGO)
h ps://c ea i ecommons.o g/licenses/by/3.0/igo/. By using he con en o his publica ion, you ag ee o be bound
by he e ms o his license. Fo a ibu ion, ansla ions, adap a ions, and pe missions, please ead he p o isions
and e ms o use a h ps://www.adb.o g/ e ms-use#openaccess.
This CC license does no apply o non-ADB copy igh ma e ials in his publica ion. I he ma e ial is a ibu ed
oano he sou ce, please con ac he copy igh owne o publishe o ha sou ce o pe mission o ep oduce i .
ADB canno be held liable o any claims ha a ise as a esul o you use o he ma e ial.
Please con ac pubsma ke [email protected] g i you ha e ques ions o commen s wi h espec o con en , o i you wish
oob ain copy igh pe mission o you in ended use ha does no all wi hin hese e ms, o o pe mission o use
heADB logo.
Co igenda o ADB publica ions may be ound a h p://www.adb.o g/publica ions/co igenda.
No es:
In his publica ion, “$” e e s o Uni ed S a es dolla s.
ADB ecognizes “China” as he People’s Republic o China.
ABSTRACT
While global alue chains (GVCs) a e conside ed an impo an de elopmen escala o ,
conce ns o e hei en i onmen al consequences ha e been inc easing. By p o iding
oppo uni ies o shi p oduc ion o de eloping economies, GVCs isk ca bon leakage.
P e e en ial ade ag eemen s (PTAs) ha e p o ed impo an d i e s o GVC ade. In his
pape we examine whe he he p esence and b ead h o PTAs is associa ed wi h
inc eased emissions embodied in GVC ade, and u he ask whe he he p esence o
en i onmen al p o isions in PTAs impac s upon emissions aded in GVCs. Ou esul s
sugges ha he p esence o a PTA be ween pa ne s is associa ed wi h inc eased
emissions embodied in GVC ade, wi h his e ec la gely he esul o inc eased GVC
ade be ween PTA pa ne s. We also show, howe e , ha he p esence o an
en i onmen al p o ision in PTAs can mi iga e his e ec , wi h en i onmen al p o isions
educing bo h he scale o GVC ade and he emissions in ensi y o ha ade.
Keywo ds: global alue chains, CO2 emissions, p e e en ial ade ag eemen s, g a i y
model
JEL codes: F13, F18, Q56
______________________
Ea ly d a s o his pape we e p esen ed in he Asian Economic In eg a ion Repo (AEIR) Theme Chap e
Wo kshop (Oc obe 2023), he Asian De elopmen Bank’s Economis Fo um (Janua y 2024), and a he UK
T ade Policy Obse a o y, Uni e si y o Sussex (Ma ch 2024).

1. INTRODUCTION
Global alue chains (GVCs) ha e become a c i ical pa o he global ade ne wo k,
d i ing economic de elopmen and indus ializa ion by enabling economies o specialize
in speci ic s ages o p oduc ion a he han en i e indus ies (Bank 2019; Taglioni and
Winkle 2016). This agmen a ion o p oduc ion has p o ided de eloping economies in
pa icula wi h a mo e accessible pa h o indus ializa ion (Baldwin 2018). Howe e ,
alongside economic bene i s, GVCs ha e been linked o a g owing sha e o global CO2
emissions. Ca bon-in ensi e p oduc ion s ages, such as in e media e manu ac u ing, can
be o sho ed o economies wi h weake en i onmen al p o ec ion, esul ing in eloca ion
a he han a educ ion in global CO2 emissions (Pe e s e al. 2011). Wi h his o sho ing
o en aking place om de eloped o de eloping economies ha ha e weake
en i onmen al s anda ds and a e mo e emissions in ensi e in p oduc ion, his ca bon
leakage can esul in highe emissions o e all. Whe eas he e idence o ca bon leakage
emains weak, he possibili y aises ques ions ega ding ade libe aliza ion policies and
he esul ing en i onmen al oo p in (An weile , Copeland, and Taylo 2001; Dean 2002;
Lopez 2017).
O e he pas ew decades, p e e en ial ade ag eemen s (PTAs) ha e acili a ed ade
libe aliza ion, which has in u n p omo ed a deepe in eg a ion o economies in o GVCs
(Dele a and Fos e -McG ego 2020). Simul aneously, PTAs ha e e ol ed o encompass
b oade issues (Ma oo, Rocha, and Ru a 2020), including en i onmen al p o isions
aimed a mi iga ing he en i onmen al impac s o ade. The cen al ques ion is whe he
PTAs, by p omo ing GVC ade, con ibu e o highe CO2 emissions, and i inclusion o
en i onmen al p o isions wi hin hese ag eemen s can e ec i ely educe emissions o
emissions in ensi y.
F om a heo e ical pe spec i e, GVCs in luence he CO2 emissions o an economy
h ough h ee p ima y channels. Fi s , a scale e ec esul s om he expec ed posi i e
impac o GVCs on p oduc i i y and p oduc ion, esul ing in highe emissions p oduced
(An weile , Copeland, and Taylo ; Wang, Wan, and Wang 2019). Second, GVCs may al e
he composi ion o p oduc ion, o en shi ing i owa d mo e emissions-in ensi e
manu ac u ing sec o s, especially in de eloping economies. Howe e , he ela ionship
be ween GVC pa icipa ion and emissions pe capi a may be nonlinea as a high enough
pa icipa ion a e may ha e a nega i e in luence on emissions (Wang, Wan, and Wang
2019). This is he s uc u e e ec , whe eby GVCs change he composi ion o p oduc i e
ac i i ies wi hin an economy, esul ing in changes in he economy’s emissions p o ile.
Thi d, he echnology e ec sugges s ha GVC pa icipa ion can lead o he di usion o
cleane echnologies, po en ially educing he ca bon in ensi y o p oduc ion (Dele a e al.
2022; G ies e al. 2018).
PTAs, pa icula ly hose wi h s ong en i onmen al p o isions, can impac all h ee o
hese channels. While PTAs p omo e GVC ade, hey can also in luence he s uc u e o
p oduc ion and encou age echnology sha ing h ough egula o y ha moniza ion and
2
c oss-bo de in es men s. Ne e heless, he p esence o en i onmen al p o isions wi hin
PTAs u he con ounds hese expec ed e ec s. Fo ins ance, en i onmen al p o isions
ha a e es ic i e may ha e a nega i e e ec on ade, ul ima ely leading o a nega i e
scale e ec h ough PTAs. Mo eo e , he ole o legal en o ceabili y o hese p o isions
emains unde explo ed wi hin a GVC con ex . Finally, such changes in PTA design also
aise ques ions o how cos s and bene i s om such ag eemen s a e dis ibu ed among
membe s o di e en income le els.
Using GVC da a om he EORA mul i egional inpu -ou pu ables (Lenzen e al. 2012,
2013) and PTA da a om Ho mann, Osnago, and Ru a (2017), we s udy he e ec o
PTAs, pa icula ly PTA b ead h and he p esence o en i onmen al p o isions, on CO2
emissions embodied in GVC ade. While PTAs a e expec ed o inc ease emissions
h ough he scale e ec , he exis ence o en i onmen al p o isions may mi iga e his. We
ind ha PTAs do inc ease CO2 emissions in GVCs. Howe e , he p esence o
en i onmen al p o isions o se s mos o he inc ease in GVC- ela ed emissions, mos
likely h ough ade es ic i e clauses. Ou esul s indica e ha en i onmen al p o isions
educe GVC emissions h ough he s uc u e and echnology e ec . In addi ion, we
explo e whe he legal en o ceabili y o hese en i onmen al p o isions makes hem mo e
e ec i e in educing emissions in GVC ade. We ind ha bo h en o ceable and non-
en o ceable en i onmen al p o isions ha e nega i e e ec s on emissions aded in
GVCs, inding in some cases ha non-en o ceable p o isions a e mo e e ec i e. Finally,
we assess whe he he e ec s o en i onmen al p o isions and PTAs in gene al o
en i onmen al consequences di e based on he income le el o he economies. We ind
ha he scale e ec p edomina es in low, uppe -middle and high-income economies, bu
he in ensi y e ec (i.e., s uc u e and echnology e ec s) is mo e p onounced in lowe -
middle income economies.
The ollowing sec ion e iews he li e a u e on he ole o GVCs in global emissions and
he e ec o PTAs in d i ing ade. I also e iews li e a u e on he consequences o
inc easing he b ead h o PTAs, especially he en i onmen al p o isions. Sec ion 3
ou lines he econome ic speci ica ion in ela ion o he scale, s uc u e, and echnology
e ec s, and desc ibes he cons uc ion o indica o s o CO2 emissions embodied in GVC
ade. Sec ion 4 p esen s he esul s and sec ion 5 p o ides a discussion and conclusion.
3
2. LITERATURE REVIEW
Impac o GVC T ade on CO2 Emissions
GVCs in ol e he agmen a ion o p oduc ion p ocesses ac oss mul iple economies, a
end which has d ama ically eshaped he global economy and ade dynamics. Thei
en i onmen al implica ions, pa icula ly ega ding CO2 emissions, ha e been s udied om
bo h he p oduc ion and consump ion pe spec i e, and ied o ecologically unequal
exchange (Al house e al. 2023; Meng e al. 2018; Mo an e al. 2013; Wang, He, and
Song 2021). GVCs accoun o an inc easing sha e o CO2 emissions and p esen he
isk o ca bon leakage, wi h ca bon in ensi e s ages o p oduc ion mo ing o economies
wi h weak en i onmen al p o ec ion and ela i ely emissions in ensi e p oduc ion
p ocesses.1 High-income economies may ou sou ce pollu ion-in ensi e p oduc ion
p ocesses o lowe -income economies, e ec i ely ans e ing he emissions bu den
while eaping he bene i s o clean inal p oduc ion (Kanemo o e al. 2012). This
phenomenon, e med “ca bon leakage,” po en ially exace ba es he en i onmen al impac
o GVCs as educ ions in emissions in one economy can be o se by inc eases in ano he
(Babike 2005) . Al hough he e idence in a o o ca bon leakage is a he weak (G ubb
e al. 2022), he po en ial o ca bon leakage is one a gumen pu o wa d o suppo he
in oduc ion o he Eu opean Union’s Ca bon Bo de Adjus men Mechanism (CBAM)
(No ds öm 2023) and emains a conce n as ca bon p ices being o ise and ca bon
p icing is in oduced in mo e ju isdic ions (Black, Pa y, and Zhunusso a 2022).
As men ioned, GVCs can a ec CO2 emissions h ough h ee main channels. The scale
e ec occu s when GVCs esul in inc eased p oduc ion, which in u n inc eases o e all
emissions (Al-Mulali and Sheau-Ting 2014; Webe e al. 2008). The s uc u e e ec
in ol es GVCs shi ing p oduc ion owa d di y sec o s, usually in manu ac u ing, which
esul s in mo e ca bon-in ensi e p oduc ion (Guo, Zou, and Wei 2010; Kumba oğlu 2011;
S öllinge 2016; Zhao e al. 2017). Finally, he echnology e ec conside s GVCs as a
sou ce o echnological change ha can educe CO2 emissions and emissions in ensi ies
(Mo ison, Pie obelli, and Rabello i 2008; Okushima and Tamu a 2010; Wang, He, and
Song 2021; Zhang e al. 2021). Gi en hese channels, he e is an expec a ion ha he
scale and s uc u e e ec o GVCs wo ks o inc ease CO2 emissions embodied in GVC
ade, while he echnology e ec dec eases emissions.
The Role o PTAs in D i ing GVC T ade and Inc easing B ead h o PTAs
The p oli e a ion o PTAs has been a c ucial d i e o GVC expansion by p omo ing ade
libe aliza ion and educing ba ie s o ade. PTAs acili a e he c oss-bo de mo emen
o goods and se ices, pa icula ly in e media e goods, which a e cen al o GVCs. This
is achie ed h ough a i educ ions, ha moniza ion o s anda ds, and in es men
p o ec ions ha enhance ade lows be ween membe economies (Baldwin 2013).
Mo eo e , he design o PTAs can in luence he ex en o which hey p omo e GVC
in eg a ion. The ho izon al dep h o b ead h o a PTA—measu ed by he
1 This alludes o he pollu ion ha en hypo hesis, which is explo ed la e in he li e a u e e iew.
4
comp ehensi eness o i s p o isions, such as he inclusion o in es men ules,
in ellec ual p ope y igh s, and dispu e se lemen mechanisms—has been shown o
co ela e posi i ely wi h GVC ade (Bo a, Jansen, and Sollede 2019; Dele a and Fos e -
McG ego 2020; Lage e al. 2020).
The expansion o PTAs beyond adi ional ade libe aliza ion e lec s he g owing
ecogni ion ha ade ag eemen s mus also add ess he en i onmen al challenges posed
by inc eased global economic in eg a ion. While ea ly PTAs p ima ily ocused on a i
educ ions and ma ke access, mo e ecen ag eemen s ha e included en i onmen al
p o isions o ensu e ha ade does no come a he cos o en i onmen al deg ada ion.
These p o isions a e designed o ha monize en i onmen al s anda ds ac oss signa o y
economies, p omo ing sus ainable de elopmen and educing he isk o nega i e
ex e nali ies, such as pollu ion om GVCs (Mo in, Dü , and Lechne 2018). The inclusion
o en i onmen al p o isions is now seen as essen ial o ensu ing ha ade libe aliza ion
suppo s no only economic g ow h bu also en i onmen al p o ec ion, pa icula ly in
indus ies deeply in eg a ed in o GVCs. In heo y, his shi in PTA design should ha e an
impac on CO2 emissions in GVC ade, as en i onmen al p o isions impose
commi men s on membe economies o uphold en i onmen al s anda ds, adop cleane
echnologies, and educe emissions embodied in ade lows (Ma ínez-Za zoso and
Ouesla i 2018).
These ends u he highligh he dual ole o PTAs wi h en i onmen al p o isions. PTAs,
by educing ade ba ie s, a e expec ed o inc ease ade among membe s which, in u n,
inc eases emissions. This posi i e co ela ion be ween ade and emissions is well
documen ed in empi ical li e a u e. A he same ime, en i onmen al p o isions in PTAs
can o se his inc ease in emissions h ough a change in s uc u e o ade o echnology
ans e ha allows cleane p oduc ion and ade.
The Rise o En i onmen al P o isions in PTAs
B andi e al. (2020) a gue ha en i onmen al p o isions wi hin PTAs a ec ade h ough
bo h ade es ic i e and libe al mechanisms. Res ic i e en i onmen al p o isions a e
designed o limi ca bon-in ensi e ade lows ei he h ough s ingen egula ions on
en i onmen al s anda ds o h ough es ic ing ade in ‘di y’ goods. Libe al
en i onmen al p o isions, on he o he hand, a e gea ed owa d acili a ing ‘g een’ ade
h ough educ ions in ade ba ie s in en i onmen al goods and acili a ing he adop ion
o clean echnologies o p omo ing eco- iendly ade p ac ices, which may enhance
compe i i eness by d i ing inno a ion and e iciency—a p emise consis en wi h he
Po e Hypo hesis (Po e 1991; Po e and an de Linde 1995). Acco ding o he Po e
Hypo hesis, well-designed en i onmen al egula ions can s imula e inno a ion, leading o
“win-win” ou comes whe e bo h en i onmen al quali y and economic pe o mance
imp o e (Mealy and Tey elboym 2022). This is analogous o he echnology e ec ou lined
ea lie .
While en i onmen al p o isions ga ne public suppo , some s udies a gue ha hey can
be used as a ba gaining ool in ade ag eemen nego ia ions as well as a mechanism o
es ic he compe i i eness o hose economies wi h he ewes en i onmen al
11
Figu e 2 shows he o e all end in GVC emissions spli o e economy pai s wi hin PTAs
and wi hou PTAs. Emissions ha esul om non-PTA based GVC ade is much g ea e
han PTA ela ed emissions h oughou he pe iod. While bo h a e s eadily inc easing,
la gely because o inc eased ade in GVCs, PTA ela ed emissions a e g owing much
as e . PTA ela ed emissions we e abou 10% o o al GVC emissions in 1990 and g ew
o abou 35% by 2010 and emained a ound ha sha e o he es o he pe iod. Fo
pu poses o compa ison, Figu e 3 shows he ends in he o al alue o GVC expo s o
1990–2015, again decomposed in o PTA and non-PTA ade. The g ow h ajec o y o
GVC expo s show a simila pa e n o emissions, al hough wi h a mo e d ama ic inc ease
in expo s a e 2000. In 1990, economy-pai s wi h a PTA ep esen ed abou 22% o o al
GVC expo s, bu his had inc eased o 46% by 2015. These wo igu es highligh he
g owing ele ance o PTA ela ed GVC ac i i ies and he emissions embodied in hem.
Emissions ha esul om GVCs g ew by 3.6% pe yea , on a e age, while expo s in
GVCs g ew by 6.5%. The g ow h a es o pai s no in PTAs we e 5.4% o expo s and
2.4% o emissions, whe eas o pai s in PTAs he numbe s we e 9.0% o expo s and
8.6% o emissions. These di e ences highligh he g owing impo ance o GVC ade
wi hin PTAs, u he sugges ing ha while emissions in ensi y has been declining o bo h
PTA and non-PTA ela ed GVC ade, he decline has been as e o non-PTA GVC ade.
Figu e 2: T end in GVC Emissions: PTA s. Non-PTA, 1990–2016
GVC = global alue chain. MT = me ic ons, PTA = p e e en ial ade ag eemen .
Sou ce: OECD TeCO2 da abase (h ps://www.oecd.o g/en/da a/da ase s/g eenhouse-gas- oo p in -indica o s.h ml)
(accessed 15 No embe 2023).
0
1,000
2,000
3,000
4,000
5,000
6,000
Millions o MT
GVC emimssions (No PTA) GVC emissions (PTA)

12
Figu e 3: T end in GVC Expo s: PTA s. Non-PTA, 1990–2016
GVC = global alue chain, PTA = p e e en ial ade ag eemen .
Sou ce: Au ho ’s calcula ions using OECD In e Coun y Inpu Ou pu Tables
(h ps://www.oecd.o g/en/da a/da ase s/in e -coun y-inpu -ou pu - ables.h ml) and OECD TeCO2 da abase
(h ps://www.oecd.o g/en/da a/da ase s/g eenhouse-gas- oo p in -indica o s.h ml) (accessed 15 No embe 2023).
Figu e 4 shows he sha e o o al GVC emissions by income le el o he expo e , wi h
abou 90% o emissions being due o expo s o uppe -middle and high-income
economies. While he combined sha e emains mos ly s able h oughou he pe iod, he e
is a sligh inc ease in he emissions sha e due o uppe -middle income economies o e
ime. The sha e o lowe -middle income economies also showed signs o an inc ease,
wi h he sha e ising om 6.2% in 1990 o 14.2% in 2016. A he same ime, he sha e o
high-income economies d opped om 51% o 41.5%. Se e al ac o s may explain hese
dynamics. Fi s , middle income economies ha e inc eased hei in eg a ion in o GVCs,
meaning ha hey accoun o a highe sha e o GVC expo s (Figu e 5). The sha e o
GVC expo s o high-income economies has allen om 83% in 1990 o 65% in 2016.
Second, echnological change in mo e de eloped economies may ha e esul ed in
inc eased emissions e iciency ha has educed hei con ibu ion o emissions. Gi en
impe ec di usion o echnologies, such de elopmen s would disp opo iona ely bene i
emissions e iciency in de eloped economies. Thi d, he changes may also be linked o
he pollu ion ha en hypo hesis, wi h de eloped economies o sho ing some o hei di ie
ac i i ies o de eloping economies. This can also be conside ed, he e o e, o ep esen
a shi in he s uc u e o GVCs be ween de eloped and de eloping economies.
0
2,000
4,000
6,000
8,000
10,000
12,000
$ billion
GVC expo s (no PTA) GVC expo s (PTA)
13
Figu e 4: Emissions in GVCs by Income Le el
GVC = global alue chain.
Sou ce: Au ho ’s calcula ions using OECD In e Coun y Inpu Ou pu Tables
(h ps://www.oecd.o g/en/da a/da ase s/in e -coun y-inpu -ou pu - ables.h ml) and OECD TeCO2 da abase
(h ps://www.oecd.o g/en/da a/da ase s/g eenhouse-gas- oo p in -indica o s.h ml) (accessed 15 No embe 2023).
Figu e 5: Expo s in GVCs by Income Le el
GVC = global alue chain.
Sou ce: Au ho ’s calcula ions using OECD In e Coun y Inpu Ou pu Tables
(h ps://www.oecd.o g/en/da a/da ase s/in e -coun y-inpu -ou pu - ables.h ml) (accessed 15 No embe 2023).
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Low Income Lowe -Middle Income Uppe -Middle Income High Income
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Low Income Lowe -Middle Income Uppe -Middle Income High Income
14
5. RESULTS
Es ima ing he Impac o he P esence and B ead h o a PTA On Emissions in GVC
Expo s
Table 1 epo s he main eg ession esul s, including esul s on he impac o he
p esence and he b ead h o PTAs on emissions embodied in GVC expo s, and he
impac o en i onmen al p o isions on GVC- ela ed emissions. The able u he epo s
esul s using a s anda d ixed-e ec s eg ession app oach and using Poisson Pseudo
Maximum Likelihood (PPML) me hods.
Following much o he ecen li e a u e using he g a i y equa ion we es ima e a s uc u al
g a i y model ha includes economy-pai and impo e - ime and expo e - ime ixed
e ec s. Resul s in Table 1 (Column 1) sugges ha he exis ence o a PTA be ween
economy-pai s is associa ed wi h an inc ease in emissions in GVC expo s o a ound
5.2%.2 This sugges s ha e en a e accoun ing o expo e , impo e and pai
cha ac e is ics, PTAs a e associa ed wi h highe emissions embodied in GVC ade. The
easons o his inc ease a e likely ela ed o he impac ha PTAs ha e on GVC ade,
wi h inc eases in GVC ade because o PTAs inc easing emissions embodied in ade.
The decomposi ion esul s ha ollow p o ide mo e insigh s.
Resul s using he measu e o PTA b ead h— ha is, he sha e o he 52 p o isions
included in he PTA ag eemen —ins ead o PTA p esence gi e simila esul s (column 2
o Table 1). The s a is ically signi ican coe icien es ima e o 0.0392 sugges s ha a mo e
om no PTA o he b oades PTA is associa ed wi h an inc ease in emissions embodied
in GVC expo s o a ound 4%.These esul s sugges ha e en comp ehensi e PTAs may
inc ease he emissions in GVC expo s, hough he e ec is ela i ely small, and we ha e
no accoun ed o any nonlinea i ies in he ela ionship be ween PTA b ead h and GVC-
ela ed emissions. To examine his po en ial nonlinea i y, we nex conside whe he he
p esence o an en i onmen al p o ision in PTAs o se s his obse ed posi i e impac o
PTA p esence on GVC- ela ed emissions.
2 Calcula ed as 𝐺𝐺𝐺𝐺𝐺𝐺(𝛽𝛽)−1.
15
Table 1: Es ima ing he Impac o he P esence and B ead h o a PTA on
Emissions in GVC Expo s
Fixed-E ec s Es ima o
PPML
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
PTA Dummy
0.0506***
0.0960***
0.0481***
0.0491***
(0.00408)
(0.00721)
(0.0180)
(0.0183)
PTA B ead h
0.0392***
0.0976***
0.0433*
0.0433*
(0.00465)
(0.0110)
(0.0256)
(0.0258)
En . P o ision
-0.0749***
-0.0655***
-0.0544***
-0.0455**
-0.0554***
-0.0456**
(0.00783)
(0.0100)
(0.0181)
(0.0210)
(0.0183)
(0.0212)
Cons an
7.324***
7.326***
7.324***
7.326***
(0.000371)
(0.000298)
(0.000360)
(0.000296)
Economy Pai FE
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Impo e -Time,
Expo e -Time FE
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Obse a ions
751,937
751,937
751,937
751,937
751,937
751,937
752,837
752,837
R-squa ed
0.995
0.995
0.995
0.995
0.995
0.995
0.995
0.995
Robus s anda d e o s in pa en heses. *** p<0.01, ** p<0.05, * p<0.1.
GVC = global alue chain, PTA = p e e en ial ade ag eemen .
Sou ce: Au ho ’s calcula ions.
Es ima ing he Impac o he P esence o an En i onmen al P o ision in PTAs on
Emissions in GVC Expo s
To conside he ole o he p esence o an en i onmen al p o ision, columns 3 and 4 o
Table 1 in oduce he dummy a iable cap u ing he p esence o an en i onmen al
p o ision. Quali a i ely, he coe icien s on he PTA a iables emain posi i e and
signi ican , hough he size o he coe icien s is somewha la ge han in columns 1 and
2.
Focusing on he coe icien on he en i onmen al p o ision dummy a iable we see ha
coe icien s a e nega i e and signi ican . Coe icien s on he en i onmen al p o ision
dummy a e simila when included alongside he PTA dummy and PTA b ead h a iable.
In he case o he PTA dummy, he coe icien on he en i onmen al p o ision a iable
sugges s ha he p esence o an en i onmen al p o ision is associa ed wi h a educ ion
in GVC- ela ed emissions o a ound 7.2%, while when including he PTA b ead h a iable
an en i onmen al p o ision is associa ed wi h a educ ion in GVC- ela ed emissions o
6.3%. While hese es ima ed educ ions a e lowe han he inc ease in emissions
associa ed wi h he p esence o b ead h o a PTA, he e ec s a e subs an ial. In he case
o he PTA dummy, he es ima ed educ ion in GVC- ela ed emissions because o he
p esence o an en i onmen al p o ision is a ound 72% o he inc ease in GVC- ela ed
emissions associa ed wi h he p esence o a PTA. These indings sugges ha while PTAs
acili a e ade and inc ease emissions, en i onmen al p o isions wi hin hese
ag eemen s play a c i ical ole in mi iga ing hese e ec s.
Es ima ing E ec s o PTAs and En i onmen al P o isions Using PPML
The esul s in he i s ou columns o Table 1 use ixed e ec s s uc u al g a i y models
and p o ide suppo o he iew ha e en as PTAs inc ease GVC- ela ed emissions
aded be ween PTA pa ne s, he p esence o an en i onmen al p o ision in hose
16
ag eemen s can o se his inc ease. The use o PPML me hods ha e become an
inc easingly popula way o dealing wi h he p esence o ze o ade lows and
he e oscedas ici y (San os and Ten ey o 2006). Fo ou pu pose, he e a e good easons
o a oiding using PPML. The dependen a iable in ou analysis, he emissions
embodied in GVC expo s, is cons uc ed using mul i egional inpu -ou pu ables and
me hods. In p inciple, he e should be no ze o obse a ions when cons uc ing hese da a,
wi h any ze o alues being he esul o missing obse a ions a he han ze o ade lows.
Including he ze o alues, he e o e, is no ecommended. In addi ion, he app oach ha
we adop elies on a decomposi ion o GVC- ela ed emissions in o a scale and in ensi y
e ec , wi h he in en ion being o decompose he e ec o PTAs and en i onmen al
p o isions in o an e ec wo king on hese wo e ms. This can be achie ed using a linea
model, such as OLS o ixed e ec s eg ession models, bu he decomposi ion b eaks
down o he nonlinea PPML model.
Despi e hese a gumen s, i is use ul o examine whe he esul s ob ained using he
s anda d ixed-e ec s model also hold when using he al e na i e PPML app oach. The
inal ou columns o Table 1, he e o e, epo esul s using he PPML es ima o . The
esul s when using PPML include bo h economy-pai and impo e - ime and expo e - ime
ixed e ec s. Columns 5 and 6 o Table 1 epo esul s om models es ima ed on he
same se o obse a ions as used in he i s ou columns, hus a oiding including missing
alues in he eg ession model, while he inal wo columns include he ull se o
obse a ions. The esul s a e consis en and indica e a posi i e impac o PTA p esence
and PTA b ead h on GVC- ela ed emissions and a nega i e impac o an en i onmen al
p o ision. Focusing on columns 5 and 6, he esul s sugges ha he p esence o a PTA
is associa ed wi h an inc ease in GVC- ela ed emissions o abou 4.9%, wi h he b oades
PTA associa ed wi h an inc ease o 4.4%. Con e sely, he p esence o an en i onmen al
p o ision is associa ed wi h a educ ion in GVC- ela ed emissions o be ween 4.4% and
5.3% depending on he speci ica ion. These esul s a e hus consis en wi h hose
epo ed using s anda d ixed e ec s models, al hough in he case o PPML he p esence
o an en i onmen al p o ision ends o mo e han o se he inc ease in GVC- ela ed
emissions associa ed wi h he p esence o a PTA. Gi en he challenges o he PPML o
ou app oach, in he es o he pape we ocus on s anda d ixed-e ec s eg ession
models.

17
Do Legally En o ceable En i onmen al P o isions Ha e Di e en ial E ec s on GVC
Expo s?
While he abo e esul s show ha he p esence o en i onmen al p o isions has a
consis en nega i e e ec on emissions, we can u he assess whe he he en o ceabili y
o such p o isions causes a di e en ial impac . Table 2 ex ends he p e ious analysis
using an in e ac ion be ween he en i onmen al p o ision a iable and whe he he
p o ision is legally en o ceable. Fo compa ison he able includes he esul s om
columns 3 and 4 o Table 1.
Focusing on he esul s on he en i onmen al p o ision a iable, he i s hing o no e is
ha he coe icien on he combined en i onmen al p o ision a iable lies be ween he
coe icien s o en o ceable and non-en o ceable p o isions as would be expec ed. When
included alongside he PTA dummy, he esul s indica e ha en i onmen al p o isions ha
a e no legally en o ceable a e associa ed wi h a dec ease in GVC- ela ed emissions o
abou 9%, while p o isions ha a e legally en o ceable a e associa ed wi h a 6%
educ ion in GVC- ela ed emissions. The wo coe icien s a e signi ican ly di e en . This
is in e es ing because he non-en o ceable p o isions seem o be mo e e ec i e in
educing emissions. This may be because non-en o ceable p o isions signal a
commi men o en i onmen al s anda ds o because hey a e backed by in e na ional
p essu e o no ms ha encou age compliance e en wi hou legal en o cemen . While
legally en o ceable p o isions a e s ill e ec i e, he smalle magni ude migh sugges ha
hei impac is mo e dependen on how s ic ly hey a e en o ced o on he speci ic legal
and ins i u ional amewo ks o he economies in ol ed.
Including he wo en i onmen al p o ision a iables alongside he PTA b ead h a iable
gi es esul s ha a e quali a i ely simila . Non-legally en o ceable PTAs a e associa ed
wi h a 7.1% dec ease on GVC- ela ed emissions, and legally en o ceable ones wi h a
5.8% educ ion. While he impac is sligh ly smalle han hose in column 2, he esul s
s ill sugges ha bo h en o ceable and non-en o ceable p o isions can be e ec i e in
educing emissions. In his case, he e a e no signi ican di e ences in he size o he wo
coe icien s. Gi en challenges in en o cing en i onmen al p o isions ela ed o di e ences
in en i onmen al s anda ds ac oss a ious ju isdic ions, as well as ha ing e ec i e legal
mechanisms ha allow en o cemen , he esul ha non-en o ceable p o isions in PTAs
a e associa ed wi h educ ions in GVC- ela ed emissions p o ides some eassu ance and
suppo o he signaling ole o en i onmen al p o isions.
18
Table 2: Reg ession Resul s Dis inguishing be ween Legally En o ceable and
Non-en o ceable En i onmen al P o isions in PTAs
(1)
(2)
(3)
(4)
PTA Dummy
0.0960***
0.0963***
(0.00721)
(0.00721)
PTA Dep h
0.0976***
0.0957***
(0.0110)
(0.0110)
En . P o ision
-0.0749***
-0.0655***
(0.00783)
(0.0100)
En . P o ision (No Legally
En o ceable)
-0.0859***
-0.0686***
(0.00873)
(0.0103)
En . P o ision (Legally En o ceable)
-0.0579***
-0.0568***
(0.00856)
(0.0111)
Cons an
7.324***
7.324***
7.326***
7.326***
(0.000360)
(0.000362)
(0.000296)
(0.000296)
Yea FE
No
No
No
No
Economy FE
Yes
Yes
Yes
Yes
Impo e -Time, Expo e -Time FE
Yes
Yes
Yes
Yes
Obse a ions
751,937
751,937
751,937
751,937
R-squa ed
0.995
0.995
0.995
0.995
Robus s anda d e o s in pa en heses. *** p<0.01, ** p<0.05, * p<0.1.
PTAs = p e e en ial ade ag eemen s.
Sou ce: Au ho ’s calcula ions.
Decomposing he E ec o PTAs and En i onmen al P o isions in o a Scale and
In ensi y E ec
A e iden i ying an e ec o PTAs and en i onmen al p o isions in PTAs on GVC- ela ed
emissions, we mo e o assess he scale and in ensi y e ec by decomposing GVC- ela ed
emissions. Resul s a e epo ed in Table 3, and he analysis ocuses on esul s using PTA
b ead h a he han he PTA dummy o easons o b e i y. In his able, models 1 and 4
shows he e ec on GVC- ela ed emissions and a e eplica ed om Table 1 and Table 2
abo e, wi h models 2 and 5 epo ing esul s wi h GVC- ela ed expo s as he a iable
(cap u ing he scale e ec ) and columns 3 and 6 esul s when GVC- ela ed emissions
in ensi y is he dependen a iable (cap u ing he emissions in ensi y e ec ).
In e ms o he scale e ec , esul s om model 2 sugges s ha a shi om no PTA o he
b oades PTA is associa ed wi h a 6.7% inc ease in GVC expo s, while he p esence o
an en i onmen al p o ision is associa ed wi h 3.9% educ ion in GVC expo s. The
co esponding esul s o he in ensi y e ec in model 3 show ha a shi o he b oades
PTA is associa ed wi h a 3.29% inc ease in emission in ensi y o GVCs and he p esence
o en i onmen al p o isions in PTAs is associa ed wi h a 2.73% educ ion. Using hese
esul s and hose om model 1 i is possible o decompose he o e all e ec o PTAs and
en i onmen al p o isions in PTAs in o a scale and in ensi y e ec . The esul s sugges
ha abou 67% (i.e., 0.0652/0.0976) o he o e all e ec o PTA b ead h on GVC- ela ed
emissions is due o an inc ease in he scale o GVC expo s in esponse o a PTA, wi h
he emaining sha e a esul o an inc ease in emissions in ensi y, likely because o a
change in he s uc u e o GVC expo s owa d mo e emissions in ensi e sec o s, such
as manu ac u ing. The scale e ec is also ound o accoun o mos o he o e all e ec
o he en i onmen al p o ision (-0.0387/-0.0655=0.59), wi h he in ensi y e ec
accoun ing o he emaining 41% o he o e all e ec . En i onmen al p o isions hus
19
seem o wo k by educing he le el o GVC expo s be ween ade pa ne s, bu also by
educing he emissions in ensi y o GVC- ela ed expo s. This la e esul would be
consis en wi h en i onmen al p o isions shi ing he s uc u e o GVC expo s owa d
cleane sec o s. The esul s do no p eclude he possibili y o en i onmen al p o isions
also esul ing in inc eased echnology di usion ha imp o es emissions in ensi y, bu we
canno dis inguish be ween he s uc u al and echnology e ec in ou esul s.
Resul s when spli ing be ween en o ceable and non-en o ceable en i onmen al
p o isions a e quali a i ely simila o he esul s in models 1–3, wi h he p esence o
en i onmen al p o isions—legally en o ceable o no —associa ed wi h lowe GVC
expo s and he emissions in ensi y o hose expo s. The esul s sugges ha he size o
he coe icien o non-en o ceable en i onmen al p o isions is la ge (in absolu e alue)
han ha o en o ceable p o isions in he case o emissions in ensi y, a di e ence ha is
s a is ically signi ican . Fo GVC expo s he e is no signi ican di e ence in coe icien s
be ween non-en o ceable and en o ceable en i onmen al p o isions, hough in his case
he coe icien is la ge (in absolu e alue) o en o ceable p o isions.
The e a e di e ences in he impo ance o he scale and in ensi y e ec s be ween
en o ceable and non-en o ceable p o isions. Fo non-en o ceable p o isions, he scale
and in ensi y e ec accoun o a simila sha e o he o e all e ec . The scale e ec
accoun s o 53% o he educ ion in GVC- ela ed emissions om an en i onmen al
p o ision, and he in ensi y e ec accoun s o 47%. In he case o en o ceable p o isions,
howe e , he scale e ec s accoun s o 79% o he o e all educ ion in GVC- ela ed
emissions, whe eas he in ensi y e ec accoun s o 21%. The esul s sugges ha
en o ceable en i onmen al p o isions ha e a much s onge impac on o e all emissions
by educing GVC expo s bu a e less success ul in educing emissions in ensi ies.
Table 3: Decomposing he E ec o PTAs and En i onmen al P o isions on GVC-
Rela ed Emissions
(1)
(2)
(3)
(4)
(5)
(6)
Emissions
Expo s
Emissions
In ensi y
Emissions
Expo s
Emissions
In ensi y
PTA Dep h
0.0976***
0.0652***
0.0324***
0.0957***
0.0666***
0.0291***
(0.0110)
(0.0104)
(0.00486)
(0.0110)
(0.0104)
(0.00482)
En . P o ision
-0.0655***
-0.0387***
-0.0269***
(0.0100)
(0.00946)
(0.00448)
En . P o ision (No Legally
En o ceable)
-0.0686***
-0.0364***
-0.0322***
(0.0103)
(0.00971)
(0.00462)
En . P o ision (Legally En o ceable)
-0.0568***
-0.0450***
-0.0118**
(0.0111)
(0.0103)
(0.00488)
Cons an
7.326***
14.30***
-6.977***
7.326***
14.30***
-6.977***
(0.000296)
(0.000263)
(0.000132)
(0.000296)
(0.000263)
(0.000132)
Yea FE
No
No
No
No
No
No
Economy FE
Yes
Yes
Yes
Yes
Yes
Yes
Impo e -Time, Expo e -Time FE
Yes
Yes
Yes
Yes
Yes
Yes
Obse a ions
751,937
751,937
751,937
751,937
751,937
751,937
R-squa ed
0.995
0.996
0.995
0.995
0.996
0.995
Robus s anda d e o s in pa en heses. *** p<0.01, ** p<0.05, * p<0.1.
GVC = global alue chain, PTA = p e e en ial ade ag eemen .
Sou ce: Au ho ’s calcula ions.
20
Impac s o PTAs and PTA P o isions on Emissions in GVC Expo s by Expo e ’s
Income Le el
We ex end he analysis by assessing whe he he magni ude o he o e all scale and
in ensi y e ec depends on he income le el o he expo e . By examining he e ec s o
PTAs and en i onmen al p o isions on emissions in GVC ade by income g oup, we a e
in e es ed in unde s anding whe he he bu den and bene i s o hese show
he e ogeneous e ec s and whe he hey all on speci ic economy g oups. One a gumen
ha we conside ele an is he idea ha since low-income coun ies end o ha e lowe
ba gaining powe in nego ia ions o e ade ag eemen s, hey may be equi ed o accep
mo e s ingen en i onmen al s anda ds in ade ag eemen s han hey would p e e . In
con as , in high-income economies en i onmen al s ingency in ade ag eemen s may
no di e subs an ially om ha al eady in exis ence. A u he a gumen is ha gi en
compa a i e ad an age low-income coun ies may end o specialize in mo e emission-
in ensi e ac i i ies in GVCs and, he e o e, hei expo s may be disp opo iona ely
a ec ed by he en i onmen al p o isions in PTAs. Con e sely, such p o isions may a o
low-income coun ies by enabling he di usion o g een echnologies ha can educe
emissions aded in GVCs. While i is no s aigh o wa d o dis inguish be ween hese
di e en hypo heses, combined hey p o ide a s ong mo i a ion o conside he
possibili y o he e ogeneous e ec s o en i onmen al p o isions in PTAs by income le el.
Resul s a e epo ed in Table 4, wi h h ee se s o each income g oup, hese being when
he dependen a iable is: (i) he logged le el o GVC- ela ed emissions; (ii) he log o
GVC expo s; and (iii) he log o he emissions in ensi y. The esul s o low-income
coun ies indica e ha a shi om no PTA o he b oades PTA is associa ed wi h an
inc ease in GVC- ela ed emissions o 58%, and wo- hi ds o his is due o he scale e ec
and he es is due o he in ensi y e ec . The e ec o an en i onmen al p o ision is o
educe he emissions embodied in PTAs by abou 20%, subs an ially lowe han he
o e all e ec o he deepes PTA. This educ ion wo ks h ough bo h he scale and
in ensi y e ec , wi h he scale e ec accoun ing o a ound 59% o he o e all e ec . The
esul s hus sugges ha en i onmen al p o isions seem o be e ec i e in mi iga ing he
e ec s o PTAs in low-income coun ies h ough egula ions ha limi he expo o
emission in ensi e goods (scale e ec ) and possibly encou aging cleane echnology ha
educes emissions in ensi y (in ensi y e ec ).
Resul s o lowe -middle and uppe -middle income economies p esen some di e ences
om hose o low-income coun ies and ela i e o each o he . The es ima ed e ec o
PTA b ead h is ound o be a ound one qua e he size o ha o low-income coun ies,
wi h esul s o bo h he lowe -middle and uppe -middle income g oups sugges ing an
inc ease in GVC- ela ed emissions o abou 11%. While in he case o he lowe -middle
income g oup his is d i en by he in ensi y e ec (accoun ing o 57% o he inc ease) in
he case o he uppe -middle income g oup he scale e ec domina es (accoun ing o
81% o he change). In o he wo ds, a PTA is associa ed wi h highe emissions in bo h
g oups, which o he lowe -middle income g oup is mainly due o an inc ease in
emissions in ensi y, pe haps e lec ing a shi in he s uc u e o GVC ade o his g oup
owa d di ie sec o s, while in uppe -middle income economies is mainly due o an
inc ease in he scale o expo s in GVCs.
27
Bo a, M., M. Jansen, and O. Sollede . 2019. Do We Need Deepe T ade Ag eemen s
o GVCs o Jus a BIT? The Wo ld Economy 42(6): 1713–39.
B andi, C., D. Blüme , and J.-F. Mo in. 2019. When Do In e na ional T ea ies Ma e o
Domes ic En i onmen al Legisla ion? Global En i onmen al Poli ics 19(4): 14–44.
B andi, C., J.-F. Mo in, and F. S ende . 2022. Do G eene T ade Ag eemen s Call o
Side-Paymen s? The Jou nal o En i onmen and De elopmen 31(2): 111–38.
B andi, C., J. Schwab, A. Be ge , and J.-F Mo in. 2020. Do En i onmen al P o isions in
ade Ag eemen s Make Expo s om De eloping Coun ies G eene ? Wo ld
De elopmen 129, 104899.
Dean, J. M. 2002. Does T ade Libe aliza ion Ha m he En i onmen ? A New Tes .
Canadian Jou nal o Economics 35(4): 819–42.
Dele a, M., and N. Fos e -McG ego . 2020. On PTAs and Bila e al T ade: Is GVC T ade
Sensi i e o he B ead h o T ade Policy Coope a ion? Economies 8(4): 84.
Dele a, M., C. Pie obelli, E. Calza, and A. La opa. 2022. Does Value Chain
Pa icipa ion Facili a e he Adop ion o Indus y 4.0 Technologies in De eloping
Coun ies? Wo ld De elopmen 152, 105788.
Downs, G. W., D. M. Rocke, and P. N. Ba soom. 1996. Is he Good News abou
Compliance Good News abou Coope a ion? In e na ional O ganiza ion 50(3),
379–406.
Ede ing on, J., and J. Minie . 2003. Is En i onmen al Policy a Seconda y T ade Ba ie ?
An Empi ical Analysis. Canadian Jou nal o Economics 36(1): 137–54.
G ies, T., R. G undmann, I. Palnau, and M. Redlin. 2018. Technology Di usion,
In e na ional In eg a ion and Pa icipa ion in De eloping Economies—A Re iew
o Majo Concep s and Findings. In e na ional Economics and Economic Policy
15(1): 215–53.
G ubb, M., N. D. Jo dan, E. He wich, K. Neuho , K. Das, K. R. Bandyopadhyay, H. an
Assel , M. Sa o, R. Wang, W. A. Pize , and H. O. 2022. Ca bon Leakage,
Consump ion, and T ade. Annual Re iew o En i onmen and Resou ces 47:
753–95.
Guo, J., L.-L. Zou, and Y.-M. Wei. 2010. Impac o In e -Sec o al T ade on Na ional and
Global CO2 Emissions: An Empi ical Analysis o China and US. Ene gy Policy
38(3): 1389–97.

28
Ho mann, C., A. Osnago, and M. Ru a. 2017. Ho izon al Dep h: a New Da abase on he
Con en o P e e en ial T ade Ag eemen s. Wo ld Bank Policy Resea ch Wo king
Pape 7981, Wo ld Bank.
Ho n, H., P. C. Ma oidis, and A. Sapi . 2010. Beyond he WTO? An Ana omy o EU and
US P e e en ial T ade Ag eemen s. The Wo ld Economy 33(11): 1565–88.
Jinnah, S., and A. Lindsay. 2016. Di usion h ough Issue Linkage: En i onmen al No ms
in US T ade Ag eemen s. Global En i onmen al Poli ics 16(3): 41–61.
Johnson, T. 2015. In o ma ion Re ela ion and S uc u al Sup emacy: The Wo ld T ade
O ganiza ion’s Inco po a ion o En i onmen al Policy. The Re iew o In e na ional
O ganiza ions 10: 207–29.
Kanemo o, K., M. Lenzen, G. P. Pe e s, D. D. Mo an, and A. Geschke. 2012.
F amewo ks o Compa ing Emissions Associa ed Wi h P oduc ion,
Consump ion, and In e na ional T ade. En i onmen al Science and Technology
46(1): 172–79.
Kolca a, D., Q. Nguyen, and T. Be naue . 2019. Does T ade Libe aliza ion Lead o
En i onmen al Bu den Shi ing in he Global Economy? Ecological Economics
163: 98–112.
Kumba oğlu, G. 2011. A Sec o al Decomposi ion Analysis o Tu kish CO2 Emissions
O e 1990–2007. Ene gy 36(5): 2419–33.
Lage , E., A. Osnago, N. Rocha, and M. Ru a. 2020. Deep T ade Ag eemen s and
Global Value Chains. Re iew o Indus ial O ganiza ion 57: 379–410.
Leon ie , W. W. 1936. Quan i a i e Inpu and Ou pu Rela ions in he Economic Sys ems
o he Uni ed S a es. The Re iew o Economic S a is ics: 105–25.
Lenzen M., K. Kanemo o, D. Mo an, and A. Geschke. 2012. Mapping he S uc u e o
he Wo ld Economy. En i onmen al Science and Technology 46(15): 8374–81.
Lenzen, M., D. Mo an, K. Kanemo o, A. Geschke. 2013. Building Eo a: A Global Mul i-
Regional Inpu -Ou pu Da abase a High Coun y and Sec o
Resolu ion. Economic Sys ems Resea ch 25(1): 20-49.
Lopez, R. 2017. The En i onmen as a Fac o o P oduc ion: The E ec s o Economic
G ow h and T ade Libe aliza ion. In In e na ional T ade and he En i onmen ,
edi ed by J. M. Dean (239–60). Rou ledge.
Ma ínez-Za zoso, I., and W. Ouesla i. 2018. Do Deep and Comp ehensi e Regional
T ade Ag eemen s Help in Reducing Ai Pollu ion? In e na ional En i onmen al
Ag eemen s: Poli ics, Law and Economics 18(6): 743–77.
29
Ma oo, A., N. Rocha, and M. Ru a. 2020. Handbook o Deep T ade Ag eemen s. The
Wo ld Bank.
Mealy, P., and A. Tey elboym. 2022. Economic Complexi y and he G een Economy.
Resea ch Policy 51(8), 103948.
Meinha , B. 2022. G eening T ade? En i onmen al P o isions in T ade Ag eemen s.
FIW Policy B ie 55.
Meng, B., G. P. Pe e s, Z. Wang, and M. Li. 2018. T acing CO2 Emissions in Global
Value Chains. Ene gy Economics 73: 24–42.
Milewicz, K., J. Hollway, C. Peacock, and D. Snidal. 2018. Beyond T ade: The
Expanding Scope o he Non ade Agenda in T ade Ag eemen s. Jou nal o
Con lic Resolu ion 62(4): 743–73.
Mo an, D. D., M. Lenzen, M., K. Kanemo o, and A. Geschke. 2013. Does Ecologically
Unequal Exchange Occu ? Ecological Economics 89: 177–86.
Mo in, J.-F., A. Dü , and L. Lechne . 2018. Mapping he T ade and En i onmen Nexus:
Insigh s om a New Da a Se . Global En i onmen al Poli ics 18(1): 122–39.
Mo ison, A., C. Pie obelli, and R. Rabello i. 2008. Global Value Chains and
Technological Capabili ies: A F amewo k o S udy Lea ning and Inno a ion in
De eloping Coun ies. Ox o d De elopmen S udies 36(1): 39–58.
No ds öm, H. 2023. Does he Risk o Ca bon Leakage Jus i y he CBAM? Wo king
Pape 2023/08, Robe Schuman Cen e o Ad anced S udies, Eu opean
Uni e si y Ins i u e.
Okushima, S., and M. Tamu a. 2010. Wha Causes he Change in Ene gy Demand in
he Economy?: The Role o Technological Change. Ene gy Economics 32, S41–
S46.
Pe e s, G. P., J.C. Minx, C.L. Webe , and O. Edenho e . 2011. G ow h in Emission
T ans e s ia In e na ional T ade om 1990 o 2008. P oceedings o he Na ional
Academy o Sciences 108(21): 8903–08.
Po e , M. E. 1991. Towa ds a Dynamic Theo y o S a egy. S a egic Managemen
Jou nal 12(S2): 95–117.
Po e , M. E., and C. an de Linde. 1995. Towa d a New Concep ion o he
En i onmen -Compe i i eness Rela ionship. Jou nal o Economic Pe spec i es
9(4): 97–118.
30
San os Sil a, J. M. C., and S. Ten ey o. 2006. The Log o G a i y, The Re iew o
Economics and S a is ics 88(4): 641–58.
Sp inz, D., and T. Vaah o an a. 1994. The In e es -Based Explana ion o In e na ional
En i onmen al Policy. In e na ional O ganiza ion 48(1): 77–105.
S öllinge , R. 2016. S uc u al Change and Global Value Chains in he EU. Empi ica 43
(4): 801–29.
Taglioni, D., and D. Winkle . 2016. Making Global Value Chains Wo k o De elopmen .
Wo ld Bank.
Wang, J., G. Wan, and C. Wang. 2019. Pa icipa ion in GVCs and CO2 Emissions.
Ene gy Economics 84, 104561.
Wang, S., Y. He, and M. Song. 2021. Global Value Chains, Technological P og ess, and
En i onmen al Pollu ion: Inequali y Towa ds De eloping Coun ies. Jou nal o
En i onmen al Managemen 277, 110999.
Wang, Z., S. Wei, and K. Zhu. 2017. Measu es o Pa icipa ion in Global Value Chains
and Global Business Cycles.
Webe , C. L., G. P. Pe e s, D. Guan, and K. Hubacek. 2008. The con ibu ion o Chinese
expo s o clima e change. Ene gy Policy, 36(9), 3572-3577.
Wo ld Bank. 2019. Wo ld De elopmen Repo 2020: T ading o De elopmen in he
Age o Global Value Chains.
Zhang, D., H. Wang, A. Löschel, and P. Zhou. 2021. The Changing Role o Global Value
Chains in CO2 Emission In ensi y in 2000–2014. Ene gy Economics 93, 105053.
Zhang, Y., J. Wang, H. Deng, D. Zhang, and Y. Wang. 2023. De eloping a
Mul idimensional Assessmen F amewo k o Clean Technology T ans e
Po en ial and I s Applica ion on he Bel and Road Ini ia i e Coun ies. Jou nal o
Cleane P oduc ion 401, 136769.
Zhao, X., X. Zhang, N. Li, S. Shao, and Y. Geng. 2017. Decoupling Economic G ow h
om Ca bon Dioxide Emissions in China: A Sec o al Fac o Decomposi ion
Analysis. Jou nal o Cleane P oduc ion 142, 3500–16.
Zhou, L., X. Tian, and Z. Zhou. 2017. The E ec s o En i onmen al P o isions in RTAs
on PM2. 5 Ai Pollu ion. Applied Economics 49(27): 2630–41.
ASIAN DEVELOPMENT BANK
ASIAN DEVELOPMENT BANK
6 ADB A enue, Mandaluyong Ci y
1550 Me o Manila, Philippines
www.adb.o g
DO ENVIRONMENTAL
PROVISIONS IN PREFERENTIAL
TRADE AGREEMENTS REDUCE
EMISSIONS TRADED IN
GLOBAL VALUE CHAINS?
Maisha Fai uz and Neil Fos e -McG ego
ADB ECONOMICS
WORKING PAPER SERIES
NO. 755
No embe 2024
Do En i onmen al P o isions in P e e en ial T ade Ag eemen s Reduce Emissions
T aded in Global Value Chains?
This pape examines whe he p e e en ial ade ag eemen s (PTAs) inc ease emissions embodied
in ade, u he conside ing whe he en i onmen al p o isions in PTAs a ec his ela ionship.
The esul s sugges PTAs a e associa ed wi h inc eased emissions embodied in ade, wi h his e ec
mainly because o inc eased ade be ween PTA pa ne s. The p esence o an en i onmen al p o ision
in PTAs, howe e , mi iga es his e ec , wi h en i onmen al p o isions educing bo h he scale o ade
and he emissions in ensi y o ha ade.
Abou he Asian De elopmen Bank
ADB is commi ed o achie ing a p ospe ous, inclusi e, esilien , and sus ainable Asia and he Paci ic,
while sus aining i s e o s o e adica e ex eme po e y. Es ablished in 1966, i is owned by 69 membe s
—49 om he egion. I s main ins umen s o helping i s de eloping membe coun ies a e policy dialogue,
loans, equi y in es men s, gua an ees, g an s, and echnical assis ance.