Local Content Requirements: Promises and Pitfalls

Ing, Lili Yan (Ed.); G ossman, Gene M. (Ed.)
Book
Local Con en Requi emen s: P omises and Pi alls
Rou ledge-ERIA S udies in De elopmen Economics
P o ided in Coope a ion wi h:
Taylo & F ancis G oup
Sugges ed Ci a ion: Ing, Lili Yan (Ed.); G ossman, Gene M. (Ed.) (2024) : Local Con en Requi emen s:
P omises and Pi alls, Rou ledge-ERIA S udies in De elopmen Economics, ISBN 978-1-003-80691-2,
Rou ledge, London,
h ps://doi.o g/10.4324/9781003415794
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/290617
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-nc-nd/4.0/
As an i-globaliza ion and geopoli ical ensions con inue o ise, he use o local
con en equi emen s (LCRs) a ound he wo ld has become mo e no iceable
han e e be o e.
The easons o adop ing LCRs ange om ensu ing domes ic supply
a ailabili y, job c ea ion, and inc easing alue added o sa egua ding na ional
secu i y. Ing and G ossman examine coun y-speci ic as well as i m-p oduc
le el exe cises o explain how LCRs educe ai compe i ion, esul ing in
lowe ade and p oduc i i y, which ul ima ely lowe s wo ld economic ou pu
and o e all human wel a e. Coun ies a ound he wo ld a e in es iga ed
wi h speci ic a en ion o he US, China, Indonesia, and esou ce-in ensi e
coun ies, including mining-in ensi e ones. The book also p esen s p oduc -
and i m-le el analyses, answe ing he ques ion o why coun ies adop ed
LCRs and how LCRs a ec he wo ld economy.
This book is a use ul esou ce ha will in e es policymake s, esea che s,
and ad anced unde g adua es in e es ed in in e na ional ade, indus ial
policy, poli ical economy, labou economics, and de elopmen economics.
Lili Yan Ing is a lead ad iso (Sou heas Asia Region) a he Economic Resea ch
Ins i u e o ASEAN and Eas Asia (ERIA). She also se es as sec e a y gene al
o he In e na ional Economic Associa ion (IEA).
Gene M. G ossman is he Jacob Vine P o esso o In e na ional Economics
in he Depa men o Economics and he School o Public and In e na ional
A ai s a P ince on Uni e si y.
Local Con en Requi emen s
P oduc ion Ne wo ks in Sou heas Asia
Edi ed by Lili Yan Ing and Fukuna i Kimu a
The Indonesian Economy
T ade and Indus ial Policies
Edi ed by Lili Yan Ing, Go don H. Hanson, and S i Mulyani Ind awa i
Wo ld T ade E olu ion
G ow h, P oduc i i y and Employmen
Edi ed by Lili Yan Ing and Miaojie Yu
Eas Asian In eg a ion
Goods, Se ices, and In es men
Edi ed by Lili Yan Ing, Ma in Richa dson and Shuji o U a a
De eloping he Digi al Economy in ASEAN
Edi ed by Lu ong Chen and Fukuna i Kimu a
COVID-19 in Indonesia
Impac s on he Economy and Ways o Reco e y
Edi ed by Lili Yan Ing and M. Cha ib Bas i
Robo s and AI
A New Economic E a
Edi ed by Lili Yan Ing and Gene M. G ossman
Local Con en Requi emen s
P omises and Pi alls
Edi ed by Lili Yan Ing and Gene M. G ossman
Rou ledge-ERIA S udies in De elopmen Economics
Fo mo e in o ma ion abou his se ies, please isi : www. ou ledge.com/Rou ledge-ERIA-
S udies-in-De elopmen -Economics/book-se ies/ERIA
Local Con en Requi emen s
P omises and Pi alls
Edi ed by
Lili Yan Ing and Gene M. G ossman

Fi s published 2024
by Rou ledge
4 Pa k Squa e, Mil on Pa k, Abingdon, Oxon OX14 4RN
and by Rou ledge
605 Thi d A enue, New Yo k, NY 10158
Rou ledge is an imp in o he Taylo & F ancis G oup, an in o ma business
© 2024 selec ion and edi o ial ma e , Lili Yan Ing and Gene M.
G ossman; indi idual chap e s, he con ibu o s
The igh o Lili Yan Ing and Gene M. G ossman o be iden i ied as he
au ho s o he edi o ial ma e ial, and o he au ho s o hei indi idual
chap e s, has been asse ed in acco dance wi h sec ions 77 and 78 o
he Copy igh , Designs and Pa en s Ac 1988.
The Open Access e sion o his book, a ailable a www. aylo ancis.
com, has been made a ailable unde a C ea i e Commons A ibu ion-
Non-Comme cial-No De i a i es 4.0 license.
Funded by Economic Resea ch Ins i u e o ASEAN and Eas Asia
(ERIA).
T adema k no ice: P oduc o co po a e names may be adema ks
o egis e ed adema ks and a e used only o iden i ica ion and
explana ion wi hou in en o in inge.
B i ish Lib a y Ca aloguing-in-Publica ion Da a
A ca alogue eco d o his book is a ailable om he B i ish Lib a y
Lib a y o Cong ess Ca aloging-in-Publica ion Da a
Names: Ing, Lili Yan, edi o . | G ossman, Gene M., edi o .
Ti le: Local con en equi emen s : p omises and pi alls /
edi ed by Lili Yan Ing and Gene M. G ossman.
Desc ip ion: Abingdon, Oxon ; New Yo k, NY : Rou ledge, 2024. |
Se ies: Rou ledge-ERIA s udies in de elopmen economics |
Includes bibliog aphical e e ences and index.
Iden i ie s: LCCN 2023029276 (p in ) | LCCN 2023029277 (ebook) |
ISBN 9781032542232 (ha dback) | ISBN 9781032542218 (pape back) |
ISBN 9781003415794 (ebook)
Subjec s: LCSH: In e na ional ade. | Fo eign ade egula ion. |
Comme cial policy. | Economic de elopmen .
Classi ica ion: LCC HF1379 .L625 2024 (p in ) | LCC HF1379 (ebook) |
DDC 382—dc23/eng/20230629
LC eco d a ailable a h ps://lccn.loc.go /2023029276
LC ebook eco d a ailable a h ps://lccn.loc.go /2023029277
ISBN: 978-1-032-54223-2 (hbk)
ISBN: 978-1-032-54221-8 (pbk)
ISBN: 978-1-003-41579-4 (ebk)
DOI: 10.4324/9781003415794
Typese in Gallia d
by Apex CoVan age, LLC
P o ec ion o sale. Lo e is no
Michelle and Han Na
LYI
Nick, Dina, Rob, Sha i, and Sam
GMG
Lis o igu es ix
Lis o ables x
Lis o con ibu o s xiii
Acknowledgemen s xi
1 In oduc ion 1
GENE M. GROSSMAN AND LILI YAN ING
2 Localiza ion measu es: a global pe spec i e 14
DOROTHEE FLAIG AND SUSAN F. STONE
3 Local con en policies in he mining sec o 48
JANE KORINEK AND PAULO DE SA
4 The unin ended consequences o high egional con en
equi emen s 87
KEITH HEAD, THIERRY MAYER, AND MARC MELITZ
5 LCR Policies in China and hei impac s on domes ic
alue added in expo s 114
KUN CAI AND ZHI WANG
6 Con o mi y o Indonesia’s LCRs wi h i s ade and
in es men commi men s 145
MICHELLE LIMENTA, LILI YAN ING, JUNIANTO JAMES LOSARI, AND
OSCAR FERNANDO
Con en s
We specially hank Cha ib Bas i, E nawa i Munadi, Siwage Dha ma, Dani
Rod ik, Jus in Yi u Lin, Miaojie Yu, Elhanan Helpman, Hal Hill, Te suya
Wa anabe, Shuji o U a a, Ha yo Aswicahyono, Kiki Ve ico, Fauziah Zen,
Iman Pambagyo, Ma i Panges u, and colleagues a he Minis y o T ade o
Indonesia, he Minis y o Finance o Indonesia, CEPR, OECD, Uni e si y
o Indonesia, CSIS, Gadjah Mada Uni e si y, Peking Uni e si y, Sun Ya Sen
Uni e si y, Liaoning Uni e si y, and Uni e si y o Peli a Ha apan o sha ing
hei insigh s on LCRs. I ana Ma kus and Li ia Naza a p o ided excellen
esea ch assis ance and Ca he in Sa i i p o ided gene ous adminis a ion sup-
po . We also hank Gilbe Gnana a hinam, Kend ick Loo, and Chelsea Low
om Rou ledge, and Fad iani T ianingsih om ERIA.
Acknowledgemen s

DOI: 10.4324/9781003415794-1
This chap e has been made a ailable unde a CC-BY-NC-ND 4.0 license.
Local con en equi emen s in heo y and p ac ice
Local con en equi emen s (LCRs) ha e been used by many coun ies, bo h
de eloped and de eloping, o p omo e he use o local inpu s and suppo
he g ow h o domes ic indus ies. Ini ially, he e m LCR (o , equi alen ly,
“con en p o ec ion”) was used o e e o a manda e ha a ce ain ac-
ion o domes ically p oduced inpu s, by alue o by olume, be inco po-
a ed in any inal good sold in he domes ic ma ke . O e ime, he ange o
policies co e ed by he e m has expanded alongside he inc eased ange o
localiza ion p ac ices used by a ious na ional and local go e nmen s. Now,
ou comes may be legally manda ed o aspi a ional. The ou comes may e e -
ence inpu sha es, employmen , i m-owne ship sha es, loca ion o R&D,
o echnology ans e . LCRs may include es ic ions on he p o ision o
ce ain se ices, eligibili y o go e nmen con ac s, local pe o mance o
compliance es s, o he loca ion o da a s o age. Aspi a ional a ge s migh
be incen i ized wi h subsidized expo o in es men inancing, wi h ax
b eaks, wi h p ice concessions o go e nmen -supplied ene gy o in as uc-
u e, wi h condi ional bailou s, o wi h o he inancial inducemen s. In his
book, we use he e m LCR b oadly o include any laws o egula ions ha
equi e o encou ages he use o locally p oduced inpu s o se ices in a
mul i-s age p oduc ion p ocess.
LCRs also play a ole in bila e al and egional ade ag eemen s, whe e
hey a e known as “ ules o o igin” (RoOs). T ade ag eemen s gene ally call
o p e e en ial a i ea men o goods emana ing om a pa ne coun y.
Bu such ag eemen s mus de ine wha i means o “emana e om,” o else
goods impo ed om ou side he membe coun ies may en e he egion
in a low- a i coun y and hen be shipped on o a high- a i coun y a e
he addi ion o only minimal o negligible local alue added. RoOs speci y
wha ac ion o he alue added o an in e nally aded good mus o igina e
wi hin he egion in o de ha he good quali y o p e e en ial ea men .
RoOs migh also u he s ipula e minimum pe cen ages om each o he
a ious coun ies wi hin he egion, as wi h ce ain p o isions o he Uni ed
S a es-Mexico-Canada T ade Ag eemen (USMCA). While pe haps o iginally
1 In oduc ion
Gene M. G ossman and Lili Yan Ing
2 Gene M. G ossman and Lili Yan Ing
in ended o hwa ansshipmen , RoOs a e egula ly used now o encou age
egional p oduc ion.
LCRs i s en e ed he a senal o ade ins umen s in Aus alia, which, in
1948, es ic ed he use o impo ed ca pa s in he local assembly ope a-
ions o B i ish mul ina ionals while o e ing concessiona y inancing based
on he ac ion o Aus alian alue added o encou age he p oduc ion o
“Aus alia’s own ca ” (Pu sell, 2001). Se e al coun ies quickly ollowed sui ,
including Canada, which ins i u ed LCRs o shield domes ic pa s p oduce s
om Ame ican compe i ion p io o he conclusion o he Canada-Ame ican
Au omo i e Ag eemen in 1965 (Wonnaco and Wonnaco , 1967; John-
son, 1971). LCR policies o os e impo subs i u ion in he au omobile
indus y soon became commonplace in La in Ame ica, whe e hey we e
in oduced in Chile, A gen ina, Mexico, and B azil (Munk, 1969; Johnson,
1967). Mo eo e , Aus alia quickly ex ended i s use o his new ins umen
well beyond he au omobile sec o , implemen ing policies o encou age
o equi e use o local inpu s in indus ies as dispa a e as pe ochemicals,
obacco, peanu oil, co ee, ui juices, indus ial machine y, and ag icul-
u al ac o s (Lloyd, 1971). S ill, LCRs we e ela i ely uncommon when
Co den (1971) and G ossman (1981) i s analyzed hei economic conse-
quences. Thei popula i y waned in he 1980s as mo e and mo e coun ies
became disillusioned wi h using a s a egy o impo subs i u ion o p omo e
de elopmen .
LCRs ha e made a oa ing comeback, pa icula ly a e he Global Financial
C isis in 2008. Be ween 2008 and 2013, almos 200 new LCR measu es we e
in oduced, acco ding o he Global T ade Ale . This igu e g ew o mo e
han 500 measu es ha we e pu in o place du ing he pe iod om 2014 o
2020. Mo eo e , he implemen a ion o hese policies has been widesp ead,
anging ac oss de eloped and de eloping economies. Qui e p ominen ly, he
Uni ed S a es has made LCRs a co ne s one o i s ecen policy o p omo e
he de elopmen o elec ic ehicles as pa o he In la ion Reduc ion Ac
o 2022. The ange o economic ac i i ies a ge ed by LCR policies a ound
he globe has expanded o include many esou ce-ex ac ing sec o s, in o ma-
ion echnology, heal hca e goods and se ices, inancial se ices, ag icul u al
p oduc s, and o he s.
Why do coun ies adop LCRs ha a o local sou cing o in e media e
inpu s and se ices? The lis o a gumen s o suppo such policies mi o s
hose o e ed o p o ec ionis ade policies mo e gene ally. Fi s , LCRs migh
a o d new job oppo uni ies in ce ain sec o s o egions o he economy.
These jobs, in u n, migh boos wages, educe unemploymen , o encou age
in es men s in human capi al. Second, LCRs, by encou aging speci ic local
ac i i ies, migh p o ide spillo e bene i s o o he ac i i ies and sec o s ia
esea ch and de elopmen o lea ning by doing. Thi d, localiza ion policies
migh encou age o manda e g ea e o eign di ec in es men , join - en u e
pa ne ships, o echnology ans e on e ms a o able o he hos coun y o
a he expense o al e na i e hos s. LCRs applied o p ima y p oduc s a e o en
In oduc ion 3
in ended o encou age highe alue-added ac i i ies, wi h he aim o p omo -
ing indigenous managemen skills and echnological knowhow.
O cou se, LCRs, like o he o ms o p o ec ion, o en ail o achie e hese
lo y goals. Whe he in oduced by well-meaning leade s o in esponse o
special-in e es lobbying, such policies o en ail o gene a e posi i e spillo-
e s o su icien magni ude o jus i y he highe cos s o domes ic sou cing.
The desi ed jobs may no ma e ialize due o inadequa e managemen , lack o
equisi e skills, una ailabili y o complemen a y inpu s, o o he easons. E en
i new job oppo uni ies a e gene a ed in a ge ed sec o s, hey may come a
he expense o employmen in o he sec o s ha po en ially o e g ea e eco-
nomic bene i . In sho , policies ha disc imina e in a o o local p oduce s o
inpu s and se ices may be subjec o he same, un a o able cos -bene i analy-
sis as wi h o he o ms o ade p o ec ion. Nume ous ales o disad an ageous
LCRs a e old in Hu baue e al. (2013), S one e al. (2015), and elsewhe e in
li e a u e. Simila c i iques o RoOs in bila e al and egional ade ag eemen s
may be ound in Cado e al. (2006), K uege (2012), Conconi e al. (2018),
Cado and Ing (2019), and elsewhe e.
The heo e ical li e a u e, beginning wi h G ossman (1981) and Dixi
and G ossman (1982), has iden i ied a pa icula isk associa ed wi h LCRs
ha dis inguishes hese policies om a i s and o he o ms o p o ec ion o
domes ic indus ies. Whe eas a i s on inal goods boos local demand o he
p o ec ed goods and he eby demand o all inpu s (including hose p oduced
locally) used in he p oduc ion o hese goods, LCRs ha aise he cos o
inpu s can easily ha e unin ended consequences. Alongside he impo sub-
s i u ion manda ed o encou aged by hese policies comes an ad e se “ou pu
e ec ”; as cos s o downs eam p oduce s ise, hese i ms likely will scale back
p oduc ion and educe hei demand o inpu s in he p ocess. The o se ing
subs i u ion and ou pu e ec s o LCRs may help o explain why empi ical
s udies o en ind disappoin ing o e en ad e se e ec s o hese policies on
employmen , alue added, and o eign in es men in a ge ed indus ies.
Ou book
This book is in ended o upda e he li e a u e on local con en equi emen s
and ules o o igin and o u he unde s anding o he expe ience wi h and
consequences o such policies and hei consis ency o no wi h he ules
es ablished by he wo ld ading sys em.
The dis inc i e ea u es o ou book a e wo old. Fi s , he esea ch epo ed
he e uses he mos up- o-da e ca alogs o LCRs ha ha e been g owing apidly
since he Global Financial C isis o 2008. Second, he esea ch complemen s
analysis o LCRs a he global le el wi h coun y- and i m-speci ic exe cises.
The emainde o his book con ains se en chap e s ha can loosely be
di ided in o h ee pa s. The i s pa , comp ising Chap e s 2 and 3, p o ides
an o e iew o majo LCR policies used globally (Chap e 2) and in min-
e al expo ing coun ies (Chap e 3). The second pa ocuses on impo an
4 Gene M. G ossman and Lili Yan Ing
ecen LCR policies in he wo ld’s wo la ges economies, he Uni ed S a es
and China. Chap e 4 conce ns he e ec s on he o ganiza ion o he No h
Ame ican au omobile sec o o he new RoOs in he USMCA while ouching
also on he implica ions o he ca indus y o B i ain’s exi om he Eu o-
pean Union. Chap e 5 add esses China’s indus ial policy ini ia i es such as
“Made in China 2025” ha a e in ended o p omo e u he indus ializa-
ion and inno a ion in ha coun y. The inal h ee chap e s shine a spo ligh
on Indonesia, a la ge, eme ging economy ha is a majo expo e o na u al
esou ces. Indonesia is in e es ing o ou pu poses because i s laws and egu-
la ions include a a ie y o LCR policies and because i p o ides a es case o
he consis ency o such policies wi h he ules ag eed by membe s o he Wo ld
T ade O ganiza ion (WTO).
We p oceed now o desc ibe he con en s and main con ibu ions o hese
chap e s in somewha g ea e de ail.
Chap e 2 by Do o hee Flaig and Susan F. S one discusses he ecen expe-
ience wi h LCRs in he wo ld economy. They begin by e iewing he ea-
sons why coun ies implemen policies ha s ipula e o incen i ize he use o
domes ic inpu s in local p oduc ion. Among he mos p ominen objec i es
ha hey ci e a e employmen objec i es and echnology ans e . Nex , he
au ho s discuss ends in he implemen a ion o LCRs, poin ing o an accel-
e a ion o usage om he pe iod o 2008–2013 o he mo e ecen pe iod o
2014–2020, as epo ed by Global T ade Ale . The au ho s ci e India, Ge -
many, and he Uni ed Kingdom as he mos in ensi e use s o LCRs, bu hey
quali y his obse a ion by poin ing ou ha coun s o usage do no accoun
o he e ogenous impac and ha he Global T ade Ale allies LCRs in mul i-
ple ju isdic ions wi hin a coun y, so ha mo e decen alized poli ies a e likely
o ha e highe coun s.
The hea o he chap e uses he OECD T ade Model, METRO, o p o-
ide a quan i a i e e alua ion o se en majo ins ances o new LCR policies,
chosen om a sample o 565 measu es ha we e conside ed o he pu pose.
The measu es unde e iew we e selec ed as hose ha could be modeled
quan i a i ely and ha we e likely o be among he mos ade dis o ing.
In o de o apply he METRO model, a policy mus speci y an iden i iable
sec o and an iden i iable es ic ion ha could be meaning ully en o ced, i
mus a ec a su icien ly la ge sec o o egion o he economy, and i mus
be a binding measu e applied whe e he domes ic sec o has capaci y o mee
he equi ed demand. Applica ion o hese c i e ia yields a good sample o
he ypes o LCR measu es ha ha e been applied ecen ly and o he ypes
o economies ha ha e applied hem. Speci ically, he au ho s ocus on (i) ax
c edi s a ailable o A gen inian ca p oduce s ha use speci ied pe cen ages o
local con en ; (ii) a equi emen imposed by B azil on he elecommunica ions
sec o ha hey use a minimum o local con en in hei 4G ne wo ks; (iii) he
p e e ence ma gins allowed by he B azilian go e nmen o public p ocu e-
men o a a ie y o na ionally p oduced goods; (i ) egula ions in oduced by
he Indonesian go e nmen ha essen ially equi e he en i e assembly p ocess
In oduc ion 5
o mo o ehicles and mo o cycles o ake place locally; ( ) a egula ion ha
makes i manda o y o Saudi A abian go e nmen al agencies o pu chase
hei medical supplies om he local indus y; ( i) a Mining Cha e in Sou h
A ica ha es ablishes a minimum local con en equi emen o mining goods
and o o al se ices used by i ms in he domes ic mining indus y; and ( ii)
he Buy Ame ica p og am ha is equi ed o U.S. s a es ha ecei e g an s o
anspo a ion unding om he ede al go e nmen .
METRO is a s a ic, global, compu able gene al equilib ium (CGE) model.
I inco po a es many coun ies, sec o s, and ac o s o p oduc ion and dis in-
guishes ou pu om each sec o acco ding o i s end use. The au ho s model
each o he LCR measu es as a es ic ion o calib a ed magni ude on he
inpu choices o sec o s in he a ec ed coun ies. A e sol ing he baseline
model unde he assump ion ha he LCR policies a e no binding, hey e-
sol e he model imposing hei calib a ed es ic ions. The simula ions p o ide
es ima es o he e ec s o he se en policies on eal GDP, ade lows, labo
income, o al disposable income, and he e ms o ade. They also gene a e
disagg ega ed es ima es o he e ec s o he policies on impo s and p oduc-
ion in he 27 sec o s cap u ed by he model.
Flaig and S one es ima e ha he economic impac s o he se en LCR meas-
u es hey s udy a e modes bu gene ally nega i e. The LCRs end o unde -
mine he long- un compe i i eness o he sec o s in which hey a e applied
while ha ing limi ed o negligible spillo e e ec s on he b oade economies.
Since he model assumes ull employmen , whe e he policy a ge s a la ge
sec o – such as he LCRs in he au omobile indus ies in A gen ina and
Indonesia – he expanded use o local inpu s necessa ily comes a he expense
o o he sec o s o he economy. As he non- a ge ed indus ies subs i u e
away om domes ic inpu s, hei demand o impo s g ows, wi h a po en ially
nega i e (albei small) e ec on he e ms o ade.
Chap e 3 by Jane Ko inek and Paolo De Sa builds on Ko inek and Ram-
doo (2017) and ocuses on laws on egula ions ha seek o s imula e g ow h
o local indus ies ups eam and downs eam om he mining sec o s in
esou ce- ich economies. The chap e begins wi h a ypology o LCRs ha
dis inguishes manda o y measu es om incen i es-based policies and supply-
side policies om demand-side policies. Demand-side policies in ended o
p omo e backwa d linkages be ween mining i ms and hei supplie s include
p e e ences o local supplie s o inpu s used in mining and nume ical a ge s
o employmen by ex ac i e i ms. Supply-side policies wi h ups eam suppli-
e s include equi emen s o p o ide aining, und capaci y de elopmen , and
conduc awa eness campaigns abou p ocu emen oppo uni ies. Demand-
side policies aimed a inc easing in e ac ions wi h i ms and indus ies down-
s eam om he mining sec o include es ic ions o axes on mine al expo s
beyond hose on p ocessed p oduc s, equi emen s ha ex ac i e i ms sell a
speci ied sha e o hei ou pu domes ically, o ax concessions ha a o local
sales. On he supply side, LCRs migh equi e ex ac i e i ms o in es in
downs eam p ocessing acili ies o o collabo a e wi h aining ins i u ions o

6 Gene M. G ossman and Lili Yan Ing
p omo e he de elopmen o needed skills. Incen i e-based policies o encou -
age o wa d linkages migh include ax concessions based on domes ic sales o
mined ma e ials o subsidized loans o capaci y in es men s. The au ho s no e
ha manda es ha e been mo e popula in A ican coun ies such as Ghana,
Sou h A ica, Tanzania, and Zambia, whe eas mo e de eloped economies such
as Aus alia, Canada, and Chile ha e elied mo e hea ily on an incen i e-based
app oach.
Ko inek and De Sa go on o highligh a numbe o common easons LCRs
in he mining indus ies ha e gene a ed disappoin ing esul s. Fi s , an inad-
equa e app ecia ion o mining i ms’ inpu s needs and o he abso p i e capac-
i y o local supplie s has led many coun ies o se un ealis ic a ge s o local
p ocu emen ha domes ic i ms ha e been unable o sa is y. Second, some
coun ies ha e employed b oad de ini ions o local con en , which p o ides
lexibili y o he sec o bu makes i di icul o assess gains in alue added
and spillo e s o he es o he economy. The a ailable e idence sugges s ha
manda o y, quan i a i e LCRs ha e ailed o gene a e signi ican g ow h in he
use o locally sou ced inpu s by he mining sec o , no ha e hey s eng hened
linkages wi h ups eam indus ies. Meanwhile, equi emen s o incen i es o
downs eam p ocessing ha e ailed in he long un when p ocessing i ms ha e
no been able o p oduce goods o su icien quali y o o achie e in e na-
ional cos compe i i eness. Finally, he au ho s no e, LCRs ha e con ibu ed
o go e nmen de ici s e en when hey appea on pape o be iscally neu al,
because hey dec ease p o i abili y in he mining sec o s and he eby educe
he go e nmen s’ eceip s om co po a e axes and oyal y paymen s.
The au ho s conclude ha when used a all, LCRs should be pa o a com-
p ehensi e policy o p omo e ins i u ional de elopmen and o os e in e sec-
o al pa ne ships. They no e ha go e nmen suppo o sec o s ha ely on
local mine al ex ac ion may exace ba e he impac s o ma e ials p ice luc ua-
ions, ha m he local ecology, and in e e e wi h di e si ica ion o he b oade
economy. Ins ead, Ko inek and De Sa sugges ha esou ce- ich coun ies
de o e g ea e a en ion o p omo ing mac oeconomic s abili y, emo ing
ba ie s o en y, p o iding a anspa en and s able egula o y en i onmen ,
and imp o ing local in as uc u e, ins i u ions, and skill le els.
Chap e 4 by Kei h Head, Thie y Maye , and Ma c Meli z analyzes he
igh e ules o o igin ha now apply o he No h Ame ican au omobile
indus y ollowing he enego ia ion o he egional ade ag eemen ha was
o me ly NAFTA and now is USMCA. NAFTA equi ed egional con en o
62.5% o alue o ca s o quali y o du y- ee en y in o one o i s membe s.
The USMCA aised his egional con en equi emen o 75% and in oduced
addi ional, binding equi emen s. The igh ening o he RoOs and he add-
ing o addi ional equi emen s clea ly we e in ended o discou age i ms om
sou cing pa s om ou side No h Ame ica, which he pa ies (and especially
he Uni ed S a es) hoped would bols e demand o No h Ame ican pa s.
In a companion pape , Head e al. (2022), he au ho s ex end he one-
inpu G ossman (1981) model o include many inpu s. Howe e , hei main
In oduc ion 7
esul , unlike wha was emphasized by G ossman, does no ely on a educ ion
in he numbe o comple ed au omobiles. Ins ead, hey make ano he impo -
an obse a ion. Fi ms in he au omobile indus y e ain he op ion o choose
he cos -minimizing sou ce o pa s, p o ided hey a e willing o sac i ice he
ea y’s a i bene i and pay he MFN a i whene e a comple ed ca c osses
a bo de . Indeed, Head e al. documen in Chap e 4 ha compliance wi h
he RoOs has declined since he in oduc ion o he s ic e USMCA ules.
I compliance a es all su icien ly, a igh e RoO in ended o expand alue
added wi hin he egion migh ha e he opposi e e ec . The au ho s e m he
in e ed-U shape ela ionship be ween he s ic ness o RoOs and he egional
alue added a “La e cu e o RoOs.”
The au ho s ake hei model o he da a, using de ailed in o ma ion on
he sou ce o engines and ansmissions o all ca models assembled in No h
Ame ica. Since he U.S. MFN a i on au omobiles is only 2.5% and only a
ac ion o he ca s assembled in Mexico a e expo ed (a ac ion ha a ies by
model), i s ands o eason ha some p oduce s will be willing o pay he a i
in lieu o sou cing mo e expensi e pa s. Simula ion o he model p edic s ha
16.9% o models ha complied wi h a binding RoO unde NAFTA will become
non-complian unde he igh e RoOs o USMCA. These i ms accoun o a
p edic ed 11.9% all in employmen in plan s ha manu ac u e pa s, o se ing
he 23.6% ise in employmen ha he model p edic s o plan s ha choose o
comply wi h a binding RoO ex pos . O e all, he model p edic s employmen
gains o only 2.3% in plan s ha manu ac u e pa s, much smalle han he 20%
gains ha would ha e been expec ed had hey assumed ha all ca lines comply
wi h he new RoOs. Had he new RoOs been se a 85%, as he Uni ed S a es
had ini ially demanded, employmen in pa s manu ac u ing ac ually would
ha e declined, acco ding o he model’s es ima es.
The au ho s also simula e he e ec s o BREXIT on he Eu opean ca indus-
y. A e B i ain’s wi hd awal om he Eu opean Union, he coun y nego i-
a ed a new pac known as he Eu opean Union and Uni ed Kingdom T ade
and Coope a ion Ac . The TCA equi es egional con en o 55% o B i ish
ca s o en e du y ee in o he Eu opean Union and o EU ca s o en e simi-
la ly in o G ea B i ain. The au ho s no e ha 85% o ca lines al eady sa is ied
his equi emen p io o TCA, sugges ing ha he plan s manu ac u ing hese
models will be li le a ec ed. On he o he hand, he MFN a i in bo h B i -
ain and he Eu opean Union is 10%, much highe han he le el in he Uni ed
S a es, sugges ing ha i ms ha did no al eady sa is y he new RoOs be o e-
hand may well choose no o comply. The simula ions again indica e ha a all
in employmen in pa s manu ac u e s se ing i ms ha choose no o comply
will o se he employmen gains in i ms ha comply wi h a newly binding
cons ain . O e all, he gains in employmen a e p edic ed o be less han 1%.
The analysis by Head e al. in Chap e 4 emphasizes ha o e ly s ic LCRs
ha a e no subjec o manda e bu a he a e suppo ed by iscal incen i es
may be coun e p oduc i e. Fi ms ha would ha e complied wi h a milde
es ic ion may op ou once he equi emen s become oo se e e.
8 Gene M. G ossman and Lili Yan Ing
Chap e 5 by Kun Cai and Zhi Wang analyzes LCR policies in China,
wi h a pa icula emphasis on he Made in China 2025 policy. The chap e
begins wi h an o e iew o China’s LCR amewo k, no ing ha hese poli-
cies we e explici p io o China’s accession o he WTO bu became mo e
opaque a e wa d. Al hough he legally manda ed LCR pe cen ages o goods
o se ices we e g adually li ed, implici localiza ion biases ing ained in he
implemen a ion o indus ial policies ook hei place. On he su ace, hese
policies ea p oduce s simila ly ega dless o na ionali y, bu in p ac ice, only
indigenous i ms can bene i om many o he p e e en ial policies, and when
o eign p oduce s a e able o pa icipa e, o en hey a e “encou aged” o
ans e echnology o o sou ce locally. The opaque na u e o he cu en
LCR policies makes hem di icul o measu e p ecisely, bu he au ho s a gue
ha hei e ec s a e e y he e ogeneous ac oss sec o s. Mo eo e , Beijing has
launched a ecen campaign o encou age he de elopmen o mo e ad anced
echnologies a home o ely less on he Uni ed S a es and o he Wes e n
supplie s. China aims o bols e indigenous i ms’ capaci y o inno a ion and
o ha e hem become global leade s in ad anced echnologies. To u he
hese objec i es, a ious go e nmen agencies in China a di e en le els ha e
implemen ed a se ies o indus ial policies, including some implici LCRs ha
bene i local i ms.
Cai and Wang a e especially in e es ed in he impac s o China’s LCR
policies on he domes ic alue added embodied in expo s. They ex end he
Koopman e al. (2012) me hodology o using wo ld inpu -ou pu ma ices o
a ibu e alue added in expo s o sou ce coun ies so as o allow o ci cum-
s ances as in China, whe e a sizable ac ion o expo ed goods is p oduced
in expo p ocessing acili ies ha a e able o impo hei aw ma e ials and
componen s du y ee. The au ho s use he inpu -ou pu ma ices published
by China’s Na ional Bu eau o S a is ics o 2007, 2012, and 2017, along wi h
de ailed ade s a is ics om China Cus oms. Using he changes ha occu ed
du ing he pe iods be ween he publica ion o hese da a, hey es ima e he
impac o China’s LCR policies on he ac ion o domes ic alue added in
o al expo s, manu ac u ing expo s, and expo s o o eign-owned i ms. In
he agg ega e, domes ic alue added in expo s ose om 64.6% in 2007 o
65.3% in 2012 and o 69.9% in 2017. The es ima ed gains a e smalle when
only manu ac u ing is conside ed, and hey mask opposing ends o no mal
and p ocessing expo s. Whe eas he domes ic alue-added sha e in no mal
manu ac u ing expo s inc eased om 2007 o 2017, he sha e in p ocessing
expo s ac ually declined. Focusing only on o eign owned i ms, hey ind a
sligh decline o e he pe iod.
The au ho s also s udy changes in domes ic alue added in expo s a he
indus y le el, compa ing only 2012 wi h 2017, because indus y de ini ions
changed a e 2007. Roughly one- hi d o he 68 indus ies, accoun ing o
20% o o al expo s, had a domes ic alue-added sha e in expo s be ween
51% and 75% in 2012. These we e mos ly capi al-in ensi e indus ies such
as basic chemicals, i on and s eel, li ing and handling equipmen , pumps,
In oduc ion 9
gene a o s, and ba e ies. Roughly hal o he indus ies, accoun ing o 25%
o o al expo s, had a domes ic alue sha e in he ea lie yea g ea e han
75%. These indus ies we e mo e labo in ensi e, including ex iles, appa el,
oo wea , lea he , and u ni u e. The lowes domes ic alue-added sha es we e
ound in he mo e echnologically sophis ica ed indus ies, such as compu e s,
communica ion equipmen , and elec onic componen s, whe e o eign own-
e ship plays a majo ole. Compa ing he es ima es o 2012 and 2017, 15 o
he 68 indus ies saw hei domes ic alue-added sha e in expo s ise by mo e
han 5%, and 4 expe ienced an inc ease o g ea e han 9%.
Cai and Wang concede ha hei me hodology does no allow hem o
es o a causal ela ionship be ween LCR policies and domes ic con en ou -
comes. S ill, hei accoun ing decomposi ion leads hem o conclude ha he
a ious LCR policies implici in China’s indus ial s a egy did no seem o
play a signi ican ole in p omo ing inc eased local con en in China’s expo s
be ween 2007 and 2017.
Chap e 6 by Michelle Limen a, Lili Yan Ing, Junian o James Losa i, and
Osca Fe nando is he i s o h ee ha ocuses on LCRs in Indonesia. Indo-
nesia is among he coun ies wi h he highes incidence o local con en es ic-
ions, wi h LCR policies da ing back o he 1950s. LCRs ha e long appealed
o Indonesian policymake s who a e keen o encou age domes ic alue added
and expand employmen in he indus ial sec o .
Limen a e al. conside in de ail whe he Indonesia’s LCR policies a e con-
sis en wi h i s commi men s unde i s a ious mul ila e al and egional ade
and in es men ag eemen s. They begin by ou lining he jus i ica ions o e ed
by he Indonesian go e nmen o i s a ious policies. These include he coun-
y’s goal o upg ading compe i i eness and elimina ing i s ade de ici , which
he go e nmen belie es can be achie ed by inc easing local alue added in
aded sec o s.
The use o LCRs began in Indonesia wi h he “Ben eng P og am” (1950–
1957), ollowed by he “Dele ion P og am” (1974–1993) and he “Na ional
Ca P og am” (1996). A e a pe iod o do mancy, Indonesia esumed in
2009 a local con en s a egy wi h he implemen a ion o he “Inc eased
Use o Domes ic P oduc ion” p og am. This policy in oduced local con en
equi emen s o goods and se ices pu chased by he go e nmen . In 2013,
Indonesia implemen ed he a o emen ioned egula ion s ipula ing a minimum
pe cen age o domes ic oil and gas as inpu in o he local p oduc ion o many
goods. Since ha ime, many membe s o he WTO ha e challenged Indone-
sia’s policies in he T ade-Rela ed In es men Measu es (TRIMS) Commi ee,
claiming ha hey un coun e o commi men s ha Indonesia made o i s
ade and in es men pa ne s. O pa icula conce n o he WTO membe s
ha e been policies egula ing he use o domes ically p oduced ene gy, as well
as policies add essed o he elecommunica ions, pha maceu ical, and e ail
sec o s.
The au ho s p oceed o e iew Indonesia’s ele an commi men s unde
TRIMS, he Gene al Ag eemen on Ta i s and T ade (GATT), he Gene al
16 Do o hee Flaig and Susan F. S one
compe i ion o he a ge indus y and lead o a de e io a ion in p oduc
quali y, as hey educe access o echnologically ad anced inpu s and p o ide
li le incen i e o in e nal inno a ion (Hu baue e al., 2013). Co up ion and
a o i ism om opaque and ad hoc policy design can also inc ease he long-
un nega i e impac o hese policies (Kun ze and Moe enhou , 2013; Weiss,
2016). The objec i es o LCRs – such as building up a compe i i e indus y
h ough s onge indus ial links, c ea ing new supplie s and backwa d link-
ages – is a ely ob ained (Hu baue e al., 2013). In mos cases, LCRs isola e
high-cos p oduce s om global compe i ion and inno a ion and esul in
insu icien incen i es o esea ch and de elopmen (R&D) in es men s.
Mos LCRs implemen ed ha e employmen as hei p ima y objec i e,
explici ly o implici ly s a ed. The use o domes ic supplie s has an immedia e
job e ec ha can be pa icula ly powe ul du ing economic down u ns. LCRs
wi h employmen objec i es encompass goals such as c ea ing new jobs, c ea -
ing highe -skilled jobs, and inc easing na ional income. Howe e , hese poli-
cies o en sac i ice job g ow h in he gene al economy o job g ow h in he
a ge ed sec o . The Uni ed S a es (US) Buy Ame ican Ac , 1933, is es ima ed
o ha e cos abou 360,000 jobs in non- a ge ed sec o s h oughou he US
economy (Dixon, Rimme , and Waschik, 2018).
Measu es a ge ing echnological de elopmen end o equi e o eign
i ms o ans e echnology o domes ic ope a ions o domes ic supplie s. To
he implemen ing economy, his echnology ans e is seen as an e icien way
o inc easing compe i i eness in wo ld ma ke s. The speci ic goals o echno-
logical de elopmen can include imp o ing echnological capaci y and spu -
ing inno a ion a he na ional, egional, o indus y le el. Howe e , hese
goals a e o en unde mined by a lack o a ailable skills.
Fac o s ha impac he abili y o mee an LCR’s s a ed policy objec i e
can be cha ac e ized ac oss ou a eas: (i) ma ke size and s abili y, (ii) policy
design and cohe ence, (iii) he es ic i eness o he LCRs, and (i ) he domes-
ic indus ial base (Kaziboni and S e n, 2021). I he domes ic ma ke is small
o uns able and canno mee he demands o he local p oduce s, he e is a
isk ha hese local p oduce s exi he ma ke , de ea ing any employmen o
echnology- ans e goals. A policy ha is oo ague is usually unen o ceable,
and i will no be e ec i e. Mo eo e , i he LCR is se a a le el ha does
no ha e a meaning ul impac on he sou cing decisions o he impo e , no
change will occu . Some LCRs a e se below exis ing sou cing le els, ha ing
no impac on ma ke decisions. Finally, i he e a e no c edible p oduce s o
he inpu , hen any LCR policy will simply lead o i ms exi ing he ma ke .
Go e nmen s o en a emp o achie e se e al policy objec i es wi h one
LCR (e.g., inc ease ou pu and employmen along wi h echnology ans-
e ). In hese ins ances, he policy o en ends up ha ing con adic o y ou -
comes (e.g., inc easing p oduc ion bu dec easing p oduc i i y i labo is
unable o implemen he ans e ed echnology). Subsequen p oduc i i y
declines, coupled wi h sho ages o su icien ly skilled labo , may lead o an
o e all lowe le el o labo demand, showing ha one policy is o en unable

Localiza ion measu es 17
o hi wo a ge s (Fang, 2020). Policymake s o en neglec o iden i y p ac i-
cal challenges ha migh nega i ely impac he e iciency and e ec i eness o
LCRs when adop ing hese measu es. The e is limi ed conside a ion o he
ac ha he economic impac o LCRs is complex and depends on se e al
a iables, including hei in e ac ion ac oss policy a eas (Lin and Weng, 2020).
The poli ical in luence o p oduce s can also a ec he le el o local pa ici-
pa ion. Ablo (2017) shows ha while LCRs ha e he po en ial o p omo e
links be ween some sec o s and he es o he economy, he deg ee o which
p oduce s suppo he go e nmen may limi he ex en o which signi ican
local con en can be achie ed. Thus, he poli ical ela ionship be ween he
go e nmen and indus y, he capaci y o local SMEs, and he echniques and
p ac ices o mul ina ional companies will all ha e a bea ing on he e ec i eness
o any LCR policy.
By limi ing compe i ion and inpu choices, he a ge i m aces a limi ed, i
no single, supplie – leading o highe inpu cos s along he p oduc ion line.
This can also a ec he quali y o he ma e ial/inpu , which can u he impac
cos s. These highe cos s a e hen passed on, in whole o in pa , downs eam,
inc easing cos s o bo h he consume and p oduce . This ul ima ely means
highe p ices o he end use .
Ano he way LCRs can ha e a de imen al e ec on he domes ic ma ke
is h ough esou ce alloca ion e ec s. Resou ces being shi ed o he a ge ed
i m/indus y become sca ce elsewhe e in he domes ic ma ke , aising cos s
o o he i ms. Ano he longe - e m spillo e in he domes ic economy ela es
o inno a ion and skills de elopmen . While e idence shows ha a ge ed
i ms will unde ake aining and de elopmen o local i ms o mee LCR
a ge s (Ramdoo, 2015), his lea es li le mo i a ion in he domes ic i ms o
inno a e, as hey ha e li le o no compe i ion o spu such inno a ion. This
can a ec he deg ee o which i ms ac oss he economy engage in inno a i e
beha io . This lack o inno a ion can c ea e he longes -las ing, mos de i-
men al e ec s o an LCR. A ecen s udy (Kingi i and Okemwa, 2022) shows
ha local con en policies ha e no had a posi i e impac on echnology de el-
opmen in he Kenyan enewable ene gy sec o .
F om he pe spec i e o public expendi u es, LCRs educe impo a i e -
enue as well as po en ial co po a e ax e enue by inc easing he ope a ing
cos s and educing he p o i abili y o mul ina ional companies (Kols ad and
Kinyondo, 2017). Imposing LCRs has he oppo uni y cos o o gone axes,
which could be used in mo e e ec i e ways o imp o e de elopmen p os-
pec s. In he case o incen i e-based LCRs, go e nmen s o en o ego e enue
om o p o ide incen i e paymen s o hese en u es, which has di ec public
expendi u e implica ions. In addi ion, pas expe iences o esou ce- ich de el-
oping coun ies indica e ha local con en policies can exace ba e key p oblems
o pa onage and en -seeking, inc easing he dange ha he esou ce weal h
will unde mine a he han help de elopmen (Kols ad and Kinyondo, 2017).
This chap e ex ends and upda es ou ea lie s udy om 2015 (S one, Mes-
sen , and Flaig, 2015). I discusses ecen ins ances o LCRs and, by modeling
18 Do o hee Flaig and Susan F. S one
ep esen a i e examples o he policy, p o ides some insigh s on he impac
hey ha e on he economy. Sec ion 2 examines ecen ends in hei imple-
men a ion. Sec ion 3 ou lines he measu es examined o his s udy and he
modeling app oach adop ed. Sec ion 4 p o ides he simula ion esul s, while
Sec ion 5 p esen s some concluding hough s.
2 Recen LCR implemen a ion
The use o LCRs has been accele a ing in ecen yea s. Acco ding o Global
T ade Ale ,1 coun ies pu in place mo e han 500 indi idual local con en
measu es om 2014 o 2020 compa ed wi h less han 200 measu es om
2008 o 2013 – a 155% inc ease (Figu e 2.1). No only has he numbe
o measu es inc eased, he way in which hey ha e been implemen ed has
changed. The less anspa en ypes o ins umen s ha e isen, wi h he Global
T ade Ale measu es a ed ambe (“likely” o cause disc imina ion) inc easing
signi ican ly o e hose a ed ed (“almos ce ainly” causing disc imina ion
agains o eign companies). Tha is, he numbe o measu es ha clea ly s a e
he ype and equi emen s o an LCR es ic ion has declined as a sha e o he
o al numbe o LCRs imposed. Only 6% o he measu es we e a ed ambe
du ing 2008–2013, whe eas 32% o he measu es we e a ed ambe du ing
2014–2020.
On he su ace, India, Ge many, and he Uni ed Kingdom appea o be
he main use s o LCRs (Figu e 2.1). Howe e , he ac ual impac o hese
measu es is much mo e complex. Fi s , as no ed, Global T ade Ale p o ides a
coun o he incidence o a measu e, no i s impac . Thus, a measu e a ec ing a
speci ic small sec o (e.g., A gen ina’s LCR on ce ain ypes o medicines om
Spain) coun s he same as a gene al measu e a ec ing a much la ge sec o
(e.g., Russian es ic ions on au o pa s o he au omo i e sec o o all ade
pa ne s). Second, a igh ly binding measu e coun s he same as one ha is
only pa ially binding o no binding a all.
In addi ion, many economies implemen hese measu es a a a ie y o ju is-
dic ional le els. La ge economies wi h decen alized economies (e.g., he US
and India) ha e many measu es implemen ed a he le el o he indi idual
s a e o e en he local le el o go e nmen . O he s ha e a much mo e cen-
alized app oach, while s ill o he s (e.g., F ance and Ge many) ha e policies
implemen ed a he sup ana ional le el. The a ie y o ins ances makes iden i-
ying and measu ing LCRs challenging.
Global T ade Ale epo s LCRs ac oss ou in e en ion ypes: labo ,
ope a ions, sou cing, and incen i izing. Labo gene ally e e s o LCRs
ied di ec ly o hi ing and employmen equi emen s. Ope a ions a e LCRs
ha ha e equi emen s conce ning a i m’s abili y/pe mission o ope -
a e in he domes ic ma ke . Sou cing e e s o he equi emen s o sou ce
inpu s (pa s and componen s) om local manu ac u e s. Incen i es a e ax
o o he go e nmen bene i s ecei ed when buying o using local inpu s,
ope a ions, o labo .
Localiza ion measu es 19
020406080100 120140 160
Uni ed S a es
Uni ed Kingdom
Tu key
Saudi A abia
Russia
Indonesia
India
Ge many
China
B azil
Aus alia
A gen ina
To al 2008-2013 To al 2014-2020
Figu e 2.1 Incidence o Local Con en Requi emen s, 2008–2020* (Selec ed
Economies)
BNDES = The B azilian De elopmen Bank, LCR = local con en equi emen .
No e: The numbe s ep esen indi idual ins ances o LCRs, so hei ac ual impac is no di ec ly
compa able. Fo example, in 2015, BNDES inanced h ee wind pa ks wi h $260 million. As
hese measu es a ec ed di e en ading pa ne s and di e en sec o s wi hin wind u bines, hey
accoun ed o 57 o he 594 measu es o local sou cing o B azil. A gen ina’s LCR on mining
coun s as one measu e ye a ec s mo e han 10 sec o s measu ed a he 2-digi le el ac oss all
ading pa ne s.
* The la ge numbe o LCRs a ibu ed o Ge many all ela e o suppo gi en unde Ge many’s
Expo C edi s p og am.
Sou ce: Global T ade Ale (2022).
Figu e 2.2 shows how he ypes o LCRs used by go e nmen s ha e changed
o e ime. Whe eas in he pe iod igh a e he Global Financial C isis (2008–
2013) LCRs ocused on ensu ing ha inpu s and labo we e sou ced locally,
he la e pe iod (2014–2020) swi ched o incen i izing i ms by o e ing ax
b eaks, p e e en ial lending, o o he pe ks ied o using local inpu s o es ab-
lishing local p oduc ion. The numbe o measu es o e ing incen i es mo e
han doubled om less han 6% o all measu es in 2008–2013 o mo e han
12% in 2014–2020.
As desc ibed by De inge e al. (2018), Global T ade Ale anked acked
policy in e en ions by hei possible damage o o eign ade and in es men .
Acco ding o his anking, LCRs ep esen ing public p ocu emen localiza ion
a e anked i h, and measu es ep esen ing o he localiza ion equi emen s a e
anked se en h (E ene and F i z, 2021). Ranking ahead o LCRs a e (i) s a e
aid, (ii) ade de ense, (iii) impo a i s, and (i ) expo axes o es ic ions,
wi h ade inance measu es in six h place (E ene and F i z, 2021). LCRs
20 Do o hee Flaig and Susan F. S one
a e also equen ly implemen ed in he o m o disc imina o y go e nmen
p ocu emen . These measu es educe he numbe o eligible i ms allowed o
compe e in a ma ke and hus dec ease ou pu and employmen while inc eas-
ing p ocu emen cos s and ma ke powe (OECD, 2020).
Measu es ela ed o da a localiza ion a e among he as es -g owing ypes o
LCR measu es. Such measu es a emp o con ol he mo emen o pe sonal
da a and local s o age and p ocessing o da a (López González, Casalini, and
Po as, 2022). As da a low is becoming an essen ial aspec o ade, ela ed
LCR measu es a ec mos sec o s wi hin an economy (OECD, 2020). Some
expe s pe cei e his ype o p o ec ionism as “pe haps oday’s g ea es h ea
o he u he libe aliza ion o he global ading sys em” (Ezell, A kinson, and
Wein, 2013).2 A he same ime, he inc easing connec ion o ade and da a
lows may also p o ide a s onge a ionale o such measu es o ensu e p i acy
and iden i y secu i y.
3 De ining measu es o be add essed
I has been a gued ha localiza ion ba ie s add o he cos o doing business
domes ically and in e na ionally, leading o a dis o ion o wo ld ade lows
and los ma ke oppo uni ies. Howe e , he s udies a emp ing o quan i y
hese impac s ac oss global ma ke s ha e been limi ed. This chap e will upda e
one such a emp (S one, Messen , and Flaig, 2015) by es ima ing he impac
LCRs implemen ed om
2008 o 2013
Labo Sou cing
Ope a ions Incen i izing
LCRs implemen ed om
2014 o 2020
Labo Sou cing
Ope a ions Incen i izing
Figu e 2.2 Type o Local Con en Requi emen s
LCR = local con en equi emen .
Sou ce: Global T ade Ale (2022).
Localiza ion measu es 21
o a se o LCRs on in e na ional ade, using he O ganisa ion o Economic
Co-ope a ion and De elopmen (OECD) ade model, METRO. This se o
LCRs is de ined om in o ma ion aken om se e al da a sou ces and cons i-
u es cu en in- o ce LCR policies ha we e pu in place om 2014 o 2020.3
Simila o he wo k unde aken in S one, Messen , and Flaig (2015), se e al
sou ces we e consul ed. These include he ollowing:
• Pe e son Ins i u e o In e na ional Economics Local Con en Requi e-
men s: A Global P oblem (Hu baue e al., 2013)
• Global T ade Ale online da abase
• Eu opean Commission (2022) Ma ke Access Da abase
• Wo ld Bank (2022) Tempo a y T ade Ba ie s Da abase
• Wo ld T ade O ganiza ion (2022) T ade Moni o ing Da abase
Mo e han 565 measu es we e conside ed o he s udy. All he iden i ied
LCRs we e hen e iewed o assess hei a ini y o quan i ica ion. The quan-
i a i e analysis p esen ed he e ocuses on measu es ha end o be he mos
ade dis o ing. These a e measu es ha es ic access o ma ke s and meas-
u es ha ende p ice p e e ences ied o a speci ic le el o domes ic con en .
Inpu measu es ha de e mine ma ke access accoun ed o mos o he meas-
u es examined o his epo . To a i e a a lis o measu es whose impac s
could be quan i ied, se e al c i e ia had o be me . Following he discussion on
localiza ion cha ac e is ics al eady, ou cha ac e is ics can be iden i ied:
1 Iden i iable sec o – many o he LCRs we e b oad s a emen s abou “sup-
po ing” domes ic sou cing wi hou di ec e e ence o a pa icula sec o
o egion o economic ac i i y. Fo example, coun ies pu in place LCRs
o go e nmen p ocu emen . These a e blanke policies ha may o may
no be implemen ed in any speci ic sec o . Gi en ha he e is no way o
iden i y which, i any, sec o s we e a ec ed by hese policies, hese measu es
we e excluded om he analysis.
2 Iden i iable es ic ion – i he es ic ion is no clea ly indica ed, i canno
be meaning ully en o ced and hus canno be modeled. Fo example, Tu -
key in oduced localiza ion equi emen s on emo e p og ammable e-SIM
echnologies wi hou s ipula ing he size o hese equi emen s.
3 Su icien size – i he sec o o egion is no signi ican , i will ha e li le
impac on he ma ke . An example is an A gen ine law obliging au omobile
uel p oduce s o use bioe hanol om he no heas o he coun y. Nei he
he bioe hanol ma ke no he no heas egion is su icien ly la ge o be
cap u ed in an economy-wide model.
4 En o ceable es ic ion – he e mus be a domes ic sec o ha can mee he
equi ed demand, and he es ic ion mus be binding (i.e., he measu e is
excluded i he domes ic con en al eady mee s o exceeds he le el called
o in he LCR).

22 Do o hee Flaig and Susan F. S one
Many o he ins umen s examined we e implemen ed a he subna ional
le el, which he model does no co e . As no ed abo e, policies ela ed o
b oad goals such as na ional secu i y o go e nmen p ocu emen can be
applied o e mul iple sec o s o no and a e a he disc e ion o go e n-
men agen s. Finally, o many o he measu es epo ed, we we e unable o
de e mine i hey a e s ill in o ce. Thus, hese we e also d opped om he
analysis.
Applying hese c i e ia, we we e able o iden i y se en LCRs ha mee he
condi ions necessa y o modeling. They p o ide a good sample o he ypes
o LCR measu es applied and he economies applying hem. Bo h de eloped
and de eloping coun ies a e included in he analysis. This s udy es ima es
he economic impac o LCRs imposed in selec ed subsec o s o au omobiles,
mining, medical supplies, elecommunica ions, and anspo . Finally, some
“buy local” p ocu emen p o isions a e a such a le el ha noncompliance
would unde mine a i m’s compe i i e posi ion, making hem a “ equi e-
men .” Thus, we include examples o wo such go e nmen p ocu emen
p o isions.
A gen ine au omo i e sec o . Since 2016, A gen ine ca p oduce s can
ob ain a ax c edi , allowing hem o de e alue-added ax (VAT). This ax
c edi is condi ional on a minimum LCR in he inal p oduc . The mini-
mum LCR is 30% o ca s, aile s, engines, and ag icul u al ehicles; 25%
o ucks; and 10% o au omobile pa s. The ax bene i is dependen on
he le el o local con en and anges om 4% o 15% o he sales alue. The
A gen ine mo o ehicle indus y has an app oxima e sales alue o $4.5
billion, so he alue o his c edi is $180 million o $675 million.
B azilian elecommunica ions sec o . In 2017, B azil implemen ed a equi e-
men ha he o e all le el o local con en in he equipmen used in i s 4G
ne wo ks mus be a leas 70%. The es ima ed size o he elecommunica-
ions sec o in B azil is mo e han $8.5 million, wi h almos 200 million 4G
b oadband subsc ibe s. The es ima ed in es men in elecommunica ions
equipmen was abou $5.8 billion in 2018, implying a po en ial ma ke o
mo e han $4 billion solely a ailable o domes ic supplie s. This equi e-
men comes on op o exis ing LCRs in he B azilian elecommunica ions
sec o (S one, Messen , and Flaig, 2015).
B azilian go e nmen p ocu emen . In 2013 and 2014, he go e nmen
o B azil inc eased he p e e ence ma gins o he public p ocu emen
o a a ie y o na ionally p oduced goods. P e e ence ma gins a e he
maximum ex en o which he p ice quo ed by a local supplie may be
abo e ha o a compe i o . The new ma gins ange be ween 9% and 25%
and include IT (15%), ac o s (20%), ai planes (9%–25%), a ious IT-
ela ed goods and se ices (15%–25%), capi al goods (15%–20%), and oys
(10%). These ma gins we e deemed su icien ly high, gi en he a ious
ma ke sizes, o i ms o be compelled o use local supplie s o emain
compe i i e.
Localiza ion measu es 23
Indonesian au omo i e indus y. To p omo e he Indonesian mo o
ehicle componen s indus y, se e al egula ions implemen ed om
2014 o 2017 equi e de ac o he en i e assembly p ocess o mo o
ehicles and mo o cycles o ake place locally. To ha e access o he
Indonesian ma ke , he LCR equi es all majo ehicle componen s and
ela ed se ices o ake place wi hin Indonesia. The Indonesian au o-
mo i e ma ke accoun s o abou 10% o Indonesia’s g oss domes ic
p oduc (GDP), o oughly $10 billion, wi h almos 25% des ined o
expo ma ke s.
Saudi A abian medical supplies. Since 2020, i is manda o y o go e n-
men agencies o p ocu e a ange o medical supplies domes ically, such as
s e ilize s, ace masks, pe sonal p o ec i e equipmen o heal h p ac i ion-
e s, s e iliza ion supplies o medical ools, and o he medical supplies. The
ma ke o medical supplies in Saudi A abia is es ima ed o be mo e han
$2 billion.
Sou h A ican mining. The Mining Cha e adop ed in 2018 by Sou h A ica
es ablishes a minimum LCR o a leas 70% o mining goods and 80% o
o al se ice expendi u e in he sec o . In addi ion, a leas 21% o min-
ing goods and 50% o se ices mus be p oduced by a domes ic company
owned and con olled by “His o ically Disad an aged Pe sons,” ano he
5% (goods)/5%–15% (se ices) mus be p oduced by companies owned by
women o you h, and 44% (goods)/10% (se ices) by companies compli-
an wi h local Black Economic Empowe men (BEE). In 2018, he mining
sec o accoun ed o $22.5 billion o Sou h A ican GDP and employed an
es ima ed 456,000 wo ke s.
US Buy Ame ica. This a - eaching p og am has many p o isions imple-
men ed on a p e e en ial- ea men basis. Howe e , anspo a ion g an s
ac oss many US s a es ha e a speci ic equi emen o local con en o be
eligible o ecei e unding. Gi en ha he main sou ce o anspo a ion
unding is he ede al go e nmen , o mos US s a es, his amoun s o an
LCR. These p og ams s ipula e ha any public p ojec unded by T anspo -
a ion In es men Gene a ing Economic Reco e y (TIGER) g an s mus
use some le el o domes ically p oduced i on, s eel, and o he manu ac-
u ed goods. The amoun s o bo h he g an s and he LCR a y by indi-
idual s a es and p ojec s wi hin s a es bu a e es ima ed o be wo h mo e
han $4 billion.
Model and da a
LCRs may ha e sho - e m bene i s in e ms o speci ic policy objec i es, bu
ad e se e ec s de elop o e ime and o en ou weigh he sho - e m bene i s.
As wi h any model, no all he impac s o he policy will be ully e lec ed in
he esul s. Howe e , using a compu able gene al equilib ium (CGE) model
allows o he cap u e o hese longe - un impac s, no o men ion he e ec
hese policies ha e on b oade economic ac i i y. The bene i o using a CGE
24 Do o hee Flaig and Susan F. S one
model in his analysis is i s abili y o cap u e impac s beyond he a ge ed sec-
o , showing he e ec s hese measu es ha e on he es o he economy as well
as he global ade en i onmen .
The METRO model (A iola e al., 2020) is based on empi ical da a and
inco po a es unique ea u es o each egion’s economic sys em. The model
is calib a ed o an augmen ed Social Accoun ing Ma ix (SAM) e sion o
he Global T ade Analysis P ojec (GTAP) e sion 10 da abase (Aguia e al.,
2019). The da abase ea u es ade lows disagg ega ed by use ca ego ies
de i ed om he OECD T ade in Value Added (TiVA) da abase as well as
Uni ed Na ions (UN) sou ces and bila e al emi ance da a om GTAP sa el-
li e da a, i.e., GMIG2 (Walmsley, Win e s, and Ahmed, 2007).4 These ca ego-
ies a e in e media e use, use by households, use by go e nmen , and use by
business/in es men .
The sec o de ail depends on he sec o co e age in he GTAP da abase,
which dis inguishes 65 sec o s wi h mo e de ail in ag i- ood, and o he sec-
o s depic ed la gely on he In e na ional S anda d Indus ial Classi ica ion
o All Economic Ac i i ies (ISIC) 2-digi le el. This sec o co e age does
no allow modeling o LCR measu es a a de ailed sec o le el. The da a-
base is agg ega ed o his s udy, as de ailed in Table 2.A1. The ag icul u e,
ood, and ex ile sec o s in he GTAP da abase a e agg ega ed; he s udy
ea u es 43 sec o s, o which 4 a e ela ed o he ex ac ion indus ies, 17
a e ela ed o manu ac u ing, and 20 a e in he se ice sec o . The da abase
dis inguishes eigh ac o s o p oduc ion, wo skilled and h ee skilled labo
ypes, capi al, land, and na u al esou ces. Coun ies a e agg ega ed o la ge
egional composi es, singling ou he ele an coun ies o he selec ed poli-
cies examined.
METRO is a compa a i e s a ic global CGE model.5 Global CGE models
link a ious ma ke s, economies, and sec o s – employing economic heo y o
show in e linkages be ween agen s, sec o s, and economies by simul aneously
de e mining p ices and quan i ies. The s eng h o METRO lies in he de ailed
ade s uc u e and he di e en ia ion o p oduc ion and consump ion com-
modi ies by use – in e media e, household, go e nmen , and capi al consump-
ion. The di e en ia ion o commodi y supply, and hus he esul ing ade
lows, by use ca ego y imp o es he abili y o depic and analyze speci ic policy
ins umen s such as LCRs. The emainde o his sec ion gi es an o e iew o
key ea u es o he model. Please e e o he METRO model documen a ion
(A iola e al., 2020) o a de ailed and comple e desc ip ion o he model da a
and equa ions.
The model is based on a se ies o egional SAMs, which a e linked h ough
ade ela ionships. Table 2.1 depic s he s uc u e o he da abase, whe e
income lows a e ead ac oss ows and columns depic expendi u es. House-
holds, o example, ecei e income om ac o se ices and emi ance in lows
and spend his income on p i a e consump ion, di ec axes, emi ance ou -
lows, and sa ings. Following accoun ing ules and depic ing a comple e and
ci cula sys em, ow and column sums mus equalize. Thus, he o al income
io al Rela ionshipse o he Da abase and Beha uc u METRO Model – S Table 2.1
Expendi u e
Income low
Commodi ies (by sec o ,
impo ed and domes ic,
by use ca ego y)
Use ca ego y:
Ac i i ies (by
sec o )
Fac o s Use ca ego y:
Household
Use ca ego y:
Go e nmen
Use ca ego y:
Capi al
Ma gins
(bila e al, by use
ca ego y)
Res o he wo ld
(bila e al, by use
ca ego y)
Commodi ies
(by sec o ,
impo ed and
domes ic, by
use ca ego y)
Use ca ego y:
Ac i i ies (by
sec o )
Fac o s (5
labo ypes,
capi al, land,
and na u al
esou ces)
Use ca ego y:
Household
Use ca ego y:
Go e nmen
Use ca ego y:
Capi al
Ma gins
(bila e al, by
use ca ego y)
Res o he
wo ld
(bila e al, by
use ca ego y)
In e media e P i a e Public In es men : Ma gins expo s: Expo s: Th ee-
inpu s: consump ion: consump ion: Fixed sha es Th ee-s age s age CET
Leon ie S one-Gea y Fixed sha es CET unc ions unc ions
inpu -ou pu u ili y unc ions
coe icien s
Domes ic supply:
To al supply om
domes ic p oduc ion
Value added:
mul i-
le el CES
p oduc ion
unc ions
Fac o income: Remi ance
Fixed sha es o in lows
ac o income
Impo a i s, expo P oduc ion Fac o axes: Di ec axes:
axes, sales axes: Ad axes: Ad A e age ax A e age ax a es
alo em and speci ic alo em a es
Dep ecia ion: P i a e sa ings: Public sa ings: Cu en accoun Fo eign sa ings:
Sha es o ac o Sha es o Residual balance on Cu en
income household income ma gins ade accoun balance
T ade and anspo
ma gins: Fixed
coe icien s
Impo s: Th ee-le el Remi ance
CES ou lows: Fixed
p opo ion o
disposable income
ma ion., CET = cons an elas ici y o ans o
elde (2013).
CES = cons an elas ici y o subs i u ion
om McDonald and Thie : Adap ed ceSou
32 Do o hee Flaig and Susan F. S one
cen age Changesoduc ion by Sec o , Pe s and P Model Resul s – E ec s on Impo Table 2.3
A gen ina– B azil– elecom Indonesia–au omobiles Saudi A abia–GP Sou h A ica–Mining
au omobiles medicals
Impo s P oduc ion Impo s P oduc ion Impo s P oduc ion Impo s P oduc ion Impo s P oduc io
Ag icul u e 2.0 −1.8 0.0 0.0 0.9 −0.6 0.0 0.0 0.0* 0.0
Coal −0.2 −3.2 0.0 0.0 0.9 −1.4 0.0 −0.1 −0.1* 0.1
Oil 2.7 −1.3 0.0 0.0 1.1 −1.2 0.1 −0.1 0.0* 0.4
Gas 7.6 −5.4 0.1 −0.1 7.3 −6.1 0.2 0.1 −0.5* 1.4
Mine als 0.8 −0.7 0.0 0.0 0.5 −1.9 0.0 0.0 −0.8* −0.8
Food 3.6 −1.7 0.0 0.0 1.9 −0.8 0.0 0.0 0.0* 0.0
Tex iles 6.0 −0.8 0.0 0.0 1.3 −2.9 0.0* −0.1 0.0* 0.0
Wood p oduc s 4.8 −0.2 0.0 0.0 3.8 −1.1 0.0* 0.0 0.0* −0.1
Pape p oduc ion, 3.7 −0.9 0.0 0.0 1.1 −2.0 0.1* 0.0 0.0* 0.0
publishing
Pe oleum, coal p oduc s 1.0 −1.0 0.0 0.0 0.9 −0.2 0.1 0.0 0.0* 0.0
Chemical p oduc s 1.6 −2.7 0.0 0.0 1.2 −2.3 −2.0* 0.4 −0.3* 0.1
Basic pha maceu icals 0.7* −3.7 0.0 0.0 2.6 −1.0 −2.3* 3.8 0.0* 0.0
Rubbe and plas ic 5.5 0.8 0.0 0.0 3.3 −1.0 0.0* 0.0 0.0* 0.0
Mine al p oduc s 4.0 0.1 0.0 0.0 2.9 −0.1 0.1 0.0 0.0* 0.0
Fe ous me als 7.6 1.3 0.1 0.0 3.5 −1.3 0.0 0.0 0.2* 0.1
Me als 6.8 −5.2 0.1 0.0 4.1 −4.2 0.0 −0.2 −0.1* −0.3
Me al p oduc s 8.4 5.7 0.1 0.0 3.2 0.4 0.1 0.0 0.1* 0.0
Compu e , elec onics 1.2* −0.8 0.0 0.0 0.3 −4.6 0.0 0.0 0.0* 0.1
Elec ical equipmen 1.4* 3.7 −0.7* 0.6 2.1 −4.8 0.0* −0.2 −0.1* 0.2
n
Machine y, equipmen 0.9* −4.2 −0.1* 0.1 1.4 −4.4 0.0 −0.1 −0.9* 0.6
Mo o ehicles and pa s −3.8* 39.2 0.0 0.0 −68.1* 43.3 0.0 0.0 0.0* 0.0
T anspo equipmen −9.7 17.1 0.0 0.0 −44.3* 23.3 0.0 −0.1 −0.9* 3.3
Manu ac u es 5.3 −0.2 0.0 0.0 3.1 −2.8 0.0 0.0 0.0* 0.0
T anspo 1.3–1.8 −0.7–−0.1 0.0 0.0 1.3–2.1 0.1–3.6 0.0 0.0 −0.3–0.0 0.0–0.1
Communica ion 2.8 −0.3 0.0 0.0 1.5 −0.3 0.1 0.0 0.0 0.0
U ili ies, cons uc ion 2.5–3.4 0.1–0.5 0.0 0.0 1.2–2.7 −0.5–0.0 0.0–0.1 0.0 −0.4–0.0 −0.1–0.3
O he se ice sec o s 1.5–2.9 −0.6–0.2 0.0 0.0 1.0–2.0 −1.0–0.0 0.0–0.1 0.0 0.0 0.0–0.1
No e: The anspo sec o co e s he h ee modes (ai , land, and sea) and hus is epo ed as a ange.
* Sec o s acing a binding local con en equi emen .
Sou ce: Model esul s.

Localiza ion measu es 33
Table 2.3 Model Resul s – E ec s on Impo s and P oduc ion by Sec o , Pe cen age Changes
A gen ina–
au omobiles
B azil– elecom Indonesia–au omobiles Saudi A abia–GP
medicals
Sou h A ica–Mining
Impo s P oduc ion Impo s P oduc ion Impo s P oduc ion Impo s P oduc ion Impo s P oduc ion
Ag icul u e 2.0 −1.8 0.0 0.0 0.9 −0.6 0.0 0.0 0.0* 0.0
Coal −0.2 −3.2 0.0 0.0 0.9 −1.4 0.0 −0.1 −0.1* 0.1
Oil 2.7 −1.3 0.0 0.0 1.1 −1.2 0.1 −0.1 0.0* 0.4
Gas 7.6 −5.4 0.1 −0.1 7.3 −6.1 0.2 0.1 −0.5* 1.4
Mine als 0.8 −0.7 0.0 0.0 0.5 −1.9 0.0 0.0 −0.8* −0.8
Food 3.6 −1.7 0.0 0.0 1.9 −0.8 0.0 0.0 0.0* 0.0
Tex iles 6.0 −0.8 0.0 0.0 1.3 −2.9 0.0* −0.1 0.0* 0.0
Wood p oduc s 4.8 −0.2 0.0 0.0 3.8 −1.1 0.0* 0.0 0.0* −0.1
Pape p oduc ion,
publishing
3.7 −0.9 0.0 0.0 1.1 −2.0 0.1* 0.0 0.0* 0.0
Pe oleum, coal p oduc s 1.0 −1.0 0.0 0.0 0.9 −0.2 0.1 0.0 0.0* 0.0
Chemical p oduc s 1.6 −2.7 0.0 0.0 1.2 −2.3 −2.0* 0.4 −0.3* 0.1
Basic pha maceu icals 0.7* −3.7 0.0 0.0 2.6 −1.0 −2.3* 3.8 0.0* 0.0
Rubbe and plas ic 5.5 0.8 0.0 0.0 3.3 −1.0 0.0* 0.0 0.0* 0.0
Mine al p oduc s 4.0 0.1 0.0 0.0 2.9 −0.1 0.1 0.0 0.0* 0.0
Fe ous me als 7.6 1.3 0.1 0.0 3.5 −1.3 0.0 0.0 0.2* 0.1
Me als 6.8 −5.2 0.1 0.0 4.1 −4.2 0.0 −0.2 −0.1* −0.3
Me al p oduc s 8.4 5.7 0.1 0.0 3.2 0.4 0.1 0.0 0.1* 0.0
Compu e , elec onics 1.2* −0.8 0.0 0.0 0.3 −4.6 0.0 0.0 0.0* 0.1
Elec ical equipmen 1.4* 3.7 −0.7* 0.6 2.1 −4.8 0.0* −0.2 −0.1* 0.2
Machine y, equipmen 0.9* −4.2 −0.1* 0.1 1.4 −4.4 0.0 −0.1 −0.9* 0.6
Mo o ehicles and pa s −3.8* 39.2 0.0 0.0 −68.1* 43.3 0.0 0.0 0.0* 0.0
T anspo equipmen −9.7 17.1 0.0 0.0 −44.3* 23.3 0.0 −0.1 −0.9* 3.3
Manu ac u es 5.3 −0.2 0.0 0.0 3.1 −2.8 0.0 0.0 0.0* 0.0
T anspo 1.3–1.8 −0.7–−0.1 0.0 0.0 1.3–2.1 0.1–3.6 0.0 0.0 −0.3–0.0 0.0–0.1
Communica ion 2.8 −0.3 0.0 0.0 1.5 −0.3 0.1 0.0 0.0 0.0
U ili ies, cons uc ion 2.5–3.4 0.1–0.5 0.0 0.0 1.2–2.7 −0.5–0.0 0.0–0.1 0.0 −0.4–0.0 −0.1–0.3
O he se ice sec o s 1.5–2.9 −0.6–0.2 0.0 0.0 1.0–2.0 −1.0–0.0 0.0–0.1 0.0 0.0 0.0–0.1
No e: The anspo sec o co e s he h ee modes (ai , land, and sea) and hus is epo ed as a ange.
* Sec o s acing a binding local con en equi emen .
Sou ce: Model esul s.
34 Do o hee Flaig and Susan F. S one
Table 2.4 A gen ina – Mo o Vehicle Sec o LCR, De ailed E ec s on Au omo i e
P oduc ion, Pe cen age Changes
(i) Mo o ehicle sec o
Quan i y
To al Use ca ego y
In e media e
inpu s
P i a e
consump ion
Go e nmen
consump ion
P ices
Use ca ego y
Capi al goods
In e media e
inpu s
P i a e
consump ion
Go e nmen
consump ion
Capi al goods
P oduc ion 39% 101% 17% 66% 28% −5% −5% −5% −4%
Expo s 51% 87% 31% 66% 38% −7% −3% −5% −4%
Domes ic 23% 175% 12% 43% 13% 0% −6% −8% −7%
Impo s −4% 3% −12% −2% −16% −1% −2% −1% −2%
To al demand 7% 15% 5% 0% 0% −5% −2% −5%
(ii) T anspo equipmen sec o
Quan i y P ices
To al Use ca ego y Use ca ego y
In e media e
inpu s
P i a e
consump ion
Go e nmen
consump ion
Capi al goods
In e media e
inpu s
P i a e
consump ion
Go e nmen
consump ion
Capi al goods
P oduc ion 17% 11% 19% 84% 28% −4% −4% −4% −4%
Expo s 41% 32% 37% 84% 44% −3% −3% −4% −3%
Domes ic 14% 11% 17% 54% 23% −4% −5% −6% −5%
Impo s −10% −12% −10% −3% −9% −2% −2% −1% −1%
To al demand 3% 6% 3% 0% −3% −1% −3%
LCR = local con en equi emen .
Sou ce: Model esul s.
B azilian elecommunica ions indus y
The measu e is modeled as a equi emen o sou ce 70% o he inpu s o elec-
onic equipmen and machine y and equipmen o he communica ions sec o
om domes ic sou ces. P io o he policy implemen a ion, hese sec o s had
local con en o a ound 30%. The elecommunica ions sec o is pa o he pos
and elecommunica ions sec o in he METRO model. Telecommunica ions
Localiza ion measu es 35
accoun o abou 80% in he b oade sec o (IBGE, 2022). To he ex en ha
pos al se ices use elec onic and machine y equipmen as an inpu , he impac
o he LCR will be o e s a ed. The model assumes ha employed echnologies
do no change. Finally, o he ex en ha i ms decide no o comply and lea e
he ma ke , he impac o he LCR is likely unde s a ed.
Resul s
Telecommunica ions accoun o 9% o GDP, and 8% o he sec o ’s inpu s
a e impo ed (Table 2.2). The LCR inc eases in domes ic p oduc ion in he
a ge ed sec o s a e epo ed in Table 2.3 and Table 2.5. The model allows
o he di e en ia ion o domes ic and expo p ices. This pe mi s us o cap-
u e he abili y o i ms o engage in p ice disc imina ion, i.e., keeping hei
expo p ices low o p o ec o e en inc ease ma ke sha e o e seas while ais-
ing domes ic p ices whe e, due o he LCR, compe i ion is es ic ed. The
elec onic equipmen sec o bene i s om inc easing domes ic demand due o
he LCR exclusi ely om inpu s in o communica ions expo s and p oduc ion
inc easing 0.6% and 0.9%, espec i ely.
This LCR builds on exis ing es ic i e policies in he B azilian elecom-
munica ions sec o (S one, Messen , and Flaig, 2015). Thus, he cos s o his
policy a e limi ed and ha e no no iceable addi ional e ec s on o he sec o s o
on he agg ega e le el o ou pu o labo income (Table 2.3). Howe e , wha
he model does no cap u e o a su icien deg ee a e he cos s imposed on he
go e nmen o moni o and en o ce he policy. I hese could be accu a ely
quan i ied, i is likely ha o e all wel a e impac s would be nega i e.
Table 2.5 B azil – LCR, Pe cen age Changes
(i) Elec ical equipmen sec o
In e media e
inpu s
P i a e
consump ion
Go e nmen
consump ion
Capi al goods
In e media e
inpu s
P i a e
consump ion
Go e nmen
consump ion
Capi al goods
Quan i y P ices
To al Use ca ego y Use ca ego y
P oduc ion 0.6% 3.1% 0.0% −0.1% 0.0% 0.0% 0.0% 0.0% 0.0%
Expo s 0.9% 2.0% 0.0% −0.1% 0.0% −0.1% 0.0% 0.0% 0.0%
Domes ic 0.6% 3.6% 0.0% −0.1% 0.0% 0.1% 0.0% 0.0% 0.0%
Impo s −0.7% −1.3% 0.0% 0.0% 0.0% −0.1% 0.0% 0.0% 0.0%
To al demand 0.0% 0.1% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
(Con inued )
36 Do o hee Flaig and Susan F. S one
Table 2.5 (Con inued)
(ii) Machine y and equipmen sec o
Quan i y
To al Use ca ego y
In e media e
inpu s
P i a e
consump ion
P ices
Use ca ego y
Go e nmen
consump ion
Capi al goods
In e media e
inpu s
P i a e
consump ion
Go e nmen
consump ion
Capi al goods
P oduc ion 0.1% 0.5% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
Expo s 0.1% 0.3% 0.0% −0.1% 0.0% 0.0% 0.0% 0.0% 0.0%
Domes ic 0.1% 0.6% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
Impo s −0.1% −0.2% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
To al demand 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
LCR = local con en equi emen .
Sou ce: Model esul s.
B azil: p e e en ial ma gins on a ious p oduc s in public
p ocu emen p ocesses
In 2013 and 2014, he go e nmen o B azil decided o inc ease he p e e -
ence ma gins o he public p ocu emen o a a ie y o na ionally p oduced
goods. P e e ence ma gins a e he maximum ex en o which he p ice quo ed
by a local supplie may be abo e ha o a compe i o . The new ma gins ange
be ween 9% and 25% and include IT (15%), ac o s wi h con inuous acks
(20%), ai planes (9%–25%), a ious IT- ela ed goods and se ices (15%–25%),
capi al goods (15%–20%), and oys (10%).
The p ice p e e ence is implemen ed as a ax b eak on domes ic p oduc s used
in go e nmen p ocu emen , in oducing a p ice di e ence be ween domes ic and
impo ed commodi ies o 10% o 20%, depending on he sec o . To de e mine he
le el o he p ice di e ence o implemen he scena io, sales axes and a i s a e
adjus ed simul aneously, keeping impo p ices cons an . The LCR unde e iew
applies o go e nmen p ocu emen , so his modeling app oach, in luencing go -
e nmen income and expendi u e, is a easonable choice. The esul s depend on
he quali y o he da a ep esen ing go e nmen consump ion in he da abase,
whe eby s a is ics on go e nmen consump ion a e gene ally a he c i ical. To he
ex en ha he go e nmen use is unde s a ed in he da abase, he esul s will be
la ge . In addi ion, gi en ha he ax b eak is implemen ed on a e age ac oss he
a ious sec o s, he esul s o any speci ic supply sec o will a y.
Resul s
Local con en in B azil’s go e nmen consump ion in he da abase is 5% o com-
pu e s and elec onic p oduc s, elec ical equipmen , and anspo equipmen
Localiza ion measu es 37
and 26% o machine y. The policy doubles local con en in hese sec o s. Local
con en is al eady high o mo o ehicles (96%) and communica ion se ices
(93%) and inc eases only ma ginally, o 97% and 95%, espec i ely. Go e nmen
impo s in he a ge ed sec o s dec ease be ween 5% and 25%.
Howe e , go e nmen consump ion in he a ge ed sec o s accoun s o a
maximum 0.01% o consump ion ac oss all use ca ego ies, so he go e nmen
sec o is oo small in his a ea o ha e a isible e ec on he agg ega e o sec o
le el. Thus, hese esul s p o ide a good example whe e, while iden i iable and
anspa en , LCRs may no ha e a no able di e ence on economic ou comes.
Indonesia: LCR in he au omo i e indus y
The LCR is modeled as a 100% LCR on mo o ehicles, ep esen ed by
wo sec o s – “mo o ehicles and pa s” and “ anspo equipmen ” – o
inal demand. Impo ed mo o ehicles being used as in e media es in o p o-
duc ion in he au omo i e indus y a e no co e ed by he LCR, assuming
hese a e pa s going in o assembly in Indonesia. Thus, he measu e is well
depic ed in he da abase and ela i ely s aigh o wa d o model. Howe e ,
he model implies ha all cu en ope a o s comply wi h he measu e. To
he ex en ha ope a o s lea e he sec o as a esul o he policy, he esul s
a e unde s a ed.
Resul s
The au omo i e indus y in Indonesia accoun s o 4% o GDP, and 32% o
au omobiles o in e media e and inal demand a e impo ed (Table 2.2).
The LCR inc eases he local con en o mo o ehicles om 63% o 89% and
local con en o anspo equipmen om 75% o 87%. As a esul , impo s
o mo o ehicles and anspo equipmen d op s ongly ac oss uses, by
68% and 44%, espec i ely (Table 2.6). Impo s o in e media es o hese
commodi ies dec ease in mos sec o s, excep in he au omo i e indus y,
which inc eases impo s o pa s o sa is y domes ic demand. As a esul
o dec easing au omo i e impo s, he exchange a e app ecia es o balance
he cu en accoun , and impo s in o he manu ac u ing sec o s inc ease
be ween 1% and 4% (Table2.3). On he coun y le el, impo s and expo s
dec ease by 2.3% and 2.5%, espec i ely, and he e ms o ade imp o e by
0.7% (Table 2.2).
The policy inc eases demand o domes ic mo o ehicles s ongly, and
domes ic p oduc ion o mo o ehicle and anspo equipmen inc eases 42%
and 23%, espec i ely (Table 2.6). Inc easing p oduc ion leads o inc easing
demand o labo in he au omo i e indus y, and wo ke s ealloca e o he
au omo i e indus ies and wages inc ease 0.2%. Re u ns o capi al inc ease
0.1%, so ha sec o s ha a e labo and capi al in ensi e expe ience inc easing
p oduc ion cos s.
A he same ime, impo p ices dec ease. Fo inpu s ac oss o he pa s
o he economy, impo p ices d op by 0.4% o 0.6% while, hey decline 9%

38 Do o hee Flaig and Susan F. S one
Table 2.6 Indonesia – LCR in he Au omo i e Indus y, Pe cen age Changes
(i) Mo o ehicles sec o
Quan i y P ices
To al Use ca ego y Use ca ego y
media e
nmen
media e
nmen
In e
inpu s
P i a e
consump ion
Go e
consump ion
Capi al goods
In e
inpu s
P i a e
consump ion
Go e
consump ion
Capi al goods
P oduc ion 42% 110% 4% −4% 43% −1% −1% −1% −1%
Expo s 32% 57% 1% −5% 24% −6% −1% −1% −3%
Domes ic 45% 153% 5% 301% 47% 3% 0% 29% 0%
Impo s −68% −62% −56% −16% −91% −9% −8% −3% −20%
To al demand 3% 7% 1% 0% 0% −1% 4% 0%
(ii) T anspo equipmen sec o
Quan i y P ices
To al Use ca ego y Use ca ego y
media e
nmen
media e
nmen
In e
inpu s
P i a e
consump ion
Go e
consump ion
Capi al goods
In e
inpu s
P i a e
consump ion
Go e
consump ion
Capi al goods
P oduc ion 23% 18% 4% −1% 238% 0% 0% 0% 0%
Expo s 28% 5% −3% −6% 109% −1% −1% −1% −5%
Domes ic 22% 19% 5% 301% 311% 0% 0% 18% 2%
Impo s −44% −3% −58% −16% −87% −1% −6% −2% −12%
To al demand 6% 13% 0% 0% 0% 0% 2% 1%
LCR = local con en equi emen .
Sou ce: Model esul s.
o mo o ehicle pa s ( anspo equipmen – 1%) due o a policy-induced
dec ease in impo demand. This p ice dec ease bene i s sec o s wi h la ge
impo sha es o in e media es such as pe oleum, chemicals, compu e s and
elec ical equipmen , elec ici y, and wa e and ai anspo a ion. Mo o
ehicle pa s a e a majo inpu , so he e ec is especially s ong o mo o
ehicles.
Localiza ion measu es 39
Dec easing expo compe i i eness and inc easing impo compe i ion,
esul ing om he cu ency app ecia ion, and inc easing p oduc ion cos s in
mos o he sec o s, ha e a nega i e impac on p oduc ion in non- a ge ed
sec o s (Table 2.3). Thus, on an agg ega e le el, p oduc ion and GDP do no
change.
Saudi A abia: local con en and go e nmen p ocu emen measu e on
p oduc s in medicines and medical supplies
The measu e is modeled as a 90% minimum equi emen o domes ic con en
o go e nmen p ocu emen in he ollowing sec o s: ex iles, wood p oduc s,
pape p oduc s, chemicals, basic pha maceu icals, ubbe and plas ic p oduc s,
and elec onic equipmen . Allowing o p oduc s ha a e no subjec o he
measu e, 10% o p oduc s in each sec o a e assumed exemp . As no ed, accu-
a e go e nmen p ocu emen da a a e di icul o ob ain. Thus, he model
ou comes depend on he accu acy o Saudi A abia’s go e nmen p ocu emen
in he METRO da abase. To he ex en ha i is unde s a ed, he impac on
go e nmen se ices, and indeed he o al impac , will be unde s a ed.
Resul s
Go e nmen p ocu emen o he a ge ed commodi ies, o which 20% a e
impo ed, accoun s o 0.7% o Saudi A abia’s GDP (Table 2.2). On he com-
modi y le el, he LCR is binding o chemicals and pha maceu icals, whe e
local con en inc eases om 66% and 40%, espec i ely, o 90%. O e all,
go e nmen impo s dec ease 3.1%, while impo s o all o he uses inc ease
sligh ly, leading o a o al impo dec ease o 0.1% (Table 2.2). The e is no
no iceable e ec on he exchange a e.
Table 2.7 shows he e ec s exempla ily o pha maceu icals. Impo s
a e subs i u ed by domes ic p oduc s in go e nmen p ocu emen and
igge domes ic p oduc ion. O he uses a e only sligh ly a ec ed and
show con as ing endencies, inc easing impo s, and lowe ing demand
o domes ic goods as he go e nmen inc eases p ices o domes ic p od-
uc s. Go e nmen consump ion is la ge enough o inc ease p oduc ion
in he sec o s whe e he LCR is binding – 0.4% o chemicals and 3.8%
o pha maceu icals. Non- a ge ed sec o s expe ience a small bu nega i e
e ec (Table 2.3).
Sou h A ica: new mining bill
The measu e is modeled as an LCR on inpu s o mining, abs ac ing om
complex company owne ship equi emen s. The model cap u es hese meas-
u es by depic ing he policies wi h espec o he domes ic con en equi e-
men o 70% on goods and 80% on se ices. Once again, we assume ha all
cu en ope a o s comply wi h he measu e. To he ex en ha companies
40 Do o hee Flaig and Susan F. S one
Table 2.7 Saudi A abia LCR – Pha maceu ical Sec o , Pe cen age Changes
lea e he ma ke as a esul o he policy, he impac s a e unde s a ed. In addi-
ion, he ex en o which he owne ship s uc u e impac s he deg ee o imple-
men a ion will impac he esul s.
Resul s
Mining con ibu es 4% o Sou h A ican GDP, and he sec o impo s 17% o
i s inpu s (Table 2.2). The LCR is binding o mos o he p oduc ion inpu s
and inc eases domes ic supply and p oduc ion in sec o s ac oss he boa d, wi h
he no able excep ions o me als and mine als (Table 2.3). The supply o min-
ing p oduc s o he domes ic ma ke inc eases 2% (Table 2.8).
While he e is no no iceable e ec on inpu p ices o p oduc ion cos s in
mining, i s ou pu declines by 0.8%. This s ems om a decline in expo ma -
ke s. Mining is an impo an expo sec o in Sou h A ica – 97% o i s o al
ou pu is expo ed, and i accoun s o 12% o he coun y’s o al expo s.
Expo s dec ease in esponse o ela i e exchange a e e ec s. As a conse-
quence, mining expo s dec ease 0.9%, d i ing he decline in p oduc ion.
US: Buy Ame ica bill enac ed by a ious s a es
The US Buy Ame ica p og am is implemen ed h ough anspo a ion g an s
ac oss many US s a es. Thus, any public p ojec unded by hese g an s mus
use some le el o domes ically p oduced i on, s eel, and o he manu ac u ed
goods. The amoun s o bo h he g an s and he LCR a y by s a e and by
p ojec s wi hin s a es.
Quan i y
To al Use ca ego y
In e media e
inpu s
P i a e
consump ion
P ices
Use ca ego y
Go e nmen
consump ion
Capi al goods
In e media e
inpu s
P i a e
consump ion
Go e nmen
consump ion
Capi al goods
P oduc ion 3.8% −0.2% 0.0% 125.0% −0.1% 0.0% 0.0% 0.0% 0.0%
Expo s 0.5% −0.2% −0.1% 71.0% −0.2% 0.0% 0.0% −4.1% 0.0%
Domes ic 4.2% −0.1% 0.0% 127.1% −0.1% 0.0% 0.0% 0.2% 0.0%
Impo s −2.3% 0.1% 0.1% −83.4% 0.0% 0.0% 0.0% −12.5% 0.0%
To al demand 0.0% 0.1% 0.0% 0.0% 0.0% −1.0% 0.0%
LCR = local con en equi emen .
Sou ce: Model esul s.
Localiza ion measu es 41
Table 2.8 Sou h A ica LCR – Mining Sec o , Pe cen age Changes
Quan i y
To al Use Ca ego y
In e media e
inpu s
P i a e
consump ion
Go e nmen
consump ion
P ices
Use Ca ego y
Capi al goods
In e media e
inpu s
P i a e
consump ion
Go e nmen
consump ion
Capi al goods
P oduc ion −0.8% −0.8% 0.1% 0.1% 0.0% 0.0% 0.0% 0.0% 0.0%
Expo s −0.9% −0.9% 0.1% 0.1% 0.0% −0.1% 0.0% 0.0% 0.0%
Domes ic 2.4% 2.4% 0.0% 0.1% 0.0% 1.8% 0.0% 0.0% 0.0%
Impo s −0.8% −0.8% 0.0% 0.0% 0.0% −0.2% 0.0% 0.0% 0.0%
To al demand −0.1% −0.1% 0.0% 0.0% 0.0% 0.0% 0.0%
LCR = local con en equi emen .
Sou ce: Model esul s.
The model does no p o ide o measu emen o indi idual g an s a he
subna ional le el. Howe e , looking ac oss hund eds o g an s, accoun ing
o he size o he g an and ela i e size o he anspo budge s, his meas-
u e is modeled as an a e age inc ease in domes ic con en o 25 pe cen age
poin s in o al go e nmen spending on anspo equipmen and se ices.
Thus, domes ic con en is aised om 61% in mo o ehicles, 5% in anspo
equipmen , and 55% in anspo se ices, o 86%, 30%, and 80%, espec i ely.
This is an a e age ac oss hund eds o indi idual policies and g an s and hus
ep esen s an a e age impac .
Resul s
Impo s accoun o 49% o US go e nmen p ocu emen spending in ans-
po equipmen and se ices (Table 2.2). The measu e inc eases domes ic
demand in he a ge ed sec o s and educes impo s. Howe e , go e nmen
p ocu emen in his a ea accoun s o only 0.002% o GDP, so he e ec is
no la ge enough o isibly impac on he sec o le el (Table 2.9 p esen s he
example o mo o ehicles). Again, while he measu e is anspa en and iden-
i iable, i illus a es he case whe e la ge domes ic ma ke s can o en a o d o
implemen his policy wi h small measu able side e ec s. Howe e , wha he
model ails o cap u e in his ins ance is he longe - e m impac s his policy
has on inno a i e beha io by i ms. By ha ing access o go e nmen con ac s
on a noncompe i i e basis, i ms ha e li le incen i e o in es in inno a i e o
cos -cu ing beha io . This educes hei compe i i e s ance is-à- is i ms ha
ace con es ed ma ke s.
DOI: 10.4324/9781003415794-3
This chap e has been made a ailable unde a CC-BY-NC-ND 4.0 license.
Objec i es o he chap e
His o ically, he mining indus y in esou ce- ich de eloping na ions has ope -
a ed as an encla e, ex ac ing aw ma e ials o expo wi h ew links wi h o he
sec o s and li le alue added o he esou ce- ich coun y. This diminished
he oppo uni y o di ec economic and social bene i s, causing many mining
coun ies o endu e undi e si ied economic s uc u es, high unemploymen
a es, and mac oeconomic amewo ks ulne able o commodi y shocks.
Legal and iscal amewo ks adop ed du ing he 1990s ocused on he capac-
i y o he mining sec o o gene a e ax and oyal y e enues whose bene i s,
i was assumed, would au oma ically ickle down o he es o he economy
(Bas ida, 2014). Howe e , his app oach ailed o app ecia e ha p ocu emen
o goods and se ices is he single la ges in-coun y economic expendi u e
o e he li e o a mining p ojec – some imes la ge han axes, sala ies, wages,
and communi y in es men combined.1
A p esen , global mining companies ollow high p ocu emen s anda ds
and end o ou sou ce hei ope a ional ac i i ies o globally compe i i e con-
ac o s. The mo e owa d g ea e ou sou cing has led o he eme gence o
global supply chains in he mining sec o , enabled by alling anspo a ion
cos s, lowe ade ba ie s, imp o ed in o ma ion and communica ion ech-
nologies, and libe alized inancial egula ions (Ös ensson, 2017).
Since he p ocu emen o goods and se ices is he single la ges expense
o e he li e o a mining p ojec , local con en equi emen s (LCRs) ha e
become poli ically a ac i e, as hey aim o espond o demands o c ea e jobs
and economic oppo uni ies. Ye in p ac ice, he design and implemen a ion o
policy amewo ks o LCRs in mining ha os e sus ainable and compe i i e
domes ic supplie s has p o en ex emely di icul .
Many coun ies ha e in oduced o ampli ied a ge s o locally supplied
goods and se ices in legisla i e and egula o y ins umen s wi hou su eying
mining companies’ p ocu emen needs, es ablishing a baseline o local sup-
ply capabili ies, o calcula ing he ade-o s om speci ic ini ia i es in e ms
o alue c ea ed o he economy. Mining companies a gue ha such a ge s
a e o en p esc ibed wi hou due conside a ion o he complex ope a ional
3 Local con en policies
in he mining sec o
Jane Ko inek and Paulo De Sa

Local con en policies in he mining sec o 49
s uc u es, s a egies, and ma ke condi ions o he indus y, and hey ha e
o en esponded wi h cau ion o sough o ci cum en compliance wi h such
a ge s. No wi hs anding, many companies ha e se up in e nal local con en
p og ams ha also se e as isk-mi iga ion ac ions and ha e de eloped con-
s uc i e engagemen wi h go e nmen s and communi ies o ein o ce hei
social license o ope a e.
This chap e d aws on some obse a ions ega ding he e ec i eness o
LCRs’ con ibu ion o os e employmen and economic di e si ica ion and
inc ease go e nmen e enues in esou ce- ich coun ies. I add esses policy
implica ions o coun ies ha aim o maximize bene i s om he mining sec-
o while ensu ing ha hei egula o y and business en i onmen s con ibu e
o sus aining he sec o ’s compe i i eness in global ma ke s.
The chap e begins wi h an in oduc o y sec ion ha p esen s cu en
de ini ions o LCRs and sugges s a ypology o local con en policies ha
a e used in he mining sec o and some conside a ions when aiming o meas-
u e hei impac s. The second sec ion e iews local sou cing and domes-
ic employmen equi emen s, ade es ic ions, and o he local con en
measu es ha ha e been in oduced in di e en coun ies o os e back-
wa d and o wa d linkages and analyses common pi alls in hei design and
implemen a ion.
The hi d sec ion includes e lec ions on ins i u ional amewo ks ele an
o coun ies ha aim o deploy LCRs. I includes a b ie discussion on he
implica ions o Wo ld T ade O ganiza ion (WTO) ules and in es men ag ee-
men s on he use o such ins umen s.
The ou h sec ion elies on a quali a i e desk e iew o exis ing li e a-
u e, esea ch pape s, and epo s on LCR law and p ac ice ha aims o
shed ligh on some o he po en ial economic impac s o he use o LCRs
in esou ce- ich coun ies. I highligh s di icul ies in measu ing he impac s
o LCRs, due in pa o a lack o da a p o iding empi ical e idence bu
also because he de ini ion o LCRs a ies om coun y o coun y, mak-
ing compa isons di icul . No wi hs anding, empi ical e idence sugges s
ha LCRs seldom achie e hei objec i es o inc easing he domes ic alue
added in he supply o goods and se ices o de eloping sus ainable linkages
bu ha e been somewha success ul in some cases in e ms o job c ea ion
and skills ans e .
The las sec ion p esen s a se ies o policy implica ions de i ed om he
p eceding analysis. I inds ha LCRs, i implemen ed, mus be aligned wi h
wha can be ealis ically achie ed wi hou h ea ening he long- e m com-
pe i i eness o he indus y. The e o e, hey should be pa o a b oade se o
public policies o le e age he sec o ’s con ibu ion o inc ease alue addi ion
h oughou he economy, job c ea ion, and economic di e si ica ion a he
han he sha e o domes ic p ocu emen . Acco dingly, many esou ce- ich
coun ies ha e mo ed om p esc ibed local p ocu emen equi emen s o
c ea ing he condi ions o inc ease he expo o mining- ela ed se ices and
in eg a ion in global supply chains.
50 Jane Ko inek and Paulo De Sa
Scope and con en o LCRs
LCRs in he mining sec o include all laws and egula ions ha p esc ibe
measu es o s imula e he use o locally sou ced goods and se ices, c ea e
job oppo uni ies, and gene a e b oade spillo e e ec s in he na ional and
local economies o esou ce- ich coun ies. Thei scope anges om manda-
o y employmen a ge s o ax exemp ions on local p ocu emen bu also
includes expo es ic ions o encou age downs eam p ocessing o domes-
ic mine als, owne ship equi emen s, he ese a ion o ce ain p ocu emen
om domes ic i ms, and demands ha esea ch and de elopmen (R&D) on
mining- ela ed echnologies be conduc ed in he coun y whe e ope a ions
ake place. An es ima ed 90% o esou ce- ich coun ies employ LCR, mos o
which a e quan i a i e a ge s o equi emen s (McKinsey, 2013).
The e is no commonly ag eed de ini ion o wha cons i u es “local con-
en .” The e m “local” has been used in di e en ways (Ko inek and Ramdoo,
2017; In e go e nmen al Fo um on Mining, Mine als, Me als and Sus ainable
De elopmen [IGF], 2018c): i can e e o geog aphic p oximi y, such as he
popula ion li ing in he icini y o a mining p ojec , al hough local supplie s
a e o en de ined as businesses egis e ed in he coun y a bo h he na ional
and communi y le els. Simila ly, local employmen is gene ally associa ed wi h
he na ionali y o he wo k o ce and can a ge a wide ange o job posi ions,
including a he managemen le el. In some coun ies, goods and se ices
mus be p o ided by i ms wi h some pe cen age o domes ic capi al owne -
ship. In o he s, mining companies may be eques ed o en e in o pa ne ships
wi h s a e-owned en i ies o local i ms o o lis a p esc ibed pe cen age o
hei sha es on na ional s ock exchanges.
The de ini ion and scope o wha cons i u es “con en ” can a y om coun-
y o coun y bu usually aims o accomplish one o mo e o he ollowing: (i)
ups eam supplie de elopmen a he domes ic and/o communi y le el; (ii)
skills enhancemen a di e en s ages o he alue chain, including he c ea ion
o job oppo uni ies and aining o he local wo k o ce; (iii) echnological
and knowledge ans e , in pa icula owa d local small and medium-sized
en e p ises (SMEs) o go e nmen agencies, o in es ing in R&D ac i i ies
in he coun y; (i ) sha ed owne ship o asse s; and/o ( ) downs eam alue
addi ion and bene icia ion o locally p oduced aw ma e ials (Ko inek and
Ramdoo, 2017).
Typology o main LCRs a ec ing he mining sec o
LCRs a e gene ally o mula ed in policy amewo ks se ing b oad o ien a ions–
such as na ional de elopmen plans and policy s a emen s – and a e codi ied
in legisla i e and egula o y ins umen s o as pa o con ac ag eemen s
nego ia ed wi h mining in es o s. The e is a wide ange o policies o p omo e
local con en , bu in simple e ms, hey can be classi ied as ei he demand-
side o supply-side policies. Demand-side policies ocus on c ea ing demand
Local con en policies in he mining sec o 51
o locally p ocu ed goods and se ices. Supply-side policies emphasize skills
de elopmen and building he capaci y o local supplie s o b ing hem up o
global s anda ds o compe i i eness on p ice, quali y, and eliabili y. Measu es
ange om manda o y o olun a y supplie de elopmen p og ams o s and-
alone o public–p i a e pa ne ships o skills de elopmen .
Measu es can be u he classi ied as manda o y (“ equi emen -based”
app oach) o ely on i ms’ olun a y p og ams o “bes e o s” o g an p e -
e ences o local economic en i ies (“incen i es-d i en” app oach). “Requi e-
men -based” policies can be u he classi ied in o wo ca ego ies. Some impose
legally binding a ge s on i ms, ei he in e ms o quan i y (e.g., he numbe
o local s a o be employed o con ac s o be awa ded o local supplie s) o
alue (e.g., a pe cen age o o al spending on local p ocu emen ). “Incen i es-
d i en” equi emen s gene ally do no se speci ic a ge s bu can also be bind-
ing. Fo example, companies may be eques ed o publish hei p ocu emen
needs o epo on he pe cen age o local employmen , al hough he le els
o local p ocu emen and employmen migh no be p esc ibed. “Bes e o s”
p o isions a e usually embedded in legisla ion bu do no subjec he compa-
nies o any speci ic quan i a i e equi emen s.
LCRs can be u he classi ied as aiming o os e g ea e linkages ups eam
o he mining sec o , o downs eam (Table 3.1). They con ain measu es
aimed a de eloping job skills and he capaci y o local supplie s, using he sec-
o as he ancho clien in he case o backwa d o ups eam linkages o as he
sou ce o inpu s in he case o o wa d o downs eam linkages.
In he case o LCRs ha aim o os e backwa d linkages o inc ease local
p ocu emen , local sou cing and domes ic employmen equi emen s a e he
mos common examples o manda o y LCRs. The i s o en manda es he
pu chase o speci ic p oduc ca ego ies o a p esc ibed olume o alue o
goods and se ices om local supplie s. They can also include ende p e e -
ences o local supplie s and a e inc easingly accompanied by he obliga ion
o p o ide p ocu emen plans, schedules, and implemen a ion epo s o
local au ho i ies (Wo ld Bank and Kaise Economic De elopmen Pa ne s,
2015). Domes ic employmen equi emen s call o he hi ing o speci ic
pe cen ages o local wo ke s and o companies o ese e some ca ego ies
o jobs exclusi ely o na ionals (and, inc easingly, o indigenous people,
women, o disad an aged g oups). They can also limi he numbe o expa-
ia es employed and manda e aining p og ams o hei eplacemen by
local wo ke s.
Go e nmen s can ac as acili a o s, p o iding incen i es o i ms o inc ease
hei local pu chases. Supplie de elopmen p og ams (SDPs) a e by a he
mos common go e nmen “incen i es-d i en” app oach o inc ease ups eam
o backwa d linkages; o he app oaches include (i) ax educ ions o i ms
ha a e pa o mining supply chains, suppo wo k o ce de elopmen , in es
in inno a ion, o ag ee o ans e echnology; (ii) g an s and schola ships
o s uden s and employees willing o upg ade hei skills o o aining
ins i u ions ha pa ne wi h indus y o de elop hem; and (iii) suppo o
52 Jane Ko inek and Paulo De Sa
Table 3.1 Illus a i e Examples o LCRs
Type Demand Side Supply Side
Policies ha aim o inc ease ups eam linkages
Requi emen Ex ac i e i ms a e Ex ac i e i ms a e equi ed o:
based equi ed o: • P o ide aining o employees
• Mee nume ical • Fund capaci y de elopmen p og ams
a ge s o local • Es ablish a “buddy sys em” whe eby
employmen pe ype local s a a e pai ed wi h expa ia es
o jobs and/o le el o di ec on- he-job aining
o compe ency o
epo on measu es
aken o hi e locally
• Ex end p e e ences
o local supplie s
o p ocu emen o
goods and se ices
Incen i es Ex ac i e i ms Ex ac i e i ms a e equi ed o
d i en a e equi ed o encou aged o:
encou aged o: • Conduc aining p og ams o
• Publish job acancies po en ial supplie s o unde s and
• Publish ende s on he needs and equi ed s anda ds o
gi en websi es and in ex ac i e i ms
he media • Conduc awa eness campaigns
• Se up o use exis ing abou key p ocu emen employmen
ne wo ks o supplie s oppo uni ies
Go e nmen s and academic ins i u ions:
• C ea e enginee ing and echnical
cu icula in conjunc ion wi h
ex ac i e i ms’ s a ed and u u e
needs
• P o ide a ge ed aining o enhance
he skills o po en ial supplie s
Go e nmen s o egula o s:
• Se up ne wo ks o supplie s and
ex ac i e i ms
• P o ide o ums o ma chmaking
be ween local supplie s and ex ac i e
i ms o os e g ea e engagemen
Policies ha aim o inc ease downs eam linkages
Requi emen Ex ac i e i ms a e Ex ac i e i ms a e equi ed o:
based equi ed o: • In es in downs eam p ocessing
• Sell a sha e o acili ies
hei aw ma e ials Concomi an ly, go e nmen s o
in-coun y egula o s may:
• Pay highe ax • P o ide ax concessions, concessional
a es on expo s o loans, and lowe impo du ies
aw ma e ials han on impo ed capi al goods o
p ocessed p oduc s subsidize ene gy, anspo , o o he
• Engage in in as uc u e
downs eam
p ocessing in o de o
ob ain expo licenses
Local con en policies in he mining sec o 53
Type Demand Side Supply Side
Incen i es
d i en
• Ex ac i e i ms a e
gi en ax concessions
i hey a o in-
coun y sale o aw
ma e ials
• Ex ac i e i ms a e
equi ed o di es
a sha e o hei
equi y i hey do no
p ocess aw ma e ials
in-coun y
Ex ac i e i ms a e equi ed o
encou aged o:
• Collabo a e wi h aining ins i u ions
o ensu e he necessa y skills o
p ocessing indus ies
Go e nmen s o egula o s:
• P o ide ax o o he concessions o
p ocesso s o in es
LCR = local con en equi emen .
Sou ce: Au ho s’ concep ion; Ko inek and Ramdoo (2017).
esea ch, de elopmen , and inno a ion h ough dedica ed unding o uni e si-
ies, esea ch cen e s, and inno a ion incuba o s.
No wi hs anding, se e al coun ies ha e in oduced p esc ip i e bene icia-
ion equi emen s ei he in he o m o ac oss- he-boa d legisla ion o du ing
nego ia ions o con ac s wi h mining i ms. These equi emen s a e essen ially
o h ee ypes: (i) axes on he expo o unp ocessed mine als; (ii) quan i-
a i e expo es ic ions on unp ocessed mine als, ei he pa ial (quo as) o
comp ehensi e bans; and (iii) expo licensing equi emen s o con ol min-
e al expo s (UNCTAD, 2017; Fung and Ko inek, 2013). In o he coun ies,
go e nmen s ha e c ea ed o manda ed exis ing s a e-owned en e p ises o
in es in he downs eam sec o . China is he mos ob ious example, wi h s a e
en i ies owning and ope a ing mos coppe p ocessing acili ies and suppo -
ing hem wi h di ec g an s, low-in e es loans, and ax incen i es (Geipel, de
Wee d , and Ala con, 2021).
P esc ip i e bene icia ion equi emen s a e o en accompanied by incen i es
o inc ease he a e o e u n o downs eam in es men s such as ax educ-
ions o exemp ions, ene gy o wa e subsidies, and concessional loans o he
p o ision o indus y-speci ic in as uc u e such as indus ial pa ks. Go e n-
men s may also o e p o ec ions o p ocessing ope a ions h ough cus oms
a i s and impo es ic ions on he p oduc s being p ocessed (IGF, 2018c).
Measu ing he impac s o LCRs
The economy-wide cos s gene a ed by local con en measu es should be
measu ed and compa ed wi h he po en ial bene i s hey aim o p o ide.
LCRs on in e media e inpu s o he ex ac i e indus ies may lead o an
inc ease in p oduc ion cos s ha will aise ou pu p ices, in he mining sec-
o in he i s ins ance.2 The inc eased p ices o aw ma e ials aise cos s

54 Jane Ko inek and Paulo De Sa
o p oduce s u he down he alue chain, educing he compe i i eness
o downs eam indus ies and ul ima ely hinde ing he de elopmen o he
wide economy (G ossman, 1981) and economic di e si ica ion. The size o
hese e iciency losses will be p opo ional o he addi ional cos s associa ed
wi h pu chasing equi ed inpu s domes ically, due o he policy, compa ed
wi h hei cos on in e na ional ma ke s (S one, Messen and Flaig, 2015).
In addi ion, such measu es can educe he amoun o axes collec ed by
he go e nmen i hey nega i ely a ec he p o i abili y o i ms and hence
sh ink he ax base.
Using manda o y quan i a i e equi emen s o de elop in e na ionally
compe i i e indus ies uns coun e o he known posi i e spillo e s om
engaging in global alue chains and he ole ade plays in hei de elopmen .
Kimu a and Obashi (2011), o example, a gue ha he success o global alue
chains in Eas Asia, especially compa ed wi h La in Ame ica, elies hea ily on
such in e linkages be ween domes ic and o eign ma ke s.
Impac s on he in es men clima e o manda o y LCRs should also be
measu ed. One o he cen al ene s o an a ac i e in es men clima e is
nondisc imina ion (OECD, 2015). Res ic ions on o eign di ec in es men
(FDI) and ade ha e been ound o esul in less FDI o e all.
LCRs: some examples
P esc ibed demand-side equi emen s ha aim o c ea e backwa d linkages
ha e been used in many A ican coun ies ecen ly; he cases o Ghana, Sou h
A ica, Tanzania, and Zambia a e ou lined he e. Coun ies such as Aus alia,
Canada, and Chile ha e p e e ed an incen i es-based app oach o de elop
compe i i e supplie s and s ong linkages be ween mining and he domes ic
economy. P esc ip i e equi emen s ha aim o os e downs eam mine als
p ocessing ha e been used in many coun ies; he cases o Indonesia and Bo -
swana a e discussed he e a some leng h.
Demand-side equi emen -based LCRs ha aim o c ea e
backwa dlinkages
Fi s -gene a ion local con en legisla ion usually equi ed “bes e o s” om
mining companies. Mo e ecen ly, especially a e 2018, a g owing numbe
o A ican coun ies ha e adop ed o e ised manda o y LCRs in hei mining
laws – in oducing manda ed quan i a i e p ocu emen a ge s, some imes
equi ing o eign in es o s o open equi y o local pa ne s. This is he case
o he Democ a ic Republic o Congo, Ghana, Namibia, Sou h A ica, and
Tanzania. Coun ies like Bu kina Faso, Mali, Kenya, Mozambique, Nige ia,
and Zambia a e e ising hei mining and in es men codes wi h he same
pu pose.
The e o s o p omo e demand-side, equi emen -based LCRs in A ica
h ough mining legisla ion p o ide aluable lessons. Fi s ly, hey ha e c ea ed
Local con en policies in he mining sec o 55
a complex landscape ac oss he con inen ha is di icul o na iga e. Mo eo-
e , many coun ies ha e p esc ibed le els o local p ocu emen beyond wha
local supplie s a e capable o mee ing, and ew e o s can be conside ed com-
ple ely success ul. Many o he quan i a i e a ge s a e aspi a ional and ha e
li le chance o being success ully implemen ed wi hou mechanisms o build
he capaci y o cu en and po en ial local supplie s (A ican Na u al Resou ces
Cen e, 2021).
The ollowing cases illus a e he challenges o designing and implemen -
ing LCRs in coun ies whe e, on many occasions, laws and egula ions a e
equen ly e ised and e ac ed, c ea ing complex egula o y amewo ks and
ins i u ional ambigui y.
(i) The case o Ghana
Ghana’s 2006 Mine als and Mining Law sough o acili a e p oduc ion link-
ages, and he 2012 Mine als and Mining (Gene al) Regula ions se quo-
as and imelines o compliance applicable o bo h mining companies and
supplie s ega ding employmen o he local wo k o ce and p ocu emen o
locally p oduced goods and se ices. The egula ions had i e main ea u es
(IGF, 2018a):
• Nume ical employmen a ge s o he numbe o allowed expa ia es, wi h
es ic ions o ce ain ca ego ies o posi ions ese ed o local s a , and
ime ames o compliance. Ghana’s egula ions equi e 100% local wo ke s
o adminis a i e o labo posi ions. Fo mo e echnical, specialis , o man-
agemen oles, es ic ions on how many o eign na ionals can be employed
a e se on a case-by-case basis.
• Manda o y p ocu emen o locally p oduced goods and se ices speci ied by a
lis published by he Mine als Commission. A i s lis , published in 2014,
included eigh p oduc s equi ed o be sou ced locally. In 2016, he num-
be o p oduc s inc eased o 19, and cu en ly 28 p oduc s a e ea u ed.
• Compulso y epo ing equi emen s manda ing mining companies o submi
a i e-yea local p ocu emen plan s a ing how much hey will buy om
local i ms and epo on i each yea , including p og ess on he lis ed goods
and se ices.
• Use o a phased app oach wi h lead imes o allow indus ial capaci y o mee
he indus y equi emen s o p oduc cos , quali y, and quan i y.
• Sanc ions o noncompliance wi h he LCRs, which is assessed annually.
The la es assessmen indica es ha local companies ha e managed o supply,
on a e age, abou hal o he p oduc s speci ied in he Mine als Commission
lis , inc easing local p ocu emen om $148 million o $394 million be ween
2014 and 2018 (A a-Quayson, 2022). Howe e , hese p oduc s a e no nec-
essa ily p oduced locally since mining companies can mee LCRs o lis ed
56 Jane Ko inek and Paulo De Sa
goods by pu chasing om eselle s. Rega ding local employmen , mos com-
panies ha e been able o mee he quo as se in he legisla ion o all p o es-
sional ca ego ies lis ed (IGF, 2018d).3
(ii) The case o Sou h A ica
Sou h A ica has one o he wo ld’s mos complex local con en legisla-
ions. The 2000 P e e en ial P ocu emen Policy F amewo k enables he
designa ion o speci ic sec o s o p e e en ial domes ic manu ac u e (o
“localiza ion”), in line wi h na ional de elopmen and indus ial policy
goals. The 2011 Bene icia ion S a egy o he Mine als Indus y iden i-
ied a ange o c osscu ing cons ain s o local bene icia ion and p oposed
a se ies o policy, legal, and egula o y measu es o inc ease alue added
in he mining sec o and acili a e job c ea ion, indus ializa ion, and eco-
nomic di e si ica ion.
The B oad-Based Socio-Economic Empowe men Cha e o he Mining
Indus y (amended in 2010), commonly known as he Mining Cha e , aimed
o ec i y he esul s o disc imina ion based on ace, sex, and disabili y. I
de ined “Black Economically Empowe ed” (BEE) en i ies as hose whe e his-
o ically disad an aged pe sons hold a minimum o 25% plus one o e o capi-
al. The cha e equi es he indus y o p ocu e om BEE en i ies acco ding
o se a ge s and o ack p ocu emen , employmen , and o he c i e ia in line
wi h social ans o ma ion a ge s (de Wee d and Geipel, 2020).
The Chambe o Mines es ablished a B oad-Based Sco eca d, which
equi es i s membe s o comply wi h a p ocu emen a ge om BEE en i ies
o 70% and 80% o goods and se ices, espec i ely. Ta ge s we e hen b oken
down in o u he subca ego ies wi h ce ain pe cen ages de o ed o di e en
g oups, including his o ically disad an aged pe sons, BEE-complian compa-
nies, and women- o you h-owned companies. Mining companies we e gi en
an implemen a ion pe iod o 5 yea s o goods, wi h in e im a ge s leading o
ull implemen a ion (Geipel, de Wee d , and Ala con, 2021).
While many o he 2010 cha e ’s a ge s appea o ha e been me , hey we e
essen ially based on company owne ship a he han he le el o alue addi ion
gene a ed by a speci ic business. Mo eo e , he e was e idence ha locally p o-
duced p oduc s o he mining sec o a imes we e displaced by impo s sold
by businesses complian wi h he a ge s (Ko inek and Ramdoo, 2017).
As a esul , he 2018 Mining Cha e s ipula ed ha 70% o all goods used
in he mining indus y mus be manu ac u ed in Sou h A ica, wi hin 5 yea s
o he law’s enac men , and no me ely pu chased om Sou h A ican– eg-
is e ed supplie s. Ye mining companies ha e a gued ha he a ge was se
oo high and ha he e a e no enough supply-side policies o suppo hem.
O e all, he disconnec be ween di e en de ini ions ac oss a ious pieces o
legisla ion, amewo ks, and sco eca ds – and he lack o open dialogue and
us be ween go e nmen and indus y – inc eased he unce ain y a ound
Local con en policies in he mining sec o 57
he in e p e a ion o he a ge s and he o e all success o he policy (Mo aka
and an Rensbu g, 2015; Whi e, 2017).
I was de e mined ha some o he complex owne ship and pa icipa ion
equi emen s we e no implemen ed in good ai h, and a Code o Good P ac-
ice o he Sou h A ican Mine al Indus y was de ised o se ou adminis a-
i e p inciples o he e ec i e implemen a ion o he mining legisla ion and he
B oad-Based Socio-Economic Cha e applicable o he mining indus y. The
code de ines e hics o conduc o ensu e he Mining Cha e is implemen ed in
good ai h and o p e en abuses such as on ing p ac ices and oppo unis ic
beha io s ha may di e po en ial bene i s om he a ge ed s akeholde s.
Fo example, he Code o Good P ac ice de ines p ac ices conside ed audu-
len , such as si ua ions in which local s akeholde s may be appoin ed o a
posi ion bu discou aged o inhibi ed om pa icipa ing in co e ac i i ies; eco-
nomic di e sion, in which economic bene i s ecei ed do no low back o he
local s akeholde in he a io speci ied in he legal documen ; and in e media -
ies le e aging hei BEE s a us ha ha e concluded ag eemen s wi h mining
companies ( on ing ope a ions) (Ko inek and Ramdoo, 2017).
(iii) The cases o Tanzania and Zambia
In Tanzania, he 1997 Mine al Policy, he 2010 Mining Ac and 2017
amendmen s – W i en Laws (Miscellaneous Amendmen s) Ac – emphasized
he de elopmen o backwa d linkages bu le implemen a ion la gely o ol-
un a y compliance. In Janua y 2018, he Minis e o Mine als p omulga ed
he Mining (Local Con en ) Regula ions, in oducing ha d quo as o he p o-
cu emen o goods and se ices om domes ically owned p o ide s, including
in he banking, inancial se ices, insu ance, and legal sec o s. Mining compa-
nies a e now equi ed o p epa e an annual local con en plan, including p o-
jec ions o p ocu emen , employmen , and aining ac i i ies. The plan mus
be upda ed annually and submi ed o app o al o he Local Con en Com-
mi ee. I is also a equi emen o p epa e a e ol ing 3- o 5-yea p og am o
R&D, de ailing planned ac i i y expendi u es and calls o p oposals o hei
implemen a ion. Licensees also need o publish a plan o echnology ans e
o he bene i o Tanzanian en i ies.
The egula ions also s ipula e ha mining companies can only e ain he
se ices o Tanzanian inancial en i ies and need app o al o he Mining Com-
mission o hi e he se ices o o eign inancial ins i u ions. They mus main-
ain a bank accoun and conduc business wi h a bank wi h majo i y Tanzanian
sha eholding and may only e ain he se ices o Tanzanian legal p ac i ione s
whose p incipal o ice is in Tanzania (He be Smi h F eehills, 2018). Di -
icul ies in implemen a ion led he go e nmen o pass in Feb ua y 2019 he
Mining (Local Con en ) (Amendmen ) Regula ions o 2019, which amended
he 2018 Regula ions in a numbe o ways. Among o he hings, he amended
egula ions educed owne ship es ic ions o inancial ins i u ions p e e ence
64 Jane Ko inek and Paulo De Sa
g ow h in he supply o e ined nickel o elec ic ehicle (EV) ba e ies o e
he nex decade.15 The nickel expo ban inc eased downs eam p ocessing
because Indonesia’s excep ional esou ce endowmen p o ided s ong incen-
i es o in es men s, mos ly om Chinese companies (Lebdioui and Bilek,
2021). Howe e , i s s anding among Wes e n mining in es o s has allen
sha ply: Indonesia is anked he ou h wo s mining ju isdic ion globally
(ou o 78 ju isdic ions), ou anked only by Venezuela, he Chubu p o ince
o A gen ina, and Tanzania, acco ding o a su ey o mining in es o s (Yunis
and Aliakba i, 2020).
The Indonesian go e nmen is p epa ing a new nickel s a egy, conside ing
he possibili y o le ying an expo ax on p oduc s wi h less han 70% nickel
con en , and limi ing he cons uc ion o smel e s o class 2 (lowe -g ade)
p oduc s. This s a egy is pa o Indonesia’s policy app oach o he ene gy
ansi ion (Hube , 2021). The coun y aims o de elop a ully in eg a ed
domes ic supply chain o nickel, om o e ex ac ion o ba e y p oduc ion
and EV assembling. The go e nmen has p oduced a oad map o EV ba e y
de elopmen and s o age sys ems h ough 2026.
Fo example, in 2021, LG Ene gy Solu ion and Hyundai om he Republic
o Ko ea (hence o h, Ko ea) join ly s a ed building a $1.1 billion EV ba e y
plan in Wes Ja a. In addi ion, he Chinese company Huayou is in ol ed in
se e al smel ing p ojec s on Sulawesi Island, including wo p ojec s wi h he
Indonesian uni o Vale, es ima ed o cos a ound $6.3 billion. Fo d Mo o
Company is nego ia ing i s in ol emen in one o hese p ojec s. Huayou is
also eaming up wi h Tsingshan and Volkswagen G oup China, wi h he goal
o supplying nickel and cobal om Indonesia o suppo he p oduc ion o
ba e ies.16
Ba e y manu ac u e s a e also in es ing. In Ap il 2022, China’s la ges ba -
e y make , Con empo a y Ampe ex Technology Co., Limi ed (CATL), PT
Aneka Tambang, and PT Indus i Ba e ai Indonesia (IBC) signed an ag ee-
men o de elop a p ojec in Indonesia’s No h Maluku P o ince ha will
ocus on nickel mining and p ocessing, ba e y ma e ials, and ba e y manu-
ac u ing, as well as ba e y ecycling. CATL is in es ing in Indonesia h ough
QMB New Ene gy Ma e ials, a join en u e wi h Tsingshan and Chinese ba -
e y ecycle GEM. Ko ean ba e y make LG Ene gy Solu ion is sepa a ely
pa ne ing wi h IBC and Aneka Tambang o de elop an end- o-end ba e y
supply chain in Indonesia.
When hese new o eign in es men s ma e ialize, Indonesia will accoun
o a ound hal o he wo ld’s g ow h in nickel p oduc ion be ween 2021 and
2025, acco ding o he In e na ional Ene gy Agency, and could become a
leading p oduce o nickel-based p oduc s, including EV ba e ies. The expec-
a ion in Indonesia is ha la ge in es men s will inc ease he economies o
scale and d i e down cos s, po en ially making Indonesia a low-cos manu ac-
u e , compe i i e in global ma ke s.

Local con en policies in he mining sec o 65
(ii) The case o Bo swana
Diamonds we e i s disco e ed in Bo swana in 1966, sho ly a e independ-
ence, wi h la ge-scale p oduc ion s a ing in 1971. Diamond mining became
he g ea es con ibu o o g oss domes ic p oduc (GDP) (cu en ly a ound
30%) and go e nmen ax e enues (cu en ly a ound 60%) (Columbia Cen e
on Sus ainable In es men , 2016). In he ea ly 1980s, he go e nmen o Bo -
swana ied o p omo e he de elopmen o a diamond-cu ing and -polishing
indus y. Howe e , global mining company De Bee s, which domina ed p o-
duc ion in Bo swana and he sale and ma ke ing o diamonds globally, did no
suppo his ambi ion, a guing ha cu ing and polishing ac i i ies we e no
economically iable in Bo swana. Th ee cu ing and polishing ac o ies we e
es ablished be ween 1980 and 1990, bu none o hem e e epo ed a p o i .
A second oppo uni y eme ged in 2005, when De Bee s’ 25-yea mining
license was due o enewal. Bo swana’s nego ia ing le e age de i ed om De
Bee s’ eliance on Debswana, a 50–50 join en u e wi h he go e nmen ,
which was esponsible o abou 60% o he company’s global supply o ough
diamonds. Bo swana ob ained a gua an ee om De Bee s ha a pe cen age
o he diamonds mined in he coun y would be alloca ed o na ional cu ing
and polishing companies and ha all so ing and aluing ope a ions would be
unde aken in-coun y (UNIDO, 2012; Ko inek, 2014). Subsequen ly, he
go e nmen in i ed o eign cu ing and polishing companies o se up ope a-
ions in he coun y wi h he p omise o a gua an eed long- e m alloca ion o
De Bee s’ diamonds a 20% o 30% below he ma ke p ice, on he condi ion
ha hey hi e and ain locals wi h cu ing and polishing skills (IGF, 2018c;
2018d). Fu he mo e, a Diamond Academy was opened by he join en u e
o ain diamond so e s and alua ion s a (Ko inek, 2014).
In 2008, he Diamond T ading Company (DTC Bo swana), a 50–50 join
en u e be ween De Bee s and he go e nmen , was es ablished o so and
alue Debswana’s ou pu and manage he supply o diamonds o he domes ic
cu ing and polishing indus y. De Bee s also ag eed o mo e i s agg ega ion
business – selec ing and mixing he diamonds om De Bee s mines o i s
cus ome s – om London o Gabo one, in he hope o c ea ing conside able
spillo e s o o he indus ies such as hospi ali y, inance, and anspo a ion,
since diamond buye s would now ha e o go o Gabo one o buy De Bee s’
diamonds (Mo is, Kaplinsky, and Kaplan, 2011).
None heless, a emaining issue is he abili y o Bo swana’s diamond-cu ing
and -polishing indus y, which emains dependen on go e nmen incen-
i es, o compe e wi h low-cos ac o ies in Asia, especially in India. Unless
in es men s in suppo in as uc u e dec ease p oduc ion cos s subs an ially,
Bo swana’s diamond-polishing sec o may no su i e in he longe e m.
Mo eo e , i s diamond p ocessing indus y is buil on access o aw diamonds,
bu domes ic diamond ese es a e expec ed o be exhaus ed in 30 o 40 yea s
(Columbia Cen e on Sus ainable In es men , 2016).
66 Jane Ko inek and Paulo De Sa
Ins i u ional amewo ks and coo dina ion
High-quali y ins i u ions in cha ge o designing, go e ning, managing, en o c-
ing, moni o ing, and e alua ing LCRs a e i al o accomplish hei objec i es.
Many egula o y en i ies do no ha e a deep knowledge o he mining supply
chain and i s po en ial o gene a e e enues, businesses oppo uni ies, and
employmen . Many a e unde s a ed o unde inanced and su e om opaque
decision-making p ocesses.
The e ec i eness o LCRs can be a ec ed by gaps in he ins i u ional ame-
wo k and lack o coo dina ion o he app o al o c oss-sec o policies and
egula ions, esponsibili y o which is sca e ed ac oss many minis ies, which
o en ope a e in silos. The e o e, some coun ies ha e es ablished dedica ed
local con en en i ies o coo dina e and moni o p og ess, d awing on he
esou ces o all ele an go e nmen agencies. This is he case o he Local
Con en Commi ee in Tanzania and he Indus y Pa icipa ion Na ional
F amewo k Au ho i y in Aus alia.
Repo ing equi emen s a e impo an ools o moni o LCRs, bu hei
e icacy depends on he capaci y o he en i y in cha ge o en o cing hem.
Because many coun ies ha e ye o enac and en o ce adequa e epo -
ing mechanisms, in p ac ice, mos exis ing LCRs enable noncompliance
(Whi e, 2017), as shown in some o he p e ious examples. To add ess
his, go e nmen s a e inc easingly asking mining companies o submi local
p ocu emen plans as pa o hei yea ly epo ing equi emen s. Ye some-
imes, hese epo s co e only he i ms’ local capaci y-building ac i i ies
and a e no sui able o moni o p og ess owa d achie ing b oade desi ed
ou comes.
Insu icien collabo a ion be ween go e nmen s, he business communi y,
and ci il socie y can lead o si ua ions in which p ocu emen a ge s a e se a
un ealis ic le els, causing en o cemen di icul ies o c ea ing condi ions ha
acili a e in luence peddling o co up ion o public o icials. LCRs can p o-
ide p e e en ial ea men o s akeholde s wi h s ong es ed in e es s o ha
a e poli ically a ilia ed and may engage in en -seeking beha io ; some exam-
ples o his e ec a e ou lined abo e. Likewise, ea o compe i ion may block
coope a ion among mining companies, leading o a duplica ion o uncoo di-
na ed ini ia i es ha p e en local supplie s om achie ing he scale needed
o become compe i i e.
Pa ne ships be ween mining i ms, go e nmen s, aining ins i u ions, and
local s akeholde s a e impo an , in pa icula when implemen ing SDPs, as
hey end o be mo e e ec i e when hey bene i om he pa icipa ion o
mul iple ins i u ional ac o s, including no ably mining i ms (A ican Na u al
Resou ces Cen e, 2021). Fo example, some subna ional go e nmen s o e
echnical assis ance o SMEs in he p ocu emen con ac p ocess o keep da a-
bases o local supplie s o educe in o ma ion gaps ha diminish hei chances
o esponding o ende s, he eby combining supply-side and demand-side
policies.
Local con en policies in he mining sec o 67
WTO ules, in es men ag eemen s, and LCRs
LCRs and o he measu es o inc ease he use o domes ic goods and se ices,
such as ade es ic ions, subsidies, a i s, and ax incen i es, can in oduce
dis o ions in a o o local p oduce s and may he e o e con a ene a num-
be o ade and in es men ag eemen disciplines (see Annex Tables 3.A1
and 3.A2). The ele an WTO commi men s ha ela e o LCRs (Colum-
bia Cen e on Sus ainable In es men , 2016; Ko inek and Ramdoo, 2017;
Ko inek and Ba os, 2012) a e:
• The Na ional T ea men Obliga ion (A icle III o he GATT) clause p e-
en s go e nmen s om disc imina ing be ween like p oduc s om local
indus ies and impo s. This applies o policies such as hose ha o ce
o eign companies o buy goods o se ices p oduced by locally owned
companies o hi e local se ice supplie s.
• The Ag eemen on T ade-Rela ed In es men Measu es (TRIMs) p ohibi s
he use o mos o ms o pe o mance equi emen s on goods, se ou in
an “illus a i e lis .” These apply o domes ic sou cing equi emen s, ei he
in he o m o lis s o goods o quo as o pe cen ages, as well as equi e-
men s o sell p oduc s domes ically. Howe e , de eloping coun ies ha e
de oga ions o some o hose commi men s ou lined in A icle XVIII o he
GATT.
• A icle XI:1 o he GATT imposes a gene al ban on quan i a i e expo
es ic ions, bu A icle XI:2 and A icle XX o e b oad-scope exemp ions
o he ban on expo quo as. No ably, A icle XX(g) allows o quan i a i e
es ic ions ela ing o he conse a ion o exhaus ible na u al esou ces on
he condi ion ha “such measu es a e made e ec i e in conjunc ion wi h
es ic ions on domes ic p oduc ion o consump ion.”
• The SCM Ag eemen p e en s go e nmen s om p o iding incen i es
and g an ing subsidies ha a e con ingen on sou cing goods domes ically.
Common subsidies, such as a ge ed ax p e e ences, may also be ac ionable
unde WTO ules.
• The Gene al Ag eemen on T ade in Se ices (GATS) egula es LCRs wi h
ega d o o eign in es men and employmen . Local equi y equi emen s
and employmen quo as a e gene ally p ohibi ed, bu hese a e only egu-
la ed o he ex en ha coun ies ha e aken speci ic commi men s.
Despi e clea ules p ohibi ing ce ain o ms o LCRs, many coun ies main-
ain hem o ha e in oduced new ones in ecen yea s. Coun ies ha joined
he WTO a e 1995 ha e been subjec o close sc u iny and igh e obliga-
ions, and some o hem, such as Kazakhs an, ha e ag eed o emo e many
LCRs du ing hei accession nego ia ions (Ko inek and Ramdoo, 2017).
Measu es imposing LCRs ha e been he subjec o signi ican exemp ions
o de eloping coun ies wi hin WTO ules, including o suppo in an indus-
ies and add ess balance-o -paymen s p oblems. Mo eo e , TRIMs apply o
68 Jane Ko inek and Paulo De Sa
es ic ions on goods; ega ding se ices, hey only apply o commi men s
con ained in coun ies’ GATS schedules.
Few complain s ega ding LCRs in he ex ac i e indus ies ha e been
b ough o he WTO, al hough some ha e been iled ega ding expo es ic-
ions (see claims ega ding Indonesia and he case agains China desc ibed
ea lie ). Dispu es ega ding he en o cemen o LCRs go h ough he WTO’s
se lemen sys em, which is usually a cos ly and long p ocess (A ican Na u al
Resou ces Cen e, 2021); mo eo e , he WTO dispu e se lemen sys em has
been e ec i ely hal ed in ecen pe iods.
In e na ional and bila e al in es men ea ies o en go beyond WTO
es ic ions, e.g., by including es ic ions on pe o mance equi emen s o
echnology- ans e and R&D p og ams o be conduc ed in-coun y. They can
also con ain ai and equi able ea men obliga ions p e en ing go e nmen s
om in e e ing wi h o eign in es o s’ “expec a ions” o hei ope a ions.
Mos ea ies do no include hese p o isions bu , when included, he obliga-
ions end o be implemen ed mo e equen ly, since hese ag eemen s usually
use in es o -s a e a bi a ion a he han he WTO’s s a e- o-s a e a bi a ion
sys em, which inc eases he likelihood o complain s being iled (Columbia
Cen e on Sus ainable In es men , 2016).
Some e lec ions ega ding he economic impac s o LCRs
Many LCRs lack a b oade policy amewo k, such as o e all poli ical econ-
omy objec i es, policy s a emen s, and na ional de elopmen plans, o sup-
po hem. Many coun ies ha e adop ed s ingen a ge s in egula o y
ins umen s ha lack de ail and cla i y and ha e a na ow scope o objec-
i es. Be o e in oducing hem, go e nmen s mus asce ain whe e he min-
ing sec o i s in ela ion o na ional de elopmen objec i es, including i s
po en ial con ibu ion o os e employmen , go e nmen e enues, and
economic di e si ica ion.
Demand-side, equi emen -based LCRs o en aim o high pe cen ages o
local con en wi hou de eloping a de ailed iew o p ocu emen spending,
es ablishing a baseline o local supplie s’ capabili ies, and quan i ying he
ade-o s om speci ic ini ia i es in e ms o alue added c ea ed (Elbo ai
e al., 2019).
Inaccu a e unde s anding o mining i ms’ p ocu emen needs and o he
abso p i e capaci y o local supplie s ha e led o he p esc ip ion o un ealis ic
a ge s, se beyond le els ha local i ms can mee , especially in coun ies wi h
a weak indus ial base and a p i a e sec o ha is small, in o mal, and wi h low
p oduc i i y. This can be a de e en o he mining indus y, especially wi hou
suppo ing go e nmen -sponso ed supply-side measu es. This was one o he
issues con on ed by Ghana’s many LCRs applied o he mining sec o .
While a b oad de ini ion o local con en p o ides mo e lexibili y o i ms
o mee a ge s and objec i es, i is di icul o assess o wha ex en LCRs c ea e
local alue-added and spillo e e ec s o he es o he economy. Owne ship
Local con en policies in he mining sec o 69
equi emen s do no always yield he bes ou come in e ms o domes ic alue
added (ACET, 2017). Local owne ship is no a ele an ac o as long as com-
panies c ea e economic oppo uni ies, and employmen , and imp o e local
labo skills (Es e es e al., 2014). Fo eign-owned bu locally based businesses
can add alue o he local economy, whe eas local sou cing o impo ed goods
usually does no . Du y- ee impo s o inpu s by domes ically owned i ms
place po en ial local p oduce s and supplie s o such inpu s a a disad an age.
Fo example, in less de eloped coun ies wi h a weak indus ial base, many
mining i ms may epo a ela i ely high pe cen age o local sou cing while,
in eali y, a la ge p opo ion o pu chases o igina e om impo s by local i ms
o ep esen a i es o o eign supplie s. This was ound in Kazakhs an be o e i
emo ed many o i s LCRs; co espondingly, when he LCRs we e emo ed,
ew local jobs we e los o displaced.
Owne ship equi emen s can also c ea e an en i onmen conduci e o
lack o anspa ency, co up ion, and a o i ism, whe e bene i s may be cap-
u ed by local eli es emb acing en -seeking beha io s and ailing o encou -
age en ep eneu ial de elopmen . They end o be ine ec i e, as o eign
companies may bypass limi s imposed on di idend dis ibu ion and exe
e ec i e con ol o join en u es h ough sha eholde s’ ag eemen s. Expe i-
ence wi h such p ac ices in Sou h A ica p omp ed he au ho i ies o ins i u e
a Code o Good P ac ice o he Sou h A ican Mine al Indus y o de ine
e hics o conduc and p e en abuses such as on ing p ac ices (Ko inek and
Ramdoo, 2017).
P esc ip i e bene icia ion (downs eam p ocessing) equi emen s ha e
o en ailed because he e is li le gua an ee ha domes ic p ocessing indus-
ies can become compe i i e in he longe e m. Fo example, in la e 2019,
Zambia was o ced o end a 15% expo ax on aw gems ones because he ax
was dec easing in es men s in he sec o , and o e all p oduc ion had allen.
Ano he high-p o ile case was Tanzania’s ban on expo s o aw gold, sil e ,
coppe , and o he me allic mine als s a ing in Ma ch 2017. The ban did no
achie e i s goal o ha ing mining companies in es in p ocessing acili ies, and
aw mined gold s ockpiled in he coun y un il he ban was li ed (Geipel, de
Wee d , and Ala con, 2021). Ano he illus a ion among many is he Indone-
sian bauxi e sec o , which has no eco e ed i s expo s since he expo ban
on bauxi e was ins i u ed in 2014.
The examples o Indonesian nickel and Bo swanan diamonds show, how-
e e , ha go e nmen s ha e mo e le e age in nego ia ions wi h o eign
in es o s ega ding p esc ip i e bene icia ion equi emen s when he coun-
y possesses an excep ional esou ce endowmen . In he case o Indonesian
nickel, he d aw o a subs an ial sha e o global ese es o high-quali y o e,
high p ojec ed demand o he mine al o ba e ies, and con inued depend-
ence on access o he mine al by China o ced in es men s in downs eam
smel e s. In Bo swana, he a o able condi ions gua an eeing access o cheape
ough diamonds g an ed o downs eam cu ing and p ocessing i ms incen-
i ized in es o s o open acili ies. Howe e , e en hese examples call in o

70 Jane Ko inek and Paulo De Sa
ques ion he longe - e m sus ainabili y o he downs eam p ocessing ope a-
ions – in Indonesia o easons o in es men clima e and in Bo swana due o
alling diamond ese es and lowe p oduc i i y among diamond-cu ing and
-polishing ope a ions compa ed wi h in e na ional compe i o s. Mo eo e ,
he global ma ke o he speci ic mine al mus p esen e y a o able ends,
wi h high p ojec ed u u e demand h ea ening a po en ial si ua ion o sca -
ci y, o such policies o be easible e en in he medium e m.
Go e nmen s a e mo e likely o achie e hei aims i hey a e ba gaining
om a a o able posi ion ei he because hey ha e s ong nego ia ing capac-
i y o because he mining company needs o enew he e ms o i s license o
concession in a mine ha is an essen ial componen o i s p o i abili y. Bo -
swana has expanded i s each h ough he alue chain o diamond alua ion,
agg ega ion, and so ing in a s ep-by-s ep ashion, in he con ex o subse-
quen license and pe mi nego ia ions (Ko inek, 2014). This g adual app oach
helped o ensu e he indus y pa ne was suppo i e o he policies and ha
esponses we e ound o supply-side cons ain s, such as he opening o a Dia-
mond Academy o ain po en ial employees o diamond so ing, aluing, and
agg ega ion jobs.
Al hough empi ical e idence sugges s ha “bes e o s” clauses do no
ha e much impac on hei own (Geipel, de Wee d , and Ala con, 2021), he
eali y on he g ound sugges s ha manda o y quan i a i e LCRs ha e ye o
gene a e signi ican amoun s o locally sou ced inpu s o s eng hened in e -
sec o linkages. Local employmen and some aining o he local wo k o ce
appea o ha e been hei mos success ul ou comes (Ellis and McMillan,
2020; Whi e, 2017).
Supply-side ini ia i es appea o wo k be e han p esc ip i e a ge s when
hey ocus on building wo k o ce skills – h ough bo h a ge ed aining and
skills- ans e ac i i ies, including in business managemen skills – and he
capaci y o local businesses o supply goods and se ices compe i i ely and
in eg a e in global supply chains (McCulloch e al., 2017). Some o he mos
in e es ing examples come om public–p i a e pa ne ships ha combine
mining i ms’ SDPs wi h go e nmen -sponso ed skills ans e and aining
ini ia i es de eloped in close collabo a ion wi h uni e si ies and echnical
cen e s. These exis in many coun ies; one example co e ed abo e comes
om Chile’s coppe sec o .
Mo e gene ally, i appea s ha coun ies applying mo e inclusi e app oaches
ha e been mo e success ul han hose ollowing mo e p o ec ionis me hods
ocused on sho - e m goals and na ow a ge s. Policies ha ail o conside
long- e m economic di e si ica ion objec i es, ocusing ins ead on na ow a -
ge s o he local sou cing o goods and se ices and he employmen o he
local wo k o ce, a e likely o gene a e businesses and skills ha a e dependen
on he mining sec o o e en on a speci ic p ojec (Weldegio gis, Die sche,
and F anks, 2021; Lebdioui, 2019). Value added, c ea ed exclusi ely wi hin
he mining indus y, pe pe ua es he ulne abili y o commodi y p ice luc ua-
ions and mac oeconomic shocks (Ma cel e al., 2016; ACET, 2017). This
Local con en policies in he mining sec o 71
is he case in pa icula o he downs eam linkage policies implemen ed in
Bo swana.
LCRs end, inco ec ly, o be conside ed iscally neu al, no p esen ing any
inancial implica ions o he go e nmen . In ac , hey can c ea e iscal sho -
comings i hey nega i ely a ec he p o i abili y o he mining sec o , educ-
ing he amoun o axes and oyal ies collec ed by he go e nmen . LCRs can
be cos ly o implemen due o he incen i es g an ed o p omo e he use o
local inpu s and de elop compe i i e local supplie businesses. Go e nmen s
o en ace a ade-o be ween maximizing ax e enues and de eloping local
con en , bu he ex a cos o adop ing LCRs can be jus i ied only i i expands
he ax base o e he long e m (Ma cel e al., 2016).
This will ail o ma e ialize when LCRs educe he indus y’s compe i-
i eness. P o ec ionis measu es can lead o he p e alence o uncompe i i e
supplie s o long pe iods, aising p oduc ion cos s and ha ing a de imen al
e ec on sec o s using mining p oduc s as inpu s (McCulloch e al., 2017).
On he o he hand, impo ed in e media e goods and se ices a e an impo -
an channel o inc ease p oduc i i y and he adop ion o new echnologies
and can play a signi ican ole in in eg a ion in global alue chains. By inc eas-
ing he specializa ion in he p oduc ion o speci ic inpu s, hey can gene a e
economies o scale ha maximize p oduc i i y and p o ide oppo uni ies o
mo e in o highe alue ac i i ies o e ime h ough upg ading (Ko inek and
Ramdoo, 2017).
Finally, LCRs can cause dis up ion in global ma ke s, in pa icula when
hey a ge mine als conside ed “c i ical” o ce ain key indus ial sec o s,
in lic ing supply isks, and a ec ing he sus ainabili y o indus ies in esou ce-
dependen coun ies. In a ew ins ances in which LCRs a e no WTO-com-
pa ible, his has led o cos ly dispu es. Fo example, beginning in 2006, China
imposed se e al es ic ions on he expo o a e ea h me als and quo as
on he expo o unp ocessed o es. This con ibu ed o a s eep inc ease in
p ices s a ing in 2010. In 2012, he Uni ed S a es ini ia ed a dispu e a he
WTO agains hese es ic ions, which he EU, Japan, and Canada joined as
complainan s. In 2014, he WTO Appella e Body decided in a o o he com-
plainan s, and China was equi ed o emo e i s expo es ic ions.
Policy implica ions: om local alue added o in eg a ing
globalsupply chains
Economy-wide impac s mus be aken in o accoun when conside ing imple-
men ing LCRs. I used, LCRs should be pa o b oade public policies, ins i-
u ional a angemen s, and pa ne ships ha public au ho i ies pu in place
o le e age he sec o ’s con ibu ion o b oade economic di e si ica ion
(Die sche, 2017). In pa icula , i should be kep in mind ha suppo ing
sec o s ha a e dependen on mine al ex ac ion may no inc ease economic
di e si ica ion, may inc ease he impac s o esou ce p ice cycles, and may ha e
subs an ial en i onmen al consequences.
72 Jane Ko inek and Paulo De Sa
Mo e gene ally, go e nmen s can adop a b oad se o ho izon al meas-
u es o emo e cons ain s o business de elopmen h oughou he econ-
omy, wi h an aim o p omo e mac oeconomic s abili y; p o ide egula o y
cla i y and s abili y; suppo SMEs; and imp o e in as uc u e in a eas like
ene gy, anspo , communica ions, in o ma ion echnology, and inance. A
suppo i e en i onmen can be c ea ed o he mining indus y and i s sup-
plie s by emo ing o e ly bu densome egula o y equi emen s, imp o ing
business in as uc u e, p omo ing skills de elopmen , s eng hening ins i u-
ional coo dina ion and collabo a ion among local supplie s, and suppo ing
inc eased access o inance, in pa icula o SMEs o en ill-equipped o access
global supply chains (A DB and Bill and Melinda Ga es Founda ion, 2015).
LCRs ha a e mo e likely o deli e on hei expec a ions a e (i) guided by
a comp ehensi e unde s anding o i ms’ p ocu emen needs, s a egies, and
capabili ies; (ii) based on a ho ough unde s anding o he local capaci ies
and bo lenecks o hei enhancemen ; and (iii) cognizan o he ac o s ha
may impac policy e ec i eness and he po en ial unin ended consequences
o such measu es.
I p esc ip i e LCRs a e o be in oduced, hey should be based on o e-
cas s o he mining indus y’s u u e needs, spending p ojec ions o speci ic
goods and se ices in exis ing and u u e mining p ojec s, and assessmen s o
he capaci y o local i ms o supply goods and se ices o su icien quali y a
compe i i e p ices. An e alua ion o he skills and compe encies equi ed by
he indus y, as well as he iming and quan i y o labo o ce equi emen s,
should be used o in o m po en ial policy e o ms in educa ion and aining in
o de o s eng hen and upg ade skills (IGF, 2018a, 2018c).
LCRs mus be aligned wi h wha can be ealis ically implemen ed, gi en he
capaci y o local supplie s, o ensu e he long- e m compe i i eness o he min-
ing indus y. Be o e he decision is made o in oduce hem, a de ailed analysis
o he p ocu emen needs o mining i ms should be unde aken, ideally in
close conjunc ion wi h he i ms, as was done by Canada and Aus alia. The
analysis should assess he ype o p ocu emen oppo uni ies ha a e a ailable
in majo p ojec s o each phase in hei li e cycle, as well as he capaci y o
domes ic supplie s and he wo k o ce o espond o he needs o hese p o-
jec s, leading o he iden i ica ion o he gaps ha need o be illed o enable
hem o ake ad an age o exis ing and po en ial oppo uni ies.
I demand-side policies a e implemen ed, hey end o wo k be e when
pu sued in conjunc ion wi h supply-side policies. Fo example, go e nmen s
can complemen egula ion es ablishing minimum a ge s o quo as o local
employees by implemen ing measu es o p omo e aining and skills de elop-
men o he wo k o ce. Al hough he o m and con en o aining plans can
be a olun a y componen o a company’s human esou ces p og am, some
coun ies ha e manda ed aining equi emen s o suppo o he de elop-
men o aining acili ies (IGF, 2018e).
P esc ip i e LCRs en ail long ime ames be ween hei announcemen and
e ec i e implemen a ion. As such, hey a e ulne able o changes in ma ke
Local con en policies in he mining sec o 73
p ices and in poli ical commi men (ACET, 2017). I coun ies none heless
wan o implemen hem, hey should be in oduced g adually, in a phased
app oach, along wi h capaci y-building e o s o allow domes ic supplie s
o adjus o hem. Go e nmen s should pe iodically assess p og ess agains
objec i es, adjus ing hem as local capaci y inc eases and knowledge abou
u u e supply and demand imp o es. Any p o ec ion p o ided o local suppli-
e s should be empo a y and disbanded g adually. Sunse clauses, which p e-
sc ibe ha a egula ion shall cease o ha e e ec a e a speci ied da e, a e good
p ac ice in local con en policies (A ican Na u al Resou ces Cen e, 2021).
Mining companies migh be equi ed o olun a ily ake he ini ia i e o
design and implemen SDPs. La ge companies a e be e placed o de elop
and implemen such p og ams, bu coope a ion among i ms, coo dina ed,
o ins ance, by a Chambe o Mines, can acili a e he pa icipa ion o
smalle companies. Examples o success ul supplie de elopmen p og ams
include Anglo Ame ican’s Zimele En e p ise P og am in Sou h A ica and
he join Newmon –In e na ional Finance Co po a ion’s Aha o Linkages
p og am in Ghana.
In some cases, mining companies suppo local supplie s by implemen ing
measu es o he han LCRs, such as (i) b eaking la ge con ac s in o smalle
ones (unbundling) so ha local i ms may p o ide a smalle po ion o he
o al con ac ende ed; (ii) pos ing all con ac s, ende oppo uni ies, and
ins uc ions o bidding p ocesses on local supplie po als; (iii) sole-sou cing
a angemen s wi h local supplie s o i ms om disad an aged g oups; (i )
s ipula ing equi emen s o ou side supplie s o subcon ac o en e in o
join en u es wi h local i ms; ( ) assigning highe p e e ence weigh ings
o local businesses in compe i i e bidding p ocesses o p o iding hem wi h
longe ime ames o bidding; ( i) p ice-ma ching, allowing local supplie s
o ma ch he p ice o o he supplie s; ( ii) suppo ing local supplie s o ob ain
he ce i ica ions necessa y o espond o ende s o compe e o con ac s;
and ( iii) using p ocu emen me hods in which bids a e awa ded o local sup-
plie s when hei p ice is wi hin a ce ain pe cen age o he bes o e – e.g., 2%
in Ghana, 10% in Tanzania, o wi hin 20% o he lowes o eign bid p ice in
Kazakhs an (Es e es e al., 2014).
Moni o ing and epo ing o policy ou comes, po en ially wi h in-buil
sanc ions o noncompliance, is key o inc easing unde s anding abou
policy design and implemen a ion. Go e nmen s and indus y should
publicly epo on p ocu emen p ocesses, con ac ed supplie s, spending,
and ax implica ions o imp o e o e sigh and accoun abili y (Pi man and
To oskainen, 2020). Moni o ing o LCRs has, in many cases, shown hem
o be ine ec i e. Howe e , Aus alia and Ghana ha e implemen ed s ong
epo ing mechanisms wi h sanc ions i i ms do no epo on hei local
con en objec i es.
Mining companies a e highly dependen on con ac o s and supplie s as a
sou ce o echnological inno a ion. Empi ical e idence shows ha local sup-
plie s ha h i e a e gene ally incumben i ms whose expe ience in he ma ke
80 Jane Ko inek and Paulo De Sa
p. 102312 www.sciencedi ec .com/science/a icle/abs/pii/S0301420721003226?
ia%3Dihub
Whi e, S. 2017. Regula ing o Local Con en : Limi a ions o Legal and Regula o y
Ins umen s in P omo ing Small Scale Supplie s in Ex ac i e Indus ies in De elop-
ing Economies. The Ex ac i e Indus ies and Socie y. 4 (2). pp.260–266. h ps://
esea ch eposi o y.mu doch.edu.au/id/ep in /34791/
Wo ld Bank, and Kaise Economic De elopmen Pa ne s. 2015. A P ac ical Guide o
Inc easing Mining Local P ocu emen in Wes A ica. Washing on, DC: Wo ld Bank.
h ps://openknowledge.wo ldbank.o g/handle/10986/21489
Yunis, J., and E. Aliakba i. 2020. F ase Ins i u e Annual Su ey o Mining Compa-
nies 2020. Vancou e : F ase Ins i u e. www. ase ins i u e.o g/si es/de aul / iles/
annual-su ey-o -mining-companies-2020.pd

Annex. Consis ency wi h
WTO ules
Table 3.A1 Consis ency o Local Con en Policies wi h WTO P o isions
Measu es Rele an WTO
P o isions
Consis ency wi h
WTO
Measu es a ec ing sou cing o inpu s
Local p ocu emen equi emen s
Quo a ela ed o
local sou cing
A pe cen age o
alue added o
speci ic olume o
in e media e inpu s
o be pu chased
locally
TRIMs illus a i e
lis pa a. 1 (a)
Quo as o
speci ic
pe cen ages
p ohibi ed
T ade balancing
equi emen s
Impo s o one
p oduc linked o
expo pe o mance
o o he p oduc s
TRIMs illus a i e
lis 1(b)
o in e nal
measu es; 2
(a) o bo de
measu es
P ohibi ed
Manu ac u ing
equi emen s
Ce ain p oduc s
a e equi ed o be
manu ac u ed locally
TRIMs illus a i e
lis
P ohibi ed
Limi a ions on
impo s
Amoun o goods and
se ices ha can be
impo ed o he
p oduc ion p ocess is
limi ed
GATT A . III.5;
GATT A .
XI.1; TRIMs
illus a i e lis ,
pa a. 2(a)
P ohibi ed
Fo eign exchange
es ic ions
Res ic he in low o
o eign exchange
a ibu able o
an in es o o
cons ain he
amoun o impo ed
in e media e goods
TRIMs illus a i e
lis , pa a 2 (b)

P ohibi ed
Excep ion o
de eloping
coun ies
GATT A s.
XII and
XVIII:B
(Con inued )
82 Jane Ko inek and Paulo De Sa
Table 3.A1 (Con inued)
Measu es Rele an WTO
P o isions
Consis ency wi h
WTO
P e e ence o In es o s o pu chase
local subs i u es local subs i u es
o impo s i
“like p oduc ” is
manu ac u ed locally
GATT A icle
III.4 (na ional
ea men ) i
(i) impo ed
p oduc s a e
acco ded less
a o able
P ohibi ed
ea men
compa ed o
local supplie s;
(ii) impo ed
goods and
he domes ic
p oduc s a e
conside ed like
p oduc s; and
(iii) measu es
a e insc ibed
in laws,
egula ions, and
equi emen s.
Owne ship equi emen s
Local equi y Some p opo ion o
pa icipa ion equi y mus be held
locally
GATS A .
XVI o
ma ke access
es ic ions
and A . XVII
o na ional
ea men , in
schedule o
commi men s
P ohibi ed only
i coun ies
ha e aken
commi men s
in hei
se ices
schedules;
o he wise no
disciplined
Employmen equi emen s
Local Speci ied employmen
employmen a ge s mus be me
a ge s
GATS A .
XVI o
ma ke access
es ic ions
and A . XVII
o na ional
ea men ,
p o ided in
schedule o
commi men s
P ohibi ed only
i coun ies
ha e aken
commi men s
in hei
se ices
schedules;
o he wise no
disciplined
Local con en policies in he mining sec o 83
Measu es Rele an WTO
P o isions
Consis ency wi h
WTO
Quo as o A maximum numbe
o eign o expa ia e s a is
employmen speci ied
Na ional Ce ain s a mus
pa icipa ion in be na ionals o
managemen a schedule o
“indigeniza ion” o
managemen mus
be se
Technology ans e equi emen s
R&D In es o s should
equi emen s commi o in es in
R&D locally
Technology Speci ied o eign
ans e echnology be used
locally
Measu es a ec ing p oduc ion
Minimum expo Ce ain pe cen age o
equi emen s p oduc ion mus be
expo ed
T ade balancing Impo s mus be a
equi emen s ce ain p opo ion
o locally p oduced
expo s, ei he in
e ms o olume o
in e ms o alue
Domes ic sales Ce ain p oduc may
equi emen s no be expo ed
Ma ke ese e Some ma ke s a e
policy ese ed o local
p oduc ion
GATS A . IV;
TRIPS A s. 3,
7, and 8; SCM
Ag eemen
A s. 2 and 8
GATT A . III.5;
GATT A .
XI.1; TRIMs
illus a i e lis ,
pa a. 2(a)
TRIMs illus a i e
lis 1(b)
o in e nal
measu es; 2
(a) o bo de
measu es
GATT A . III.5;
GATT A .
XI: 1; TRIMs
illus a i e lis
2(c)
GATT A . III.4
P ohibi ed only
i coun ies
ha e aken
commi men s
in hei
se ices
schedules;
o he wise no
disciplined
P ohibi ed only
i coun ies
ha e aken
commi men s
in hei
se ices
schedules;
o he wise no
disciplined
P ohibi ed
No disciplined
P ohibi ed
P ohibi ed
P ohibi ed
P ohibi ed
(Con inued )
84 Jane Ko inek and Paulo De Sa
Measu es Rele an WTO
P o isions
Consis ency wi h
WTO
P oduc
manda ing
equi emen s
Some p oduc s o be
expo ed by he
hos ing coun y only
GATT A . III.5;
GATT A .
XI: 1; TRIMs
illus a i e
lis 2(c)
P ohibi ed
Licensing
equi emen s
In es o s o ob ain
license o
p oduc ion in he
hos coun y
GATT A . XI.1 P ohibi ed
Technology
ans e s
In es o s a e
commi ed o a
speci ied embodied
echnology
TRIPS A s. 3,
7, and 8; SCM
Ag eemen
A s. 2 and 8
Disciplined
O he measu es
ele an o
LCPs
S a e ading
en e p ises
Fo eign i ms o en e
in o join en u e
wi h SOEs
A icle XVII
o GATT,
applicable when
SOEs en e
comme cial
ope a ions
P o ision does
no egula e
obliga ions o
o eign i ms
o en e in o
join en u e
wi h SOEs
Subsidies o
suppo local
supplie s
Go e nmen s gi e
inancial incen i es
o local supplie s o
a o local p oduc s
o e impo s
SCM A . 3.1(b) Ac ionable
i speci ic;
o he wise
non-ac ionable
Subsidies o R&D
and inno a ion
Go e nmen policies
suppo R&D and
inno a ion
SCM A . 8.2 Ac ionable
i speci ic;
o he wise
non-ac ionable
A . = A icle, GATS = Gene al Ag eemen on T ade in Se ices, GATT = Gene al Ag eemen
on Ta i s and T ade, LCP = local con en policy, R&D = esea ch and de elopmen , SCM =
Ag eemen on Subsidies and Coun e ailing Measu es, SOE = s a e-owned en e p ise, TRIMs
= T ade-Rela ed In es men Measu es, TRIPS = Ag eemen on T ade-Rela ed Aspec s o In el-
lec ual P ope y Righ s, WTO = Wo ld T ade O ganiza ion.
No e: Excep ions o de eloping coun ies – de eloping coun ies a e pe mi ed o e ain TRIMs
ha cons i u e a iola ion o GATT A . III o XI, p o ided he measu es mee he condi ions o
GATT A . XVIII, which allows speci ied de oga ion om he GATT p o isions o he economic
de elopmen needs o de eloping coun ies.
Sou ce: Ko inek and Ramdoo (2017), adap ed om G eenaway (1992); McCulloch e al. (2001).
Table 3.A1 (Con inued)
Local con en policies in he mining sec o 85
Table 3.A2 Measu es No P ohibi ed by WTO Rules
Measu es Rema ks
Measu es a ec ing impo s
Ta i measu es WTO does no p ohibi a i s. Coun ies mus
bind hei a i s and can modi y hei a i a es
wi hin he ange i bound a i s a e di e en o
applied a i s.
Non a i measu es (o a Gene ally p ohibi ed (QRs, licensing, e c.) bu
quan i a i e na u e) wi h he excep ion o imposi ion o impo
quo as o BOP pu poses (A . XVIII:B). This is
empo a y in na u e.
Measu es o suppo en e p ises
Exchange a es No WTO ag eemen deals exp essly wi h exchange
a es, al hough GATT A . XV conce ns
exchange a angemen s.
Go e nmen Pe mi ed, excep i a coun y is membe o he
p ocu emen GPA.
Expo inance/ Allowed bu may be conside ed an expo subsidy
insu ance/gua an ees i hey a e g an ed a p emium a es insu icien
o co e long- e m ope a ing cos s and losses.
P oduc ion subsidies Allowed i nonspeci ic*
T ade inance No p ohibi ed
Measu es o p omo e echnology
Technology- ela ed No p ohibi ed
equi emen s o FDI
(e.g., echnological
ans e )
Suppo o R&D/ Unless speci ic, o he wise pe mi ed
inno a ion
Human capi al No p ohibi ed
de elopmen
Employmen o local No p ohibi ed
labo
Regional assis ance No p ohibi ed
In es men incen i es
Expo pe o mance No p ohibi ed
equi emen as
a condi ion o
in es men
Equi y equi emen No p ohibi ed
byFDI
Measu es subjec o disciplines unde speci ic ci cums ances
C edi subsidies No p ohibi ed, p o ided hey a e no p oduc o
sec o speci ic
(Con inued )

86 Jane Ko inek and Paulo De Sa
Table 3.A1 (Con inued)
Measu es Rema ks
Tax subsidies/holidays No p ohibi ed, p o ided hey a e no p oduc o
sec o speci ic
Clus e s/EPZ/SEZ No specially egula ed by a pa icula WTO
Ag eemen ** bu may be subjec o disciplines
when measu es con a ene o he WTO
disciplines (e.g., subsidies, e c.). Fiscal acili a ion
p o ided in SEZ is no p ohibi ed.
Con ingency measu es
Sa egua d measu es These measu es allow coun ies o apply impo
An i-dumping measu es es ic ions in pa icula ci cums ances, p o ided
Coun e ailing hey can p o e hei economy/economic ac o s
measu es a e a ec ed by (i) a su ge in impo s (sa egua d);
(ii) a p oduc ha is being sold below i s no mal
p ice on he domes ic ma ke by an expo ing
coun y (dumping); and (iii) a dis o ing e ec
o a subsidy by a o eign go e nmen .
A . = A icle, BOP = balance o paymen s, FDI = o eign di ec in es men , GATT = Gene al
Ag eemen on Ta i s and T ade, GPA = Go e nmen P ocu emen Ag eemen , QR = quan i a i e
es ic ions, R&D = esea ch and de elopmen , SCM = Ag eemen on Subsidies and Coun e ail-
ing Measu es, SEZ = Special Expo Zone, WTO = Wo ld T ade O ganiza ion.
* The WTO Ag eemen on Subsidies and Coun e ailing Measu es disciplines he use o subsi-
dies. The disciplines only apply o “speci ic subsidies,” ha is, o subsidies a ailable only o an
en e p ise, indus y, g oup o en e p ises, o g oup o indus ies in he coun y ha gi es he
subsidies. They can e e o domes ic o expo subsidies.
** SEZ is men ioned in a oo no e o GATT A . XVI and in he SCM, excluding om he de ini-
ion o a subsidy one o he iscal acili a ion measu es p o ided o SEZs – an exemp ion om
impo du ies and axes on goods expo ed om SEZs.
Sou ce: Ko inek and Ramdoo (2017).
DOI: 10.4324/9781003415794-4
This chap e has been made a ailable unde a CC-BY-NC-ND 4.0 license.
1 In oduc ion
When he T ump adminis a ion launched i s e ision o he ea y go e ning
ade be ween he US and i s neighbo s, he US nego ia o s emphasized he
need o s ic e ules o o igin. The US T ade Rep esen a i e, Robe Ligh -
hize , epo edly asked his coun e pa s o aise he egional con en equi e-
men (RCR) o 85%, a la ge inc ease om he le el se in 1993 (62.5%).1
Canada and Mexico balked a such a high a e, and he h ee pa ies inally
se led on an inc ease o 75%, bols e ed wi h addi ional binding equi e-
men s. The poli ical appeal o s ic e o igin ules lies in he hope ha hey
will inc ease domes ic employmen in he pa s indus y. Ligh hize (2020)
acknowledged his in en , w i ing “The USMCA ebalances he NAFTA o
p omo e inc eased p oduc ion in he Uni ed S a es and No h Ame ica.”
F om an economic s andpoin i is ha d o jus i y one ous es ic ions on
sou cing. I he goal is me ely o limi impo s o pa s, hen a i s on pa s
would be a mo e e icien ool. While ade ag eemen s ha lack a common
ex e nal a i need some ule o o igin o p e en back-doo en y o he
high- a i ma ke ia he low- a i coun y, his issue was no ele an in he
USMCA nego ia ion o wo easons. Fi s , because he ac ual di e ences in
a i s we e small, so much smalle con en es ic ions would be su icien o
p e en his a i -hopping.2 Second, i was he lowe - a i membe , he US,
ha was asking o he s ic e ules.
Going back o he wo k o G ossman (1981), economis s ha e in es iga ed
whe he , e en as p o ec ionis de ices, s ic ules o o igin could ail o achie e
hei goals. G ossman’s P oposi ion 3 (p. 591) s a es ha small inc eases in
local con en equi emen s ha e ambiguous e ec s on indus y alue added,
de ined as he sum o alue added in componen s and in inal goods.
Whe eas he con en p o ec ion policy causes an inc ease in he ou pu
o domes ic componen s, i will no mally esul in a concomi an con-
ac ion o inal good p oduc ion. Which e ec will domina e depends
on how sensi i e in e media e good p oduc ion is o changes in i s ou -
pu p ice, and how sensi i e inal good p oduc ion is o changes in he
p ice o i s in e media e inpu .
4 The unin ended consequences o
high egional con en equi emen s
Kei h Head, Thie y Maye , and Ma c Meli z
88 Kei h Head, Thie y Maye , and Ma c Meli z
In his chap e , we ex end he G ossman app oach o ake in o accoun he
e y la ge numbe o di e se pa s ha go in o mode n manu ac u ed goods
such as au omobiles. Fo each pa , he i m decides whe he o sou ce i om
inside he egion (whe e he e is a ee ade ag eemen ) o om ou side coun-
ies. The co e ade-o he i m aces is ha wi hin- egion sou cing helps i
comply wi h ules o o igin (RoO), bu necessi a es o going oppo uni ies o
ob ain cheape pa s elsewhe e. In sec ion 4, we gi e an o e iew o he heo-
e ical model de eloped in Head e al. (2022) ha analyzes hese ade-o s.
We show ha RoOs gene a e compe ing incen i es o pa sou cing wi hin a
Regional T ade A ea (RTA). E en hough he ules a e in ended o eloca e
p oduc ion o pa s wi hin he RTA, hey can ha e he opposi e e ec when
hey a e o e ly es ic i e. This main esul does no wo k ia declines in inal
goods p oduc ion, as in G ossman (1981). Howe e , we also quan i y he
nega i e impac o highe cos s induced by he RoOs o pa p oduc ion.
This quan i ica ion exe cise p edic s how any gi en RoO would a ec ma ke
sha e changes and he associa ed p oduc ion and employmen changes ac oss
all ehicle plans selling in he egion. D awing on he a ac i e agg ega ion
p ope ies o ou model, we de i e a e age p ice, ma ke sha e, p oduc ion,
and pa employmen changes ac oss g oups o ca lines – including he g oup
o all ca lines assembled wi hin he egion.
This chap e is o ganized as ollows. Sec ion 2 p o ides an o e iew o
ecen changes in ules o o igin ha impac ed he au o indus y in No h
Ame ica and Eu ope. The ollowing sec ion p esen s empi ical pa e ns o
sou cing in No h Ame ica ha in o m he model and he way we quan-
i y i . Sec ion 4 summa izes he key mechanisms o he model de eloped in
Head e al. (2022). We hen es ima e he model o i he pa e n o sou cing
obse ed a he le el o indi idual ca models p io o he 2020 changes in
RCRs. Sec ion 5 desc ibes how we use ha i ed model o e alua e he impac
o coun e ac ual RoOs. Sec ion 6 epo s he e ec s o changing hose ules
o bo h NAFTA and he EU-UK ade ag eemen .
2 Changing ules o o igin in No h Ame ica and Eu ope
Rules o o igin in he au o indus y we e i s in oduced in he 1965 Au o
Pac be ween Canada and he Uni ed S a es. To a oid non-US companies se -
ing up sales en e p ises in Canada o se e he US ma ke , i was ag eed ha
only ca s wi h 50% con en om he US and Canada would bene i om he
new a i - ee egime.3 In he nego ia ion o he No h Ame ican F ee T ade
Ag eemen in 1991, he Ame ican side sough a mo e es ic i e ule. I win
(2017) desc ibes he ini ial nego ia ing posi ions and how hey eached he
peculia egional con en equi emen o 62.5%:
Rules o o igin we e pa icula ly impo an in he case o au omobiles.
The US au o indus y wan ed high No h Ame ican con en ules o
ensu e ha Mexico did no become an expo pla o m o Japanese
Unin ended consequences o high egional con en equi emen s 89
o o he o eign p oduce s who would simply send pa s o Mexico o
assembly and hen ship he ehicles in o he Uni ed S a es. .. . Fo
NAFTA, he Uni ed Au o Wo ke s pushed o an 80 pe cen ule, Fo d
and Ch ysle 70 pe cen , and Gene al Mo o s 60 pe cen . Mexico and
Canada wan ed o keep he 50 pe cen equi emen in he US-Canada
FTA, bu eluc an ly accep ed 60 pe cen . US nego ia o s had p omised
au o p oduce s a numbe highe han 60 pe cen o p e en hei oppo-
si ion. While hey we e able o pe suade Mexico o go o 65 pe cen ,
Canada emained i m a 60 pe cen and so he nego ia o s spli he di -
e ence and a i ed a a 62.5 pe cen ule.
I win goes on o desc ibe how he US comp omise led o an apoplec ic call o
he US ade nego ia o om Fo d’s CEO, who el be ayed by he ailu e o
ob ain he p omised 65%. The case poin s o he cen al impo ance assigned
o ules o o igin as well as he p esump ion ha US p oduce s would bene i
om a s ic e ule o o igin han he one he US had se led on o NAFTA.
When P esiden T ump’s nego ia o s se ou o eplace NAFTA, one o
hei ocal poin s was s ic e ules o o igin o he au o indus y. E en u-
ally, Canada, he US, and Mexico ag eed in 2019 o eplace he 1994–2020
NAFTA wi h a new ag eemen called he USMCA (in he Uni ed S a es).
Ligh hize (2020) o e ed he ollowing jus i ica ion o s ic e ules o o igin:
The USMCA ebalances he NAFTA o p omo e inc eased p oduc ion
in he Uni ed S a es and No h Ame ica and o ensu e ha non-pa ies
do no gain unwa an ed bene i s h ough he ag eemen . The USMCA
ea u es inno a i e ules o o igin o au omobiles and au omobile pa s
ha , once ully implemen ed, will c ea e s ong incen i es o in es and
manu ac u e in he Uni ed S a es and No h Ame ica.
The new ag eemen de o ed 39 pages in an appendix o he new ules, so we
canno do ull jus ice o hei complexi y he e. The ollowing we e he main
ways in which he equi emen s o quali ying o a i - ee ea men became
mo e di icul o he au o sec o :
1 The minimum No h Ame ican egional con en equi emen (RCR) was
inc eased o 75% ( om 62.5%).
2 A new labo alue con en (LVC) ule equi es ha 40% o 45% o au o
con en be made by wo ke s ea ning a leas $16 pe hou .
3 Se en y pe cen o bo h he s eel and he aluminum going in o each ca
mus o igina e in No h Ame ica.
4 Six “supe -co e” pa s – including engines and ansmissions – mus hem-
sel es comply wi h he 75% RCR.
The new equi emen s a e clea ly in ended o discou age i ms om
sou cing pa s om ou side No h Ame ica: i he ehicles cu en ly
96 Kei h Head, Thie y Maye , and Ma c Meli z
3.2 No h Ame ican inpu cos sha es (AALA da a)
Ou sou ce o da a ega ding a ia ion in egional cos sha es is based on
annual epo s manda ed by he Ame ican Au omobile Labeling Ac (AALA)
o 1992. The law equi es ha “A label wi h he US/Canada con en pe cen -
age and ela ed addi ional in o ma ion mus be displayed on hese ehicles up
o he ime o i s e ail sale.” Acco ding o AALA, each new passenge mo o
ehicle mus be labeled wi h he ollowing in o ma ion:
1 The pe cen age o US/Canadian equipmen (pa s) con en
2 The name and pe cen age con en o any coun ies o he han he US and
Canada ha indi idually con ibu es 15% o mo e o he equipmen con-
en (wi h a maximum o wo coun ies)
3 The coun ies o inal assembly, engine manu ac u e, and ansmission
manu ac u e.
The da a a e a ailable in PDF o m on he AALA websi e.8 In o ma ion on
componen supplie s o he han he US and Canada begins in 2011. The cos
sha e da a is epo ed by AALA a he ca line le el, which usually co esponds
o a b and-model assembled a a speci ic ac o y. AALA o en p o ides mo e
de ail o ca lines, wi h in o ma ion such as engine size.
We ep esen he model-le el AALA da a as a collec ion o cumula i e den-
si ies in Figu e 4.3. These a e plo ed wi h he o iginal da a pooled o e he
2011–2020 pe iod. We plo he CDFs sepa a ely o he ca s ha a e he mos
020406080
100
0.0 0.2 0.40.6 0.
81
.0
No h Ame ican con en sha e
F ac ion wi h < x
J/K/G−made
Canada
RCR
NAFTA
RCR
USMCA
Figu e 4.3 NAFTA Regional Cos Sha e by Loca ion o P oduc ion (CDFs)

Unin ended consequences o high egional con en equi emen s 97
po en ially a ec ed by he RoO, i.e., hose p oduced in Canada, Mexico, and
he US. We also p esen sepa a e densi ies o he Japanese, Ko ean, and Ge -
man b ands ha a e p oduced in NAFTA (J/K/G make). Finally, we also plo
a densi y o he models sold in he US bu assembled in Japan, Ko ea, o
Ge many (J/K/G made).
The AALA epo s gi e es ima es o he sha e o pa s cos s, no accoun -
ing o assembly cos s. In o de o compa e hose numbe s o he RCR,
we he e o e need o add on he egional cos s a ibu able o assembly.
Figu e 4.3 compu es he o e all egional cos sha e unde he assump ion
ha inal assembly amoun s o 15% o he o al p oduc ion cos o each
egionally made ca .
Fou main indings eme ge. The majo i y o ca lines in each NAFTA coun-
y ha e cos sha es ha indica e compliance wi h he 62.5% RCR p esc ibed
by he o iginal NAFTA. Second, compliance is highes in Canada, lowes in
Mexico, and in e media e in he US. Ca b ands headqua e ed in he h ee
majo ca -p oducing coun ies ou side NAFTA ha e lowe NAFTA inpu
sha es e en when p oducing in NAFTA. Finally, No h Ame ican cos sha es
o ca s assembled ou side No h Ame ica end o be e y small.
4 A heo e ical model o pa s sou cing
As we p e iously discussed, ules o o igin (RoO) can gene a e compe ing
incen i es o he loca ion o pa p oduc ion wi hin a egional ade a ea
(RTA). Those ules a e in ended o eloca e he p oduc ion o pa s wi hin
he RTA; bu when hey a e o e ly es ic i e, he impac on egional sou c-
ing will be e e sed and pa sou cing will be eloca ed ou side he egion.
We now ske ch a simple model based on ou companion pape Head e al.
(2022) ha illus a es why RoOs will induce such a hump-shaped esponse o
ha egional pa sha e. In o de o ocus on he sou cing decision o pa s
and he in ui ion o his hump-shaped esponse – which we call he La e
cu e o RoOs – we keep he loca ion o assembly ixed. Ou companion
pape shows how RoOs will also impac ha assembly loca ion choice and
how o e ly es ic i e RoOs will no only lead o lowe egional pa sou cing
bu also induce inal good p oduce s o eloca e assembly ou side he egion.
4.1 Model s uc u e
The po en ial o he downwa d-sloping segmen o he RoO La e cu e,
whe e s ic e RoOs lead o educ ions in he egional pa sha e, a ises when
inal good i ms (a ca line p oduce in ou da a) make sou cing decisions o
many pa s. Al hough we would echnically only need a minimum o wo
pa s o highligh his e ec , we de elop a model wi h a con inuum o pa s
due o i s analy ical ac abili y. And i also i s well wi h ou empi ical applica-
ion in which ca p oduce s make sou cing decisions on a e y la ge numbe
o pa s.
98 Kei h Head, Thie y Maye , and Ma c Meli z
Each ca pa can be sou ced om ei he wi hin he egion a one cos
o ou side he egion, deno ed Fo eign, a a di e en cos . Each pa cos
o egional and Fo eign p oduc ion is modeled as a s ochas ic d aw om a
Weibull dis ibu ion wi h pa ame e θ ≥ 1.9 We no malize he mean cos o
egional p oduc ion o 1. The mean cos o he Fo eign-sou ced pa s is δ >
0. This pa ame e a ies ac oss i ms. Fi ms wi h δ > 1 ha e a lowe egional
p oduc ion cos o pa s on a e age. As we men ioned ea lie , we igno e
he assembly loca ion choice in o de o ocus on he pa -sou cing decision
( egional o Fo eign); and we he e o e do no model he associa ed assembly
cos s un il he quan i ica ion in sec ion 6.
F ee T ade (No Rules o O igin) When he e a e no RoOs, a i m δ decides
whe he o sou ce each pa om ei he wi hin o ou side he egion based
on whiche e cos is lowe . This is he i m’s un es ic ed pa -sou cing choice,
which we deno e wi h a subsc ip U. The esul ing sha e o egionally sou ced
pa s is gi en by he p obabili y ha he egional cos o a gi en pa is lowe
han he Fo eign cos . Gi en ou dis ibu ional assump ions o he Weibull
cos d aws, ha p obabili y and esul ing sha e is:
dh h
U()sc
dg
--
11 (1)
Fi ms wi h highe δs ha e a compa a i e ad an age in egional pa p oduc ion
and hence sou ce a highe sha e o hei pa s domes ically. This sou cing deci-
sion hen leads o a o al pa s cos (agg ega ing o e bo h he egional and
Fo eign pa s) o CU(δ) = χU(δ)1/θ.
As we will see, hese cos di e ences will be inconsequen ial o a i m’s
esponse o a RoO, because ha will only depend on how a RoO inc eases he
i m’s cos abo e his benchma k CU(δ).
Rules o O igin A RoO manda es ha i ms sou ce a minimum ac ion o
hei pa s χR egionally, o else i will ace a Mos Fa o ed Na ion (MFN) a i
a e on he inal good expo ed wi hin he RTA. We model his addi ional cos
as an a e age a i τ > 1 incu ed ac oss all inal good uni s p oduced. In he
quan i ica ion in sec ion 5, we will cons uc his a e age a i a e based on
he sha e o a ca line’s wi hin-RTA expo s ela i e o all i s o he sales. I a
i m chooses o comply wi h he RoO and a oid he a i , i sou ces p og es-
si ely mo e expensi e pa s egionally ( ela i e o o eign-sou ced) un il he
minimum h eshold is me . In ou companion pape , we show how he sou c-
ing choices o comply wi h a RoO χR a e equi alen o he ones he i m would
make i a a i we e imposed on o eign pa s (wi h he a i e enue subse-
quen ly eba ed back o he i m). We also desc ibe he connec ions be ween
a RoO speci ied as a egional pa χR and a RoO speci ied as a egional cos
sha e λR: a manda ed minimum cos -sha e o egionally p oduced pa s. Bo h
ypes o RoOs ha e quali a i ely iden ical e ec s on egional pa sou cing
Unin ended consequences o high egional con en equi emen s 99
because he e is a mono onic ela ionship be ween χR and λR. This connec ion
is impo an in he quan i ica ion because RoOs o ca s in NAFTA and he
EU-UK TCA a e speci ied as cos sha es.
When a binding RoO χR > χU(δ) is manda ed, he i m’s o al pa cos
inc eases om CU(δ) o:
CRRR
dd d
d
dd
d
,
d
cd
d
dd
11
1 (2)
This ep esen s an inc ease in he i m’s o al pa cos ela i e o i s un e-
s ic ed (lowe bound) cos CU(δ) gi en by he a io

CCC
RRU
,,/()
(0
=
(0
>1
This cos a io cap u es he compliance cos penal y associa ed wi h he
RoO χR. I is ep esen ed in he op panel o Figu e 4.4 as a unc ion o he
RoO χR o h ee di e en i ms. An icipa ing ou empi ical applica ion, we
use ou i ed dis ibu ion o δs ac oss NAFTA-assembled ca lines. Fi m 2 has
δ2 = 0.12, which is he median δ ( ep esen ing a 12% a e age cos ad an age
o NAFTA-p oduced pa s).10 We hen show wo o he i ms (δ1 and δ3) ha
a e, espec i ely, a he 5% and 95% pe cen ile o ha empi ical dis ibu ion.
Fo any gi en i m – a gi en δ – he e is a ange in which i s un es ic ed sou c-
ing choice χU (δ) is abo e χR and he e o e complies wi h he RoO. The e is
no cos associa ed wi h compliance, so

C
(χR, δ) is a i s lowe -bound o 1. We
deno e his case complian -uncons ained. As he RoO χR ises abo e χU (δ),
compliance wi h he RoO en ails a cos compliance penal y

C
(χR, δ) > 1. As
an icipa ed, his cos penal y hen inc eases mono onically wi h he RoO χR:
compliance becomes inc easingly cos ly as he RoO becomes mo e es ic-
i e. Looking ac oss i ms, we see ha , as expec ed, he compliance cos wi h
a gi en RoO χR is always highe o i ms wi h lowe δ whene e hey a e no
uncons ained: hose i ms ha e a compa a i e ad an age in Fo eign-sou ced
pa s, so complying wi h a gi en RoO is mo e expensi e.
4.2 Compliance
As we men ioned, a i m δ can choose no o sa is y he RoO χR and ins ead pay
he a e age a i τ. I will do so whene e he compliance cos is g ea e han
he a i penal y:

C
(χR, δ) ≥ τ. In his case, we label he i m as non-complian ,
and i hen e e s o i s uncons ained pa sou cing wi h egional sha e χU
(δ) and associa ed cos CU (δ) = χU (δ)1/θ. The ho izon al line in he op panel
o Figu e 4.4 shows he example o a 6.2% a i penal y. Con inuing wi h ou
an icipa ed empi ical applica ion, his ep esen s he non-compliance a i ha
would be paid on a e age ac oss all ehicles assembled in Mexico based on
he empi ical p opo ion o Mexican-assembled ehicles ha a e expo ed o
100 Kei h Head, Thie y Maye , and Ma c Meli z
Figu e 4.4 Compliance Cos and Sou cing Decision o 3 Fi ms
i s NAFTA pa ne s, he Uni ed S a es and Canada, and hei associa ed MFN
a i s.
The bo om panel o Figu e 4.4 shows he egional pa sha e chosen by he
h ee i ms, gi en hei compliance decision. When he RoO χR is low enough,
all h ee i ms a e complian -uncons ained and choose hei un es ic ed pa
Unin ended consequences o high egional con en equi emen s 101
sha e χU (δ). This co esponds o he case o no compliance cos penal y,

C
(χR,
δ) = 1, in he op panel. As he RoO χR inc eases, i m 1, ollowed by i m 2
and hen i m 3 become complian -cons ained: The compliance cos penal y

C
(χR, δ) ises abo e 1 bu emains below he a i penal y τ. In his case, he
i ms choose he egional sha e χR o comply wi h he RoO. This is cap u ed
by he 45-deg ee inc easing line in he bo om panel: a chosen egional sha e
equal o he RoO. As he RoO χR u he inc eases, i m 1 and hen i m 2
choose non-compliance: he cos penal y is highe han he a i penal y. In
hose cases, hei chosen egional pa -sha es d op back o hei ini ial un e-
s ic ed le els χU (δ). No e ha i m 3 will ne e choose o be non-complian :
Complying wi h e en he mos es ic i e RoO o 100% is s ill less cos ly han
he a i penal y. We label i ms o his ype as always-complie s.
4.3 La e cu e o ules o o igin
Se ing aside hose i ms ha a e always-complie s, we see in Figu e 4.4 ha
inc easing a RoO om 0% o 100% will ini ially induce i ms o inc ease hei
egional pa -sha e – when hey a e complian -cons ained – bu will hen
induce hose i ms o sha ply educe hei pa -sha e once he RoO ises
abo e a h eshold whe e he i ms choose non-compliance. In ou companion
pape , we show ha his non-mono onic esponse, in his indi idual i m case
an in e ed-V, equi es a i m-sou cing decision o e mul iple pa s. When
he e is a single pa , ha non-mono onic sou cing esponse disappea s:
Inc easing he RoO can ne e induce a i m o educe i s egional pa -sha e.
And we also show ha as we smoo h ha in e ed-V sou cing esponse a he
i m le el o e a se o i ms wi h he e ogeneous δ, hen he a e age egional
sou cing sha e becomes a smoo h in e ed-U La e cu e. So long as we
exclude he always-complie s, hen he a e age egional pa -sha e e u ns o
i s ini ial (χR = 0) le el as he RoO inc eases o i s 100% upwa d bound. When
we conside he ull se o i ms including always-complie s, hen he a e age
egional pa -sha e emains abo e i s ini ial le el as he RoO inc eases o i s
upwa d bound.11
5 Simula ing policy changes in he model
The model delinea ed in he p e ious sec ion p o ides key quali a i e insigh s.
Mos impo an ly, i demons a es he unin ended consequences o an o e ly
s ic se o ules o o igin. When he cos o compliance is highe han he pen-
al y o non-compliance, i ms will op in o non-compliance, cu ing egional
inpu use down o hei uncons ained le els. The key unanswe ed ques ions
a e whe he ecen policy changes pu No h Ame ica in o his ange o coun-
e -p oduc i e ules. Answe ing his ques ion equi es us o calib a e se e al
di e en dimensions o he e ogenei y. We do his by inding pa ame e alues
ha induce he bes i be ween ou simula ed da a and he obse ed da a o
he p e-USCMCA pe iod, when he RCR was 62.5%.

102 Kei h Head, Thie y Maye , and Ma c Meli z
When aking he model o he da a, we ha e o ake a s and on he le el
a which he con en decision is made. While he model e e s o “ i ms,” he
AALA epo s show ha di e en ca lines owned by he same i m use e y
di e en sha es o No h Ame ican inpu s. Fo example, he made-in-Mexico
Fo d Fies a uses 80% No h Ame ican pa s, whe eas he US-assembled Fo d
Mus ang has 46% o i s pa s o igina ing in No h Ame ica. The Volkswagen
Gol R, made in Ge many, has only 1% o No h Ame ican pa s, bu he Gol
GTI assembled in Mexico has 42%. The US-assembled VW Passa has 61%
o he e sion wi h a 2.0-li e engine (made in Mexico) and jus 30% o he
3.6-li e e sion (engine impo ed om Ge many).12 Thus, he da a sugges
ha he con en decision is aken in esponse o a ia ion in ela i e cos s (δ in
he model) a he le el o speci ic ca lines. The ac ual decision-make could
be a plan manage o global headqua e s. In he model, i does no ma e
whe he he decision is cen alized, because p o i maximiza ion implies ha
cos s should be minimized o each ca line. The e is a single compliance deci-
sion o all he ehicles ha come ou o he same p oduc ion line, ega dless
o hei inal des ina ion. This assump ion comes om obse a ion in he IHS
Ma ki da a ha i is ex emely a e o he same ca line o sou ce a gi en
engine o ansmission om mo e han one coun y. Also, he AALA da a
p o ide single NAFTA sha es o each ca line.
I is impo an o simula e he model a he ca line le el because he a -
i penal y o non-compliance (τ in he model) a ies g ea ly ac oss ca lines
because o hei di e en sales des ina ions. Fo example, he Fo d Mus ang
has 2018 sales o 76,000 uni s in he US. These ca s will no pay any a i
penal ies o non-compliance wi h USMCA ules, no will he oughly 12,000
uni s headed o Aus alia and China.13 Only he 7,600 Mus angs sold in Can-
ada and he 1,900 sold in Mexico will ace MFN a i s as a penal y o non-
compliance wi h he USMCA RoO. The si ua ion o he Fo d Fies a made
in Mexico is e y di e en . The company sends he lion’s sha e o i s o al
p oduc ion (66,000 ca s) o i s USMCA pa ne s: o he US (52,000 ca s)
and Canada (1,200 ca s). Meanwhile, only 4,500 Fies as s ay in Mexico. The
o e whelming dominance o expo sales o NAFTA pa ne s gi es he Fies a
plan e y s ong compliance incen i es, as compa ed o he Mus ang. We
cap u e his impo an sou ce o he e ogenei y by using he IHS Ma ki da a
o compu e a i penal ies o e e y ca line.
The a i penal y ends o be much lowe han he MFN a i s because
la ge sha es o ou pu in he egional plan s o a ca make end o s ay wi hin
he coun y o p oduc ion o go o ma ke s ou side he egion (as in he
Mus ang example). Table 4.2 p o ides mo e g anula in o ma ion o he 20
la ges a i penal ies.
We use a simula ion o ou model o es ima e he unde lying he e ogene-
i y pa ame e s. The idea is ha ca lines ecei e hei compa a i e ad an age
“d aws” acco ding o a pa icula “guess” o he mean and s anda d de ia ion
o δ. A he same ime, hey d aw a pa ame e de e mining he impo ance
o assembly cos s o ha ca line. Then he simula ed ca lines each decide
Unin ended consequences o high egional con en equi emen s 103
Table 4.2 Top Ta i Penal ies o USMCA Ca lines in 2018
B and Model Assembly coun y Ta i penal y sh. es o RTA
Che ole Sil e ado Mexico 1.23 0.96
Toyo a Tacoma Mexico 1.22 0.97
Nissan NV200 Mexico 1.22 0.99
Ram 2500/3500 Mexico 1.21 0.94
Ram P oMas e Mexico 1.20 0.92
GMC Sie a Mexico 1.20 0.99
Ram 1500 Mexico 1.19 0.96
GMC Sie a Canada 1.18 0.80
Me cedes-Benz Sp in e Uni ed S a es 1.09 0.73
Che ole Sil e ado Canada 1.05 0.29
Volkswagen Gol Spo Wagen Mexico 1.03 0.96
Che ole C uze Mexico 1.03 0.93
Nissan No e Mexico 1.03 0.86
Volkswagen Gol Mexico 1.03 0.81
GMC Te ain Mexico 1.03 0.98
Toyo a Co olla Canada 1.03 0.85
In ini i QX50 Mexico 1.03 0.93
Buick Regal Canada 1.03 0.97
Dodge Jou ney Mexico 1.03 0.94
Dodge Cha ge Canada 1.03 0.89
No e: Head e al. (2022) p o ides he o mula used o compu e he ca line-le el a i penal y in a
way ha akes in o accoun ma ke sha e changes in esponse o a i changes.
whe he o comply wi h a con en equi emen o 62.5%. Depending on he
assembly cos sha e, his RCR con e s o a pa icula pa s cos s sha e (λR in
he model), which in u n con e s o an implied sha e o egional pa s (χR
in he model). I compliance is oo cos ly ela i e o he a i penal y, hen
he ca line selec s i s uncons ained cos , minimizing No h Ame ican pa s
sha e. The esul is a ec o o pa s cos s sha es eme ging om he simula ed
model. Recognizing ha he model is an app oxima ion, and he da a epo -
ing in AALA is a om pe ec , he simula ion builds in andom measu emen
e o .14 The esul is a simula ion-based dis ibu ion o No h Ame ican pa s
sha es, which we compa e o he ac ual dis ibu ion om he AALA epo s.
We quan i y he disc epancy in e ms o he sum o squa ed de ia ions be ween
model and da a. The algo i hm hen epea s he p ocedu e o a la ge g id o
di e en guesses o he pa ame e s, selec ing he ones ha achie e he bes
i be ween simula ion and obse a ion. Head e al. (2022) p o ides a mo e
o mal desc ip ion o his p ocedu e o es ima ing he model pa ame e s.
The es ima ed pa ame e s allow he dis ibu ion o he simula ed ca lines
o igh ly i he dis ibu ion o No h Ame ican con en epo ed by AALA.
To p o ide ex e nal alida ion o he quan i ied e sion o he model, we
ollow he common p ac ice o conside ing a ea u e in he da a ha was no
pa o he o iginal momen -ma ching exe cise. Fo his pu pose, we compa e
he implied RoO compliance a es (also e e ed o as p e e ence u iliza ion
104 Kei h Head, Thie y Maye , and Ma c Meli z
a es) o au o ade (HS 8703) o hose ha eme ge om he simula ion
based on he calib a ion desc ibed ea lie . As shown in Table 4.1, he ue a e
o p e e ence u iliza ion o US-made ca s en e ing Canada was 97% in 2019
(be o e he change in he egional con en equi emen in 2020). The cali-
b a ed model ob ains a a e o 92%. Thus, ou model is able o closely mi o
he dis ibu ion o No h Ame ican con en a es a he ca line le el and also
ma ch easonably well he RoO sa is ac ion a es obse ed o agg ega e ade
lows wi hin No h Ame ica.
A e ob aining he bes - i alues, we can sol e he model o any po en-
ial RCR. This equi es compu ing how each indi idual ca line will espond
o a s ic e RCR. Depending on hei pa ame e d aws, hey migh inc ease
egional pa s sha es jus enough o ma ch he new equi emen , o hey
migh op in o non-compliance. Based on his decision, he change in cos s
( om inc easing egional con en in esponse o a s ic e ule) o he a i
penal ies ( om op ing no o comply wi h a s ic e ule) will ealloca e ma -
ke sha e owa ds o eign ca lines, as well as hose domes ic ca lines ha we e
no complying be o e he s ic e ule. In compu ing he changes in his s ep,
we ake ad an age o he agg ega ion p ope ies o he cons an elas ici y o
subs i u ion (CES) demand sys em. This p o ides an exac agg ega ion o
he esul ing changes in he p ice index and employmen in he nex sec ion.
6 Quan i ica ion o he impac o RoO changes
In his sec ion, we use ou model, wi h pa ame e s chosen o i he dis ibu-
ion o egional con en by No h Ame ican ca lines, o quan i y he e ec s
o wo ecen changes in RoOs. The i s is he igh ening o RoOs o No h
Ame ican ehicle ade, which was one o he mos salien ea u es o he
USMCA. The second is he applica ion o ules o o igin o UK–EU ade,
equi ed by B i ain’s exi om he cus oms union in he inal B exi deal.
We e alua e changes in he s ic ness o he RoO, as measu ed by changes in
he RCR o he enac ed policies. We also conside al e na i e RCR le els ha
migh ha e been chosen. Fo each policy change, we epo ou comes o g oups
o ca lines based on hei compliance decisions be o e and a e he RoO changes.
Fo example, he i s g oup in each able is he one o ca lines ha comply exac ly
wi h he old RoO bu hen decline o comply wi h he new RoO. The i s nume -
ical column shows he sha e o ca lines in each g oup (in pe cen ). The las ou
columns epo he simula ed changes induced by he change in he RCR. These
ou come a iables comp ise he pe cen age changes in he p ice index, he g oup’s
ma ke sha e, he weigh ed a e age egional pa s sha e, and employmen .
6.1 USMCA
Table 4.3 desc ibes he simula ed ou comes o he USMCA inc ease in he
RCR om 62.5% o 75%. Acco ding o he calib a ed model, jus o e a
hi d o ca lines swi ch om complying uncons ained o complying a he
Unin ended consequences o high egional con en equi emen s 105
Table 4.3 Inc ease in RCR om NAFTA (62.5%) o USMCA (75%)
Compliance s a us unde : Pe cen changes in
NAFTA USMCA Sha e o P ice Mk Pa s Pa s
(RCR = 62.5%) (RCR = 75%) ca lines sha e sha e Emp.
Comply- Non-complian 16.90 0.57 −1.05 −10.40 −11.85
cons ained
Comply- Non-complian 7.10 0.27 −0.16 0.02 −0.40
uncons ained
Comply- Comply- 7.30 1.32 −3.23 20.97 15.53
cons ained cons ained
Comply- Comply- 34.50 0.21 0.00 8.26 8.03
uncons ained cons ained
Non-complian Non-complian 8.30 0.00 0.65 0.00 0.65
Comply- Comply- 25.80 0.00 0.65 0.00 0.65
uncons ained uncons ained
All All 100.00 0.28 −0.20 2.80 2.30
No es: “Sha e o ca lines” e e s o he pe cen age o all domes ic ca lines in he co esponding
s a us uple. “Pa s sha e” is a quan i y-weigh ed a e age o he sha es o pa s om NAFTA o i-
gins ac oss egionally assembled ca lines. “Pa s Emp.” is employmen in pa s manu ac u e o
domes ically assembled ehicles.
minimum equi ed le el o 75%. These ca lines will inc ease hei egional
pa s sha es by abou 8%. The inc ease in a e age cos s o he g oup is jus
one i h o a pe cen . The e is no disce nible educ ion in ma ke sha e o
his g oup, and i s employmen ises by almos he same amoun as i s a e age
pa s sha es. G ea e employmen gains a e eco ded by he 7.3% o ca lines
ha we e jus complying a 62.5% and aise hei egional con en up o 75%.
These ca lines inc ease hei pa s sha es (X) by 21%, sligh ly mo e han he
o e all cos change o 0.75/0.625 − 1 = 20%. The implied ise in employmen
is jus unde 16%. The dampening comes om he 3% ma ke sha e educ ion
o his g oup, which i sel ollows om hei 1.32% ise in hei a e age p ice.
The inc ease in employmen o he cons ained complie s is mos ly o se
by a educ ion in employmen by ca lines ha s op complying once aced wi h
he 75% RCR. The o e all employmen gain is jus 2.3%, much lowe han he
nai e expec a ion o 20% (0.75/0.625 = 1.2) ha would ollow om assuming
ha all ca lines mechanically comply wi h he RoO. While he employmen
gains a e modes , so a e he p ice inc eases aced by consume s: he p ice index
o egionally assembled ca s ises by jus 0.28%. As p edic ed by he con ex
cos cu es shown in Figu e 4.4, he e will be a highe cos o u he ises in
he RCR.
Table 4.4 epo s he esul s o a coun e ac ual ise in he RCR om 75%
o 85% ( he o iginal US ask du ing he USMCA nego ia ions). The las ow
o he igh mos column gi es an in e es ing message o policy. I shows ha
had he US succeeded in nego ia ing an 85% RCR, his would ha e educed
112 Kei h Head, Thie y Maye , and Ma c Meli z
7 Policy implica ions and discussion
The USMCA was welcomed by he chie lobbyis o Canadian au o pa s
manu ac u e s, Fla io Volpe. In an in e iew, he con ended, “Tha deal
[USMCA] ... is he bes single posi i e hi o supplie business ac oss No h
Ame ica in he his o y o he au o business. We hink he e’s going o be 25%
mo e in absolu e olume bough om local supplie s.” In addi ion, he head
o he Mexican au o pa s indus y associa ion p edic ed a en pe cen inc ease
in p oduc ion in Mexico’s pa sec o .19 In con as , he calib a ed e sion o
ou model implies a much smalle e ec o 2.3% (Table 4.3, bo om ow).
Wha is i abou ou model ha implies much lowe employmen gains
om RoO inc eases han nai e calcula ions? The key poin is ha complying
wi h a s ic ule o o igin is a choice. The bene i is p e e en ial a i access o
he o he No h Ame ican ma ke s. Howe e , so long as he US main ains i s
2.5% MFN a i on inished ca s, his is no a huge penal y. Mo eo e , some
Ge man ac o ies in he US may ca e a mo e abou hei sales in o he ma -
ke s – such as China, o example – han hey do abou losing sales in Mexico
o Canada. I b inging ansmission sou cing o No h Ame ica will add o he
cos s and make he ehicle non-compe i i e in China, he i m migh p e e
no o comply on sales o Mexico o Canada and hen sou ce engines om
Eu ope as well i he only eason i had only sou ced locally was o comply
wi h he old NAFTA ules.
The esul s om ou quan i ica ion sugges ha he old NAFTA ule and
he cu en TCA ule a e bo h unde he pa s employmen -maximizing le els.
Howe e , he o iginal T ump adminis a ion demand o 85% would ha e been
coun e -p oduc i e e en om a pu ely p o ec ionis s andpoin . Ou esul s
also sugges he 100% con en equi emen s o ba e ies o EVs a e likely o
lowe employmen while signi ican ly aising he cos s o EV adop ion.
No es
1 Husisian e al. (2018) no e he 85% p oposal in hei o e iew o he USMCA.
2 Felbe may e al. (2019) p esen e idence ha his a gumen applies o mos ules
o o igin.
3 Anas akis (2005) p o ides a book-leng h ea men o his pionee ing egional
ag eemen .
4 “Inside he B exi deal: he ag eemen and he a e ma h” Geo ge Pa ke , Pe e
Fos e , Sam Fleming and Jim B unsden, Financial Times Janua y 21, 2021.
5 “Wha ’s d i ing he EU on ules o o igin?” Jim B unsden, Financial Times Oc o-
be 29, 2020.
6 By con as in he main manu ac u ing coun ies ou side No h Ame ica – Japan,
Ko ea, and Ge many – he USA-USA pai ing is used o jus 1% o ca s.
7 Figu e 4.2 applies he g ea ci cle o mula o calcula e he dis ance be ween engine
(o ansmission) ac o ies and he inal ehicle assembly ac o y. Since engines and
ansmissions a e oo hea y and bulky o ai shipmen , oad, ail, o sea dis ances
would be mo e accu a e. Pas wo k inds high co ela ions be ween g ea ci cle
and ac ual oad dis ances wi hin coun ies. Fo in e con inen al ade, ai ou es
di e ge in a mo e se e e way om sea ou es. Thus, we should expec ha any

Unin ended consequences o high egional con en equi emen s 113
measu emen e o is la ge o long dis ances, bu we see ela i ely li le ade a
dis ances o e 2,000 km.
8 www.nh sa.go /pa -583-ame ican-au omobile-labeling-ac - epo s
9 The pa ame e θ go e ns he a iance o he cos d aws. As θ inc eases, he a iance
dec eases. In he limi , as θ goes o in ini y, he a iance goes o ze o, and he e is
no a ia ion in he cos d aws a ound hei mean.
10 We also se θ = 4.
11 Hypo he ically, i he dis ibu ion o δs is such ha i is domina ed by always-
complie s, hen i is possible o he a e age egional pa -sha e o mono onically
inc ease wi h he RoO. Howe e , we show ha his is no he case o NAFTA.
12 All hese pe cen ages a e cos sha es om he 2019 AALA epo .
13 The a i s China imposes on US expo s do no depend on hei No h Ame ican
con en .
14 Among he sou ces o e o a e he AALA exemp ion o epo ing Mexico con en
i i is below 15%. Addi ional measu emen e o comes om ounding, which he
law pe mi s o he nea es 5%. We also in end o he e o o cap u e de ia ions
om he con inuum assump ion in he model. Since many pa s ha e non-negli-
gible cos sha es, a i m ha in ends o “jus comply” will in ac be obse ed o
o e -comply depending on he sha e o he las pa .
15 The Ve ge, Augus 8, 2022
16 Business Inside Augus 10, 2022
17 Renaul , Peugeo , Sea , and Skoda a e examples o popula b ands in Eu ope ha
a e no o e ed in he US.
18 The e a e some complexi ies in he UK-EU TCA as ega ds elec ic ehicles.
19 Reu e s, Oc obe 1, 2018.
Re e ences
Anas akis, D. (2005). Au o Pac : C ea ing a Bo de less No h Ame ican Au o Indus y
1960–1971. To on o: Uni e si y o To on o P ess.
Felbe may , G., F. Te i, and E. Yalcin (2019). Rules o o igin and he p o i abili y o
ade de lec ion. Jou nal o In e na ional Economics 121, 1032–1048.
G ossman, G. M. (1981). The heo y o domes ic con en p o ec ion and con en
p e e ence. The Qua e ly Jou nal o Economics 96(4), 583–603.
Head, K., T. Maye , and M. Meli z (2022). The La e cu e o ules o o igin.
Manusc ip .
Husisian, G., A. Gomez-S ozzi, and A. Al a ez (2018). In e na ional ade: A new
dawn o No h Ame ican ade. Technical epo , Foley LLP.
I win, D. A. (2017). Clashing O e Comme ce. Chicago: Uni e si y o Chicago P ess.
Ligh hize , R. E. (2020). 2020 T ade Policy Agenda and 2019 Annual Repo . Wash-
ing on, DC: O ice o he US T ade Rep esen a i e.
DOI: 10.4324/9781003415794-5
This chap e has been made a ailable unde a CC-BY-NC-ND 4.0 license.
1 In oduc ion
Due o he apid de elopmen o global alue chains (GVCs) o e he las
h ee decades, he “Made in” label ypically applied o manu ac u ed goods,
a ibu ing hem o a speci ic economy, has become an a chaic symbol, as mos
manu ac u ed p oduc s a e now “Made in he Wo ld” ( hey a e p oduced a
s ages in se e al coun ies, wi h alue added a each s age). The ise o GVCs
has signi ican ly changed he na u e and s uc u e o in e na ional ade and
in es men and b ough conside able bene i s o China and is he majo d i e
behind China’s apid indus ializa ion.
Howe e , he g ow h o GVCs has slowed since 2012, a e a quick eco e y
ollowing he Global Financial C isis (Wo ld T ade O ganiza ion, 2019). Bakas
(2019) poin s o his decline as sugges ing ha he wo ld has en e ed a pe iod o
“slobaliza ion.” The GVC pa icipa ion a e in China has pla eaued since 2007
and was below he wo ld a e age in 2019 (Asian De elopmen Bank, 2021).
The ade wa be ween he US and China in 2018 u he wo sened China’s
in e na ional en i onmen . The US go e nmen i s a emp ed o educe US
impo s om China h ough highe a i s, hen p oceeded o impose s ic
expo con ols o cu o key high- ech componen s supply o Chinese high-
ech i ms such as he ban o semiconduc o sales o Huawei. Fo some hawkish
membe s o he US Cong ess, undoing 40 yea s o e e -close economic ela-
ions wi h China and olling back US eliance on Chinese ac o ies was always
one o hei poli ical objec i es. An “Economic P ospe i y Ne wo k” o like-
minded coun ies, a concep ini ially p oposed by he T ump adminis a ion
and inhe i ed and s eng hened by he Biden adminis a ion, aims o con ince
Wes e n i ms o ex ica e hemsel es om China and ins ead pa ne wi h i ms
headqua e ed wi hin membe coun ies o he ne wo k based on he common
goal o educing economic dependence on Beijing (“ iend sho ing”). Eco-
nomic na ionalism and a ious p o ec i e measu es a e on he ise.
China has made s a egic mo es o p epa e i sel o his less a o able in e -
na ional economic en i onmen . Beijing has announced a dual-ci cula ion
economic s a egy ha emphasizes domes ic consump ion as he majo ehi-
cle o economic de elopmen . Since Beijing launched a campaign o de elop
5 LCR Policies in China and hei
impac s on domes ic alue added
in expo s
Kun Cai and Zhi Wang
LCR policies in China 115
mo e ad anced echnologies a home and ely less on he Uni ed S a es and
o he Wes e n supplie s in 2012, i has been pu suing i s own o m o “made
in China” o mo e han a decade. Achie ing echnological independence
om he Wes , especially he Uni ed S a es, has been a s a ed goal o he
Chinese go e nmen and ea i med by he cu en leade .1 Beijing is pu suing
wo key objec i es: (1) elimina ing i s dependence on o eign coun ies o
c i ical echnologies and p oduc s and (2) encou aging domes ic indigenous
i ms o bols e hei own capaci y o inno a ion in o de o become lead-
e s in ad anced echnologies. To achie e such key objec i es, a ious go -
e nmen agencies a di e en le els in China p oposed and implemen ed a
se ies o indus ial policies, including some implici local con en equi emen s
(LCRs), o encou age domes ic p oduc ion and inno a ion.
This chap e e iews hese policies and measu es he changing end o
domes ic con en in China’s expo s om 2007 o 2017 based on de ailed ade
s a is ics om he China Cus om adminis a ion and he mos ecen na ional
inpu -ou pu ables (IOTs) published by Na ional Bu eau o S a is ics (NBS). We
also seek o assess he implica ions o a ious implici LCR measu es p oposed in
China on domes ic con en in Chinese expo s. The chap e is o ganized as ol-
lows. Sec ion 2 e iews he majo indus ial policies and implici LCR measu es
in ecen yea s p oposed by Chinese cen al and local go e nmen s and by majo
manu ac u ing indus ies. Following Koopman e al. (2012), Sec ion 3 ou lines
he concep ual amewo k o es ima ing domes ic alue added (DVA) in a coun-
y’s expo s when p ocessing expo s a e impo an . We ex end hei me hodol-
ogy o decompose p oduc ion ac i i ies in o pu e domes ic, adi ional ade and
GVC ac i i ies a he coun y/sec o le el based on na ional IOTs. Sec ion 4 p e-
sen s he majo empi ical esul s and uses hem o e alua e he impac o China’s
LCR measu es on domes ic con en in Chinese expo s. We ind no empi ical
e idence ha hose policy measu es implemen ed by he Chinese go e nmen in
ecen yea s ha e played any signi ican ole in p omo ing domes ic con en in i s
expo s, a leas a he agg ega e le el. Sec ion 5 concludes.
2 Recen local con en equi emen policy de elopmen
inChina2
A local con en equi emen is a measu e ha suppo s he use o local inpu s in
he p oduc ion o goods o se ices as a p econdi ion o gaining ma ke access
o ob aining inancial incen i es. Coun ies ha ha e aken hese measu es hope
o compel o eign companies o sou ce om local i ms o p omo e he de el-
opmen o hei own indus ies. This suppo o local inpu s incen i izes i ms
o selec hei supplie s based on hei na ionali y a he han quali y and cos .
The scope o LCR measu es is no clea ly de ined. Hes e meye and
Nielsen (2014) classi ied LCR policies in o h ee ca ego ies: licensing, go -
e nmen p ocu emen and inancial incen i es. Hu baue e al. (2013) belie e
LCRs can ake many o ms, including p ice p e e ences awa ded o domes ic
i ms ha bid on go e nmen p ocu emen con ac s, manda o y minimum
116 Kun Cai and Zhi Wang
pe cen ages equi ed o domes ic goods and se ices used in p oduc ion,
impo licensing p ocedu es designed o discou age o eign supplie s and dis-
c e iona y guidelines ha bo h encou age domes ic i ms and discou age o -
eign i ms. They iden i ied 117 LCR measu es implemen ed ac oss he wo ld
since he 2008 inancial c isis and poin ed ou he dis inc i e cha ac e is ics
ha LCRs ha e compa ed wi h o he ade policies. Following hei wo k,
S one e al. (2015) p o ide a quan i a i e analysis o he localiza ion ba ie s
o ade. They g oup a ious LCR measu es in wo dimensions, which a e he
a ge ed ma ke and iden i ied bene i s. The a ge ed ma ke s include inpu s,
owne ship, labo , go e nmen p ocu emen and da a. The bene i s include
ma ke access, p ice p e e ence, ax policies, go e nmen unds and domes ic
b anding schemes. The OECD had deemed ecen “Made in XX” o “Buy
XX” p og ams ini ia ed by some coun ies as localiza ion ba ie s o ade.3
The LCR measu es applied by China be o e i s WTO accession we e explici .
Fo ins ance, p e e en ial a i and ax incen i es we e p o ided based on he
pe cen age o local inpu s, and o eign en e p ises we e o ced o ollow he
manda o y echnology ans e equi emen s. A e 2001, explici LCR pe cen -
ages o goods o se ices we e g adually li ed. Howe e , implici localiza ion
ade ba ie s ing ained in he implemen a ion o indus ial policies eme ged.
These implici LCRs aim o p omo e he inno a i e capaci y o China and o
cul i a e indigenous domes ic companies. On he su ace, hese policies ea
p oduce s equally ega dless o na ionali y, while in p ac ice, o eign p oduce s
may be encou aged o conduc localiza ion s a egies olun a ily, o i may be
he case ha only indigenous i ms uly bene i om hese p e e en ial policies.
Due o hei opaci y and co e ness, hese LCRs can be di icul o iden i y.
The he e ogenei y o localiza ion policies is p ominen ac oss sec o s. A
sec o -by-sec o app oach is aken o p esen he localiza ion policies in China,
which ake he o ms o ma ke access, subsidies, licensing and go e nmen
p ocu emen . We ocus on he au omobile, in eg a ed ci cui s (IC), elecom-
munica ions, pha maceu ical and medical equipmen indus ies because hey
a e o i al impo ance o China’s indus ial sys em, and o ha in Made in
China 2025, speci ic a ge s o he localiza ion a e o ma ke sha e a e se
o h o many c i ical ma e ials, p oduc s and p ocessing equipmen in hese
indus ies. By looking a he LCR policies in hese sec o s, we hope o shed
some ligh on he oles hey played in p oduc ion ac i i ies and hei e ec on
he changing end o domes ic con en s in Chinese expo s.
2.1 Au o indus y
The au o indus y is one o he pilla indus ies in he Chinese economy. The
alue added o au o ehicle and au o pa s p oduc ion accoun s o app oxi-
ma ely 2% o China’s GDP.4 In 2021, bo h he domes ic sales and p oduc ion
o ca s in China eached 26 million, which means ha app oxima ely 32.5%
o o al wo ld au o p oduc ion is conduc ed in China and 31.8% o global
au omobile sales a e conduc ed in China, making China he la ges ca manu-
ac u ing and consump ion coun y.
LCR policies in China 117
Au o manu ac u ing in China elies hea ily on locally made componen s.
Figu e 5.1 p esen s he alue o he impo ed au o pa s pe ehicle and
-
5,000
10,000
15,000
20,000
25,000
30,000
-
2,000
4,000
6,000
8,000
10,000
12,000
14,000
16,000
Numbe s o ca s p oduced ( igh axis)Impo ed au o pa s pe ca (le axis)
housands o ca s
Impo ed au o pa s pe ca p oduced (dolla s)
Figu e 5.1 Impo ed Au omobile Pa s and Au omobile P oduc ion o Majo P oduc-
ing Coun ies in 2020
No es: Au o p oduc ion includes bo h comme cial and passenge ehicles. Impo ed au o pa s pe
ca a e calcula ed as o al impo ed au o pa s di ided by he numbe o ca s p oduced in a coun y.
Sou ces: Au ho s’ calcula ion based on p oduc ion s a is ics om he In e na ional O ganiza ion
o Mo o Vehicle Manu ac u e s (OICA). h ps://www.oica.ne /p oduc ion-s a is ics/. UN
Com ade Da abase. h ps://com ade.un.o g/da a/.
Table 5.1 MNEs and Thei Join Ven u es in China
MNEs Coun y Join en u es Yea o Sha eholdings
in China es ablishmen
Volkswagen Ge many SAIC 1984 50% by Shanghai Au o
Volkswagen Indus y Co. (SAIC),
40% by Volkswagen, 10%
by Volkswagen China
FAW 1991 60% by China FAW G oup
Volkswagen Co., (FAW), 20% by
Volkswagen, 10% by Audi
AG, 10% by VW China
JAC 2017 50% by Volkswagen, 50%
Volkswagen by JAC G oup (JAC)
In 2020, VW b ough in
50% s akes o JAC G oup
and inc eased he sha e in
JAC VW o75%.
Daimle Beijing 1983 51% by Beijing Au omobile
Benz Co., LTD (BAIC), 49%
by Daimle
Fo on 2011 50% by Beijing Fo on, 50%
Daimle by Daimle
(Con inued)

118 Kun Cai and Zhi Wang
Table 5.1 (Con inued)
MNEs Coun y Join en u es Yea o Sha eholdings
in China es ablishmen
Fujian 2007 35% by Beijing Au o, 15%
Daimle by Fujian Au o, 50% by
Daimle
BMW BMW 2003 50% by BMW, 50% by
B illiance Au o. In 2022,
BMW inc eased s akes o
75%.
Hyundai Ko ea Beijing 2002 50% by Beijing Au o, 50%
Hyundai by Hyundai
Toyo a Japan FAW 2003 35% by FAW, 50% by
Toyo a Toyo a, 15% by Tianjin
FAW Xiali Co. LTD
The cu en a io is 38%
by FAW, 32% by Toyo a,
30% by Tianjin FAW
Toyo a (TFTM).
GAC 2004 50% by Guangzhou
Toyo a Au omobile G oup Co.
LTD (GAC), 30.5% by
Toyo a, 19.5% by Toyo a
China
Honda Guangqi 1998 50% by GAC, 40% by
Honda Honda, 10% by Honda
Technical Resea ch
Indus y (China)
In es men
Dong eng 2003 50% by Dong eng Mo o
Honda G oup Co. LTD (DFG),
40% by Honda, 10% by
Honda China
Nissan Dong eng 2003 50% by DFG, 50% by
Nissan G oup o China
GM US SAIC GM 1997 50% by SAIC, 50% by GM
SAIC-GM- 2002 50.1% by SAIC, 44% by
Wuling GM, 5.9% by Guangxi
Au o ( o me ly called
Wuling)
SGM 2004 25% by SAIC, 25% by GM
No som China, 50% by Shanghai
GM
Fo d Changan 2001 50% by Changan, 35% by
Fo d Fo d Asia Paci ic Mo o
Holdings LTD, 15% by
Fo d China
JMC Fo d 1997 41% by Nanchang Jiangling
In es men Co., LTD,
32% by Fo d, o he s by
public sha eholde s
Tesla – 2018 100% by Tesla
Sou ce: Collec ed om o icial websi es o he companies in he able and he en e p ises big da a
pla o m ope a ed by Baidu, h ps://aiqicha.baidu.com/.
LCR policies in China 119
he numbe o ca s p oduced in majo ca -manu ac u ing coun ies. Each ca
p oduced in China used app oxima ely $1,021 o impo ed componen s, a
less han Canada ($14,564), F ance ($9,792), Ge many ($9,200) and he
US ($6,806). O he eme ging economies, such as Mexico ($6,901), Russia
($5,330) and B azil ($2,615), also use a highe alue o impo ed pa s han
China. Among he op 15 au o-manu ac u ing coun ies, Japan has he low-
es alue sou ced om o e seas, app oxima ely $778. The impo ed alue pe
ca in Ko ea ($1,154) and India ($961) is oughly he same as ha in China.
One eason ha au o componen s and pa s emain la gely locally sou ced
is ela ed o he au o ma ke access policy in China. In 1994, he Au omo i e
Indus ial Policy was published, which equi es o eign ca make s o o m a
join en u e (JV) o gain access o he Chinese ma ke . An equi y cap o 50%
is also se o o eign sha eholde s. Table 5.1 lis s he join en u es and he
sha eholding o mul ina ional en e p ises (MNEs) in China. The equi emen s
o JVs and he equi y caps ha e g adually elaxed in ecen yea s. BMW aised
i s equi y sha e o 75%, becoming he i s o eign company o ha e a majo i y
sha e in au o JVs in China. In 2019, Tesla became he i s o eign au omobile
company wi h a wholly owned Giga ac o y in China, as shown in Table 5.1.
In ecen yea s, iscal subsidies ha e been p o ided o new-ene gy ehicle
(NEV) p oduce s, which is ano he one o China’s e o s o p omo e domes-
ic au o p oduc ion. The subsidies a e no exclusi ely p o ided o indigenous
manu ac u e s, bu domes ic p oduce s seem o ha e bene i ed he mos .
Table 5.2 shows he op 20 companies ha ecei ed he mos iscal subsidies
om 2017 o 2020. Each yea , hey collec i ely accoun o app oxima ely
90% o o al subsidies o NEVs.
In 2021, China published he P o isions on he Secu i y Regula ion o Au o-
mobile Da a, laying down he ules ha all impo an da a mus be s o ed
wi hin he geog aphic bounda ies o China. The so-called impo an da a
include da a o pedes ian low, a ic s eam, da a o elec ic au omobile
cha ging ne wo ks, ideos and images o ca pla es and d i e s’ aces. Any
c oss-bo de ans e o such da a mus be examined by local au ho i ies.
Complying wi h he egula ions on au o da a, many MNEs, such as Tesla,
ha e se up da a cen e s5 in China o acili a e he localiza ion o da a s o age
and p ocessing.
2.2 In eg a ed ci cui (IC) indus y
The goal o IC-p omo ing schemes in China is o cul i a e comp ehensi e
and ma u e domes ic IC supply chains, co e ing all p oduc ion s ages and
achie ing chip sel -su iciency. S a is ics show ha he global p oduc ion sha e
o China’s IC indus y was only app oxima ely 5% in 2020, a behind ha
o he US (47%), Ko ea (20%), Japan (10%) and Eu ope (10%). Addi ionally,
China is mo e concen a ed on ela i ely capi al-in ensi e ac i i ies, including
ma e ials, wa e ab ica ion, assembly, packing and es ing. The alue added
cap u ed by China accoun s o app oxima ely 9% o he o al semiconduc o
alue chains.6
120 Kun Cai and Zhi Wang
Subsidies
,
million)
1448.91
776.33
549.93
496.38
275.74
232.18
190.75
187.98
170.9
136.93
135.53
122.72
118.87
105.32
100.47
82.22
79.06
75.06
68.46
63.56
(CNY
5890.34
e
_99ae13aa81d04a358
2017
Company
BAIC Mo o
DFAC
Geely Au o
y Mo o Che
Zhong ong Bus
Dayun Au o
Guang ong Manu ac u
Changan Au o
BYD Au o
Sunlong Bus
Golden D agon Bus
V
SAIC Mo o
Jiangnan Au o
CRRC Times E
Shuchi Bus
Fo on Au o
o al
Yaxing Bue
Joylong Au o
y Au oic o V
Xinchu eng Au o
/2021/a
T
Subsidies
,
million)
696.06
636.35
468.71
369.93
341.56
340.52
253.3
(CNY
465.22
245.46
227.23
220.93
215.56
187.71
152.81
112.98
109.16
92.38
91.32
83.63
63.57
6075.8
.cn/zwgk/wjgs/a
eManu ac u
V
mii .go
way Mo o s
all Mo o
://wap.
2018
Company
DFAC
BAIC Mo o
Geely Au o
BYD Au o
JMC Mo o
y Mo o Che
SAIC Mo o
Changan Au o
Jiangnan Au o
Guang ong
JAC Mo o
Golden D agon Bus
inne
Sunlong Bus
BMW
CRRC Times E
o al
W
Yu ong Bus
ea WG
Zhong ong Bus
Fo on Au o
. h ps
T
echnology
Subsidies
, (CNY
million)
1134.74
966.35
890.39
801.73
783.86
741.16
599.59
343.26
315.32
275.76
267.54
266.12
175.32
166.04
161.01
159.81
142.51
128.95
121.79
91.06
9578.99
ma ion T
, 2017–2020
BAIC Mo o
Golden D agon Bus
V
Geely Au o
SAIC Mo o
G
CRRC Times E
y Mo o
BYD Au o
Sunlong Bus
-VW
2019
Company
u ong BusY
DFAC
Zhong ong Bus
Changan Au o
all Mo o ea W
Che
SAIC
Fo on Au o
BMW
Anhui Ankai Au o
way Mo o s
y and In o
V Subsidies Bene icia ies
inne
axing Bus
o al
W
GAC Mo o
Y
T
y o Indus
Subsidies
,
million)
2123.37
2046.87
951.32
830.56
550.41
539.57
410.86
379.02
338.97
303.97
300.87
248.74
238.44
195.62
166.74
150.89
110.42
105.16
80.59
66.6
(CNY
10537.17
om he Minis
op 20 NE
.h ml
T
all Mo o
oyo a
5.2
esla Shanghai
GAC Mo o
y Mo o
Based on da a
2020
Company
BYD Au o
ea W
T
G
Che
JAC Mo o
SAIC Mo o
Yu ong Bus
Geely Au o
BAIC Mo o
SGMW Au o
Xiaopeng
WM Mo o
Haima Mo o
T
able
-
T
GAC
Golden D agon Bus
Hozonau o
Zhong ong Bus
DFAC
Lixiang Au o
:
o al
ce
T
Sou
e0443700 e100ce
LCR policies in China 121
Table 5.3 In es ed Fi ms and he Sha eholdings o Na ional IC Fund Phase I (accessed on Ap il 24, 2022)
Segmen ed
p oduc ion
Fi ms Sha eholdings
o ICF
Subsc ibed capi al
(CNY million)
Wa e ab ica ion Semiconduc o Manu ac u ing No h China (Beijing) Co po a ion (SMNC) 32.00 10291.20*
Semiconduc o Manu ac u ing Sou h China Co po a ion (SMSC) 14.56 6341.55*
Huahong Wuxi 29.00 3698.40*
Yang ze Memo y 24.09 13558.42
Shanghai Huali In eg a ed Ci cui Co po a ion 39.19 11600
Ningbo Semiconduc o In e na ional Co po a ion (NSI) 13.55 60
Beijing Yan Dong Mic oelec onic Technology Co., L d. (YDME) 18.84 113
Silex Mi oSys ems 30.00 600
Compound
semiconduc o
Sanan Op oelec onics 6.47 -
Beijing Cen u y Golden Ligh Semiconduc o Co., L d. (CENGOL) 9.61 29.57
Packing and
es ing
JCET 13.31 -
Tong u Mic oelec onics Co., L d. (TFME) 15.13 -
JCET Shaoxing 26.00 1300
Specialized
equipmen and
key componen s
NAURA Technology G oup Co., L d. (NAURA) 7.48 -
Hangzhou Changchuan Tech (CCTECH) 6.76 -
Shanghai P ecision Measu emen Semiconduc o Technology, Inc. (PMISH) 7.30 100
RSIC scien i ic ins umen (Shanghai) Co., L d. 8.78 37.58
Sky Technology De elopmen Co., L d. Chinese Academy o Sciences (SKY) 19.73 -
Na ional Silicon Indus y G oup (NSIG) 20.84 -
Ma e ials Jiangsu Xinhua Semiconduc o Ma e ial Technology (Xinhua Semiconduc o ) 23.56 306.29
BDS a Na iga ion 8.57 -
Chip design Unisoc (Shanghai) Technologies (Unisoc) 13.96 705.88
Shanghai AisinoChip Elec onics Technology Co., L d. (AisinoChip) 21.06 13.5
Empy ean Technology 11.1 48.19
Telink Mic oelec omics Shanghai (Telink) 11.94 21.49
No e: * indica es he o iginal igu es a e in USD and he numbe s a e con e ed using he o eign exchange a e o 6.70 RMB/USD.
— indica es da a no a ailable.
Sou ce: Based on he inancial s a emen o public companies; he en e p ise big da a pla o m ope a ed by Baidu, h ps://aiqicha.baidu.com/.
128 Kun Cai and Zhi Wang
ac i i ies o adi ional ade. The las pa is alue added embodied in expo s
o in e media e goods and se ices, which will be used in p oduc ion ac i i ies
in o he coun ies and in ol es c oss-coun y p oduc ion sha ing, so i can
be de ined as GVC ac i i ies. I includes bo h he simple GVC (DVA c osses
bo de s only once) and complex GVC (DVA c osses bo de s mul iple imes)
ac i i ies de ined in Wang e al. (2017).
The DVA decomposi ion esul s om Equa ion (5) will be he same based
on a na ional IO able as using a global ICIO able i he e is no need o
sepa a e he pa o DVA ha is i s expo ed bu e en ually e u ns home.
I is also s aigh o wa d and compu es domes ic con en s in a coun y’s
g oss expo s as he sum o he second and he hi d pa o Equa ion (5).
I is impo an o no e ha his me hod is limi ed because i elies on na ional
IOTs. Al hough i can decompose p oduc ion in o pu e domes ic, adi ional
ade and GVC ac i i ies and compu e DVA in expo s, including di ec and
indi ec alue-added expo s ia ups eam o downs eam sec o s, i canno es i-
ma e DVA i s expo ed bu e u ned home ia impo s, indi ec expo s o
DVA ia hi d coun ies, dis inguish simple and complex GVC ac i i ies, decom-
pose bila e al ade lows, measu e double coun ing due o in e media e goods
ade and ace o eign alue added and/o e ical specializa ion by coun y
sou ces. These quan i a i e measu es mus be ob ained om ICIO ables.
3.2 Domes ic con en in expo s – when p ocessing expo s12 a e impo an
The p oduc ion o p ocessing expo s o en has a e y di e en in ensi y in he
use o impo ed inpu s han ha in domes ic inal sales and no mal expo s. To
e lec such he e ogenei y, one needs o keep ack sepa a ely o he IO coe -
icien s o he p ocessing expo s and hose o domes ic inal sales and no mal
expo s. The ex ended IO able wi h a sepa a e accoun o p ocessing expo s
is ep esen ed by Table 5.7.
In such an ex ended IO able, domes ic p oduc ion has been sepa a ed
in o wo pa s: (1) p oduc ion o domes ic inal demand and no mal expo s
ep esen ed by supe sc ip D and (2) p oduc ion o p ocessing expo s ep e-
sen ed by supe sc ip P. Ma hema ically, he model can be speci ied as Equa-
ions (6) and (7),
IA A
I
XE
E
YE
E
DD DP P
P
DP
P
dd
g
g
g
j
l
|
|
d
g
g
g
j
l
|
|=d
g
g
g
j
l
|
|
0 (6)
AXEAEY M
MD PMPP M
-
90
=== (7)
whe e
A
DD =
Z
DD (
XE
P
−−
)
1
and
A
DP =
Z
DP (EP)−
1
deno e he inpu coe -
icien ma ix o he p oduc ion o domes ic use and no mal expo s and he
inpu coe icien ma ix o he p oduc ion o p ocessing expo s, espec i ely.

LCR policies in China 129
Table 5.7 Inpu -Ou pu Table wi h Sepa a e P oduc ion Accoun s o P ocessing T ade
  In e media e use 
  P oduc ion P oduc ion Final use G oss
o domes ic o p ocessing (C + I + ou pu o
use &no mal expo s G + E) impo s
expo s
DIM 1,2, ..., N 1,2, ..., N 1 1
1
P oduc ion o .
domes ic use .ZDD ZDP YE
DP
−XE−P
& no mal .
Domes ic expo s (D) N
in e media e 1
inpu s .
P ocessing . 0 0 EPEP
expo s (P) .
N
1
.
In e media e inpu s om .ZMD ZMP YMM
impo s .
N
Value added 1 VDVP
G oss ou pu 1 XE−PEP
Sou ce: Adop ed om Koopman e al. (2012).
The analy ical solu ion o his ex ended IO model is
g k b1
XEbPgIA
bb
DD ADP kgYE
DP
bk
h =h h (8)
hEP g0IP
g h hh
h
gE
h
I s Leon ie in e se can be compu ed as ollows:
s1
dIAss
DD ADP gdLL
DD DP g
L=g h =gh (9)
g
0IPD
g
h
gLL
PP g
h
ds
IAs
()
s1
DDDAPg
IAsD
g
()
1
DD
=h
gh
0Ig
IA IA
DD D
=s
()
s
s1DDDP
A
I
()
d
g
g
g
g
h
h
s1
0
LIA A
I
DD DP DD DP
PD PP
=ss
d
g
g
g
g
h
h=d
g
g
g
g
h
h
s
0
1
130 Kun Cai and Zhi Wang
Subs i u ing Equa ion (9) in o Equa ion (8), we ha e:
XE IA YE
AA
E
PDDDPD
DDPp
-=--
()
+-
--
() ()
11
1 (10)
Subs i u ing Equa ion (10) in o Equa ion (7), he o al demand o impo ed
in e media e inpu s can be compu ed as
MY AIAY
EA AAEAE
MMDDDD
PMDDDDPP MP p
-=
--
()
+- +
--
()
()
11
1
(11)
MY AIAYEA
AAEAE
MMDDDDPM
DDDDPP MP p
-= --
()
+-
+
--
()
()
11
1
I has h ee componen s: he i s e m is o al impo ed con en in inal
domes ic sales and no mal expo s, and he second and he hi d e ms a e
indi ec and di ec impo ed con en in p ocessing expo s, espec i ely.
DVA (GDP) a he indus y le el can be compu ed as:
DVA
DVAAA IA IA A
I
AIA
D
P
T
V
DV
PDD DD DP
V
DDD
s
d
s
d
s
d
g
g
j
l
|
|
d
dd11
0
s
d
s
+
d
d
g
g
j
l
|
|
d
d
1
1
AIAAA
YE
E
V
DDDDPV
P
T
DP
P
(12)
whe eA
D
and
A
P a e n × 1 ec o s ha deno e he di ec alue-added coe i-
cien ec o s o domes ic sale/no mal expo s and p ocessing expo s, espec-
i ely. Equa ion (12) can be ew i en as:
DVAGDP AIAY
EA AAEAE
V
DDDDP
V
DDDDPP V
PP
==
--
()
+- +
--
()
()
11
1
(13)
DVAGDP AIAYEA
AAEAE
V
DDDDP
V
DDDDPP V
PP
== --
()
+- +
--
()
()
11
1
I also has h ee componen s: he i s e m is DVA in inal domes ic sale and
no mal expo s, and he second and he hi d e ms a e indi ec and di ec
DVA in p ocessing expo s, espec i ely. No ice ha Y
D = Y + Y e + Y ei; i we
inse i in o Equa ion (13) and ea ange, we can decompose each indus y’s
domes ic alue added, o GDP, in o he ollowing ou pa s:
DVAGDP AIAYAIAYE
AIAY
V
DD V
DDe P
V
DDei
== -+
--
()
+- -
--
-
() )(
()
11
1EEA AAAE
Pi V
DDDDPV
PP
()
+- +
()
-
()11 (14)
LCR policies in China 131
whe e
E
P and
E
Pi a e p ocessing expo s o inal and in e media e p oduc s,
espec i ely, and
EE
P Pi
+
=
E
P by de ini ion. Equa ion (14) is an ex ension
o he GDP decomposi ion Equa ion (5) wi h an addi ional ou h e m ha
is gene a ed om p ocessing expo p oduc ion.
4 Da a sou ces and es ima ion esul s
To es ima e he ex ended IO able ha accoun s o p ocessing expo s sepa-
a ely, we closely ollow he Quad a ic P og aming p ocedu es desc ibed in Sec-
ion 2.3 o Koopman e al. (2012).13 The pu pose o hese p ocedu es is o
minimize a quad a ic penal y unc ion subjec o a se ies o accoun ing iden i ies
and adding-up cons ain s based on o icial s a is ics. S anda d na ional IOTs
a e used o de e mine sec o -le el p oduc ion and ade, and in o ma ion om
ade s a is ics is used o de e mine he ela i e p opo ion o p ocessing and
no mal expo s wi hin each sec o ; hus, all a ailable da a a e used o spli he
na ional economy in o p ocessing and non-p ocessing blocks, each wi h i s own
IO s uc u e.
A e desc ibing he da a sou ces, we epo he es ima ion esul s o
China’s sha e o domes ic con en in i s p oduc ion and g oss expo s a he
agg ega e le el, by i m owne ship, by majo des ina ion coun ies and by
manu ac u ing indus ies.
4.1 Da a
In e -indus y ansac ion and di ec alue-added da a a e om China’s
2007, 2012, and 2017 benchma k IOTs published by he NBS o China,
while de ailed ade da a om 2007 o 2017 a e om he Gene al Cus oms
Adminis a ion o China. The ade s a is ics a e i s agg ega ed om he
8-digi HS le el o China’s IO indus ies. We pa i ion bo h impo s and
expo s in a gi en commodi y classi ica ion in o di e en pa s based on
he dis inc ion be ween p ocessing and no mal ade in he ade s a is ics
and on he UN BEC classi ica ion. A summa y o such pa i ions as a pe -
cen age o China’s o al expo s and impo s along wi h he sha e o p o-
cessing expo s du ing 2007–2017 is epo ed in Table 5.8 and Table 5.9.
The UNBEC classi ies each HS 6-digi p oduc in o one o h ee ca ego ies:
in e media e inpu s, capi al goods and consump ion p oduc s. In Table 5.8,
we u he decompose he i s wo ca ego ies in o wo subca ego ies: p o-
cessing impo s used o he p oduc ion o p ocessing expo s ha canno
be sold o domes ic use s by egula ion and non-p ocessing impo s used
o domes ic sale and no mal expo s. Columns (1) o (5) in Table 5.8 sum
o 100%; Columns (1) o (3) and (4) o (6) in Table 5.9 sum o 100%.14
These da a a e impo an pa ame e s in ou op imiza ion model and he key
in o ma ion o unde s and ou es ima es o domes ic and impo ed con en
sha es in Chinese expo s, especially hei c oss-sec o he e ogenei ies and
changing ends o e ime.
132 Kun Cai and Zhi Wang
4.2 Domes ic con en s in agg ega e expo s
Table 5.10 p esen s he esul s o he decomposi ion o agg ega e o eign
and domes ic alue-added sha es in 2007, 2012 and 2017. The es ima ed
agg ega e DVA sha e in China’s o al g oss expo s was 64.2% in 2007,
65.2% in 2012 and 69.8% in 2017. Such numbe s o me chandise expo s
Table 5.8 Majo Impo s Sha e Pa ame e s Used in Domes ic Con en Es ima ion,
2007–2017
Yea Impo ed in e media es % Impo ed capi al goods % Impo ed P ocessing
inal expo s as %
Fo p ocessing Fo no mal Fo p ocessing Fo no mal consump ion o o al
expo s use expo s use % expo s
(1) (2) (3) (4) (5) (6)
2007 37.8 47.3 8.3 4.0 2.6 51.6
2008 32.3 53.4 8.1 3.3 2.9 48.1
2009 30.8 53.9 9.8 2.1 3.4 49.8
2010 29.0 55.1 10.4 1.8 3.7 48.0
2011 25.6 58.4 10.3 1.6 4.1 45.2
2012 24.9 59.4 9.7 1.4 4.6 44.1
2013 24.9 59.7 9.6 0.9 4.9 41.6
2014 26.5 57.1 10.0 0.9 5.6 39.6
2015 27.1 55.6 9.9 0.9 6.5 36.8
2016 25.3 56.2 10.1 0.7 7.7 35.7
2017 23.6 58.5 9.6 0.7 7.6 35.1
Sou ce: Au ho s’ calcula ions based on China Cus om ade s a is ics and Uni ed Na ions B oad
Economic Ca ego ies (UNBEC) classi ica ion scheme.
Table 5.9 Majo Expo s Sha e Pa ame e s Used in P oduc ion Decomposi ion
2007–2017
Yea No mal expo s % P ocessing expo s %
 Sha e o Sha e o Sha e o Sha e o Sha e o Sha e o
in e media es capi al consump ion in e media es capi al consump ion
goods goods goods goods
 (1) (2) (3) (4) (5) (6)
2007 49.92 13.39 36.69 33.28 34.26 32.46
2008 53.06 14.28 32.66 32.68 35.12 32.21
2009 46.80 15.23 37.97 31.20 36.34 32.46
2010 48.39 14.78 36.83 31.92 37.21 30.87
2011 49.38 15.01 35.61 31.82 37.34 30.84
2012 47.72 15.86 36.42 30.45 38.19 31.36
2013 46.95 15.86 37.19 31.75 36.57 31.67
2014 47.27 16.27 36.46 31.63 34.93 33.44
2015 46.87 16.87 36.26 32.77 36.37 30.86
2016 46.72 16.74 36.54 33.59 36.01 30.40
2017 47.86 16.59 35.54 33.85 35.86 30.29
Sou ce: Au ho s’ calcula ions based on China Cus om ade s a is ics and Uni ed Na ions B oad
Economic Ca ego ies (UNBEC) classi ica ion scheme.
LCR policies in China 133
we e 60.1%, 59.5% and 64.4%, espec i ely. Fo manu ac u ing p oduc s
only, hese es ima ed sha es a e lowe om 59.2% in 2007 o 63.5% in
2017. In gene al, he es ima ed di ec domes ic alue-added sha es a e
less han one- hi d o he o al domes ic alue-added sha es. Howe e ,
he di e ence be ween he di ec o eign alue-added sha e and he es i-
ma ed indi ec o eign alue-added sha e was ela i ely small, indica ing
ha mos o he o eign con en comes di ec ly om impo ed o eign
inpu s and gene a es much less indi ec alue added compa ed o domes ic
alue added.
I is in e es ing ha he DVA sha es in no mal and p ocessing expo s
end in opposi e di ec ions: he DVA sha e in China’s no mal manu ac-
u ing expo s inc eased om 82.9% in 2007 o 84.7% in 2017, bu his
sha e declined in p ocessing expo s om 36.6% in 2007 o 27.7% in 2017,
which is comple ely di e en om he p e ious decades (mo e domes i-
cally p oduced inpu s we e used in China’s p ocessing expo s be ween
1997 and 2007, he DVA “sha e inc eased om 20.7% in 1997 o 37.0%
in 2007, up by mo e han 16 pe cen age poin s” (Koopman e al., 2012)).
This indica es ha he inc ease in he DVA sha e be ween 2007 and 2017
in China’s o al expo s (app oxima ely 5.3 pe cen age poin s inc ease) was
mainly d i en by he decline in p ocessing expo s in China’s o al me -
chandise expo s (dec eased om 51.6% in 2007 o 35.1% in 2017) and
Table 5.10 Domes ic and Fo eign Values Added: P ocessing s No mal Expo s (in
Pe cen o To al Expo s)
No mal expo s P ocessing expo s Weigh ed sum
2007 2012 2017 2007 2012 2017 2007 2012 2017
To al expo s (including
se ice sec o s)
To al o eign alue added 15.3 14.6 12.7 63.0 70.0 71.9 35.8 34.8 30.3
Di ec o eign alue added 4.9 4.7 4.8 58.0 66.4 69.5 27.7 27.2 24.0
To al domes ic alue added 84.7 85.4 87.3 37.0 30.1 28.1 64.2 65.2 69.8
Di ec domes ic alue added 28.6 30.4 30.4 9.5 8.9 9.3 20.4 22.5 24.1
All me chandise
To al o eign alue added 16.7 16.6 14.9 63.0 70.0 72.0 39.9 40.5 35.6
Di ec o eign alue added 5.6 5.6 5.9 58.1 66.5 69.7 31.9 32.8 29.0
To al domes ic alue added 83.3 83.4 85.1 37.0 30.0 28.0 60.1 59.5 64.4
Di ec domes ic alue added 23.4 22.1 22.3 9.4 8.9 9.2 16.4 16.2 17.6
Manu ac u ing goods
( ood p ocessing sec o s
a e excluded)
To al o eign alue added 17.1 17.1 15.3 63.4 70.3 72.3 40.8 41.4 36.5
Di ec o eign alue added 5.7 5.8 6.1 58.4 66.8 69.6 32.7 33.7 29.8
To al domes ic alue added 82.9 82.9 84.7 36.6 29.7 27.7 59.2 58.6 63.5
Di ec domes ic alue added 22.5 21.4 21.6 9.4 8.9 9.2 15.8 15.7 17.0
Sou ce: Au ho s’ es ima es based on China’s 2007, 2012, 2017 benchma k inpu -ou pu able pub-
lished by he Bu eau o Na ional S a is ics and O icial China ade s a is ics om China Cus oms.

134 Kun Cai and Zhi Wang
he inc ease in expo s in se ices (inc eased om 14.4% in 2007 o 19%
in 2017). The inc ease in he DVA sha e in China’s no mal me chandise
expo s only played a ela i ely mino ole (inc eased only app oxima ely
1.1 pe cen age poin om 84% in 2007 o 85.1% in 2017). This empi ical
inding may indica e ha he a ious indus ial policies and implici LCR
measu es p oposed in China du ing ecen decades played no signi ican
ole in p omo ing DVA in China’s o al expo s, a leas a he agg ega e
le el du ing 2007–2017.
4.3 Domes ic con en in expo s by i m owne ship
The e is a signi ican change in expo s uc u e by i m owne ship be ween
2007 and 2017. The sha e o p i a e i ms inc eased d ama ically, om 21.3%
in 2007 o 44.2% in 2017, mo e han doubling wi hin 10 yea s. A he same
ime, he sha e o bo h s a e-owned and o eign-in es ed en e p ises (SOEs
and FIEs) declined om 18.9% o 9.1% and 56.3% o 45.3%, espec i ely.
Bo h p i a e i ms and FIEs a e he majo playe s in China’s expo success,
and one may be in e es ed in he DVA sha e in hei expo s. Howe e , since
he e is no in o ma ion on sepa a e inpu -ou pu coe icien s by i m owne -
ship, we canno meaning ully dis inguish o eign e sus local i ms wi hin a
sec o . Ins ead, we p o ide an es ima e o he DVA sha e o agg ega e me -
chandise expo s by i m owne ship. By cons uc ion, he di e ences ac oss
i ms o di e en owne ships a e d i en en i ely by di e en deg ees o hei
eliance on p ocessing expo s wi hin a sec o and he di e ence in he sec o
composi ion o hei o al expo s (bo h a e obse ed di ec ly om he cus-
oms ade s a is ics).
Es ima es o he DVA sha es by i m owne ship o China’s me chandise
expo s a e p esen ed in Table 5.11. The esul s show ha expo s by wholly
o eign-owned en e p ises exhibi he lowes sha e o DVA, ollowed by
Sino- o eign join - en u e companies (dec eased om 44.2% and 56.7% in
2007 o 43% and 52.7% in 2017, espec i ely). Expo s om Chinese p i a e
en e p ises embodied he highes DVA sha es (80.7% in 2007, 77.8% in 2012
and 81.1% in 2017), while hose om s a e-owned i ms we e in he middle
(app oxima ely 70% in he h ee yea s). I is also in e es ing o obse e ha
he a ia ion o DVA sha e in no mal expo s is ela i ely small by di e en
i m owne ships and o e he h ee benchma k yea s (be ween 82% and 85%).
The weigh s o p ocessing expo s a e he decisi e ac o behind he di e -
ence in DVA sha e be ween p i a e i ms and FIEs (p i a e i ms only ha e
app oxima ely 10% o hei expo s as p ocessing expo s, while app oxima ely
wo- hi ds o expo s om FIEs a e p ocessing expo s). No e ha hese a e
es ima ions based on he cu en ly a ailable in o ma ion; be e es ima es can
be de i ed once in o ma ion on I/O coe icien s by i m owne ship becomes
a ailable.
Compa ed o Table 5.11 o Koopman e al. (2012), he mos no iceable
ea u e o his able is he ela i ely s able DVA sha e in expo s p oduced
LCR policies in China 135
by FIEs om 2007 o 2017 (i sligh ly declines). F om 2002 o 2007, he
DVA sha e inc eased by mo e han 10 pe cen age poin s. I seems ha FIE
expo e s did no sou ce mo e o hei in e media e inpu s wi hin China a e
2007. This inding p o ides u he e idence ha he implici LCR measu es
we desc ibed in Sec ion 2 did no a ec mos FIEs’ decision o sou ce hei
p oduc ion inpu s ou side China a he agg ega e le el. Thei use o impo ed
inpu s inc eased du ing his pe iod.
4.4 Domes ic con en in Chinese expo s by ading pa ne s
By assuming ha DVA sha es wi hin a gi en sec o and expo egime (no -
mal/p ocessing) a e he same o all des ina ion coun ies, we can u he
es ima e he DVA sha e in China’s expo s o each o i s majo ading pa -
ne s. Howe e , he a ia ion by des ina ion is d i en solely by China’s expo
s uc u e (sha e o p ocessing expo s and sec o composi ion) o each o i s
ading pa ne s. The esul s o China’s o al me chandise expo s o each o
i s majo ading pa ne s a e epo ed in Table 5.12 in inc easing o de o he
es ima ed weigh ed DVA sha e in 2017.
Hong Kong, Singapo e, Japan, he Uni ed S a es and Ko ea a e he op
i e in bo h 2012 and 2017, wi h less han 60% o China’s DVA embodied
in i s impo s om China. China’s expo s o all eme ging economies and
de eloping coun ies embodied much highe DVA han i s expo s o OECD
coun ies. The di e ence is mo e han 10 pe cen age poin s.
In e es ingly, he DVA sha e uni o mly inc eased in China’s non-p ocessing
expo s o all i s majo ading pa ne s om 2012 o 2017, while i uni o mly
declined o p ocessing expo s o all coun ies. This in o ma ion sugges s ha
he LCR policies we discussed in Sec ion 2 did no educe expo ing i ms’
sou ces o aw ma e ials, pa s and componen s a ound he wo ld, a leas in
China’s p oduc ion o p ocessing expo s.
4.5 Domes ic con en in Chinese expo s by indus ies
To see i he e a e in e es ing pa e ns a he sec o le el, Table 5.13 epo s, in
ascending o de o he weigh ed DVA sha e o 2012, he alue-added decom-
posi ion in Chinese manu ac u ing expo s by indus y in 2012 and 2017,
espec i ely, oge he wi h he sha es o p ocessing ade and o eign in es ed
i ms in each sec o ’s expo s and he sec o ’s sha e in China’s o al me chan-
dise expo s. Because he sec o classi ica ions a e mo e consis en be ween
2012 and 2017 han hose in 2007 due o a new e sion o indus ial classi i-
ca ion in China (CSIC, 2002 o CSIC, 201115), we p esen he esul s o each
sec o o he yea s 2012 and 2017 only.
Thi een ou o he 68 manu ac u ing indus ies epo ed in Table 5.13
had a sha e o DVA in hei expo s o less han 51% in 2012, collec i ely
accoun ing o 34% o China’s o al expo s. I is wo h no ing ha mo e
han hal o hese indus ies a e ela i ely sophis ica ed and high- ech, such as
136 Kun Cai and Zhi Wang
s
s o al
e o expo
m owne ship
s
by i
in China’
expo
32.1
12.4
9.1
2.1
44.2
100.0
36.5
15.0
12.4
2.5
33.6
Sha
100.0
38.0
17.8
18.9
4.0
21.3
100.0
m Owne ship (%), 2007, 2012 and 2017
o al T
domes ic
alue added
43.0
52.7
72.7
78.5
81.1
64.5
40.9
53.0
69.8
74.1
77.8
59.6
44.2
56.7
72.0
73.1
80.7
60.2
eigh ed sum
ec Di
domes ic
eau o Na ional S a is ics and O icial China
W
alue added
12.5
15.1
20.6
20.1
21.3
17.6
11.4
14.3
19.2
19.8
20.8
16.2
13.5
15.4
19.9
19.1
22.2
16.5
o al T
domes ic
alue added
28.0
29.2
22.4
32.5
30.2
28.2
30.0
31.2
28.4
33.9
29.8
30.1
published by he Bu
36.1
38.5
39.4
41.8
42.2
37.1
s by Fi
P ocessing
chandise Expo
ec
domes ic
alue added
able
9.0
9.6
9.2
9.4
9.5
9.2
8.7
8.8
10.3
9.4
9.6
8.9
Di
-ou pu
11.4
8.8
9.8
9.5
10.3
10.0
T
domes ic
o al
alue added
o al Me
84.8
84.8
84.2
83.5
85.6
85.2
82.6
82.5
81.7
82.3
84.3
83.4
83.8
83.6
83.4
83.1
84.9
83.4
p ocessing
alue Added in T
one-
ec
N
Di
domes ic
alue added
22.0
22.5
23.2
21.2
22.3
22.3
21.8
21.8
21.8
21.9
22.3
22.1
23.8
22.9
23.4
22.4
23.5
23.5
s
s 2007, 2012, 2017 benchma k inpu ’
s in
e o Domes ic V
e o
.
Sha
p ocessing
expo
o al expo
73.5
57.8
18.7
9.9
8.0
36.2
79.2
57.5
22.3
16.9
12.0
44.7
83.0
59.7
25.9
24.3
9.7
50.1
eign owned
ms
ms
ms
ms
ms
ms
ms
om China Cus oms
ms
eign owned
ms
Sha
e i
e i
e i
eign owned
ms
’ es ima es based on China
ms
ms
5.11
Wholly o
Join en u
Collec i ely owned i
ms
ableT
P i a e i
All i
2012
Wholly o
Join en u
Collec i ely owned i
Wholly o
Join en u
Collec i ely owned i
ms
2017
S a e owned i
S a e owned i
ms
Au ho s
P i a e i
All i
All i
:
2007
S a e owned i
P i a e i
ceSou
ade s a is ics
LCR policies in China 137
, 2012 and
2017
11.3
2.0
6.1
19.1
4.5
1.8
1.6
2.2
14.3
2.1
1.7
1.9
1.7
1.3
1.6
3.0
3.1
1.5
1.9
5.6
3.2
6.1
2.3
s o
100.0
% o expo
he wo ld
cen ages
2012
14.0
2.0
7.6
17.6
4.3
1.7
1.3
2.0
14.8
2.1
1.7
1.8
1.6
1.6
1.7
3.8
2.4
0.8
2.2
5.5
3.2
3.9
2.5
100.0
, as Pe
58.1
59.0
59.7
63.7
63.7
74.0
76.6
74.8
75.1
76.3
64.6
ne s ading Pa
eigh ed sumW
2017
51.3
53.3
59.1
64.4
67.0
67.6
68.4
71.3
74.6
76.4
77.1
79.9
78.0
2012
41.9
50.6
56.6
54.5
54.7
56.6
58.7
59.4
60.0
64.5
64.1
68.4
66.3
68.4
69.6
70.3
71.2
71.5
73.2
74.2
75.2
75.5
76.0
59.6
eau o Na ional S a is ics and O icial China
s o i s Majo T
2017
28.8
24.7
29.8
28.1
29.3
28.0
28.0
27.8
28.1
27.1
27.6
27.8
29.6
28.9
27.2
26.7
29.1
25.3
28.3
27.8
27.3
29.6
29.7
28.3
P ocessing
2012
28.8
27.4
32.3
29.9
31.8
32.7
30.1
29.6
29.9
31.8
31.3
32.8
33.7
32.6
28.2
29.7
32.1
31.6
30.1
30.4
30.8
33.5
33.4
30.2
published by he Bu
chandise Expo
able
86.8
-ou pu
on p ocessing
2017
83.8
83.0
85.6
85.3
85.2
84.5
84.5
85.5
85.6
85.8
85.2
85.1
84.5
85.0
84.7
83.9
85.3
85.9
85.3
85.4
85.7
85.4
85.2
oss Me
N
2012
81.7
82.2
85.3
84.1
82.0
81.6
82.4
82.8
83.9
83.5
84.1
83.9
82.3
82.6
82.6
82.1
81.7
82.8
84.5
83.7
83.3
83.9
84.0
83.4
e in Chinese G
% o p ocessing
2017
59.2
51.0
50.4
46.3
46.9
44.6
36.8
36.7
36.8
31.8
31.3
29.3
24.9
19.0
14.6
17.0
16.9
14.8
18.8
14.2
9.3
16.7
13.3
36.3
s
alue Added Sha
expo
2012
75.2
57.7
54.1
54.6
54.4
51.1
45.3
44.0
44.3
36.8
37.8
30.3
32.9
28.6
23.9
22.6
21.1
22.1
20.7
17.8
15.3
16.8
15.8
44.7
2017
s 2012 and 2017 benchma k inpu ’’ es ima es based on China
.
Domes ic V
h A ica
ince
ope/Cen al Asia
om China Cus oms
o
5.12
e
Au ho s
ea Rep
n Eu
:
able
gion desc ip ion
ce
ea
-Saha an A ica
Re
Y
Hong Kong
Japan
Uni ed S a es
Mexico
es EU15
T
Singapo
Taiwan p
Aus alia/NZ
Res o OECD
Res o Sou heas Asia
Thailand
B azil
Indonesia
RUS
Middle Eas /No
o al
Ko
Eas EU12
W
Res o La in Ame /Ca ibbean
India
Philippines
Sub
Res o Sou h Asia
Eas e
T
Sou
ade s a is ics
240 Lili Yan Ing and Rui Zhang
Gopina h, G., Neiman, B., 2014. T ade adjus men and p oduc i i y in la ge c ises.
Ame ican Economic Re iew 104, 793–831.
G ossman, G.M., 1981. The heo y o domes ic con en p o ec ion and con en p e -
e ence. The Qua e ly Jou nal o Economics 96, 583–603.
Head, K., Maye , T., Meli z, M., 2022. The La e cu e o ules o o igin. h ps://
schola .ha a d.edu/si es/schola .ha a d.edu/ iles/meli z/ iles/hmm_ oo_la e _
sha ed.pd
Ju, J., K ishna, K., 2005. Fi m beha iou and ma ke access in a ee ade a ea wi h
ules o o igin. Canadian Jou nal o Economics/Re ue Canadienne d’économique 38,
290–308.
K ishna, K., I oh, M., 1988. Con en p o ec ion and oligopolis ic in e ac ions. The
Re iew o Economic S udies 55, 107–125.
Lahi i, S., Ono, Y., 1998. Fo eign di ec in es men , local con en equi emen , and
p o i axa ion. The Economic Jou nal 108, 444–457.
Qiu, L.D., Tao, Z., 2001. Expo , o eign di ec in es men , and local con en equi e-
men . Jou nal o De elopmen Economics 66, 101–125.
Yang, C., 2021. Rules o o igin and au o-pa s ade. h ps://chenying-yang.com/
RoO_au opa s.pd

A.1 Equilib ia wi h and wi hou LCR
In he equilib ium wi hou LCR, he domes ic p ice index o composi e inpu
pu chased by sec o k om sec o s,Q
D
ks
,, is gi en by
PC
z
D
ks s
s
i
is
ks
s
ss
,
,.uj
h
h
h
h
y
j
j
j
j
y
y
y
y
y
y
y
y
h
jj
h
h
h
hh
1
1
1
1
b
(A1)
In con as , he same p ice index in he equilib ium wi h LCR is gi en by:
PC
z
C
z
D
ks s
s
i
is
ks
i
is
s
s
s
,
,
1
1
nn n
NB
sCNC
kks
D
ks s
s
i
is
ss
s
k
PC
,
,
,
j
j
k
k
k
]
1
1
1
1
OG
nzz kF
PC
ks
D
ks s
s
i
ss
s
,
,
,,
j
j
k
k
k
]
=
1
1
1
1
OG
n
ii
S
ks
i
i
S
ks
z
C
z
s
s
s
j
j
k
k
k
]
*
,,
11
1
1
n
s
kF,
(A2)
whe e
ΩΩ
NB
s
C
s
, and
ΩNC
s deno e he se s o i ms ha ind hei LCR non-
binding, i ms ha decide o comply wi h hei binding LCR, and i ms ha
Appendix A
Appendix: Desc ibing he ull
equilib ium
242 Lili Yan Ing and Rui Zhang
decide no o comply, espec i ely. We can cha ac e ize he compliance deci-
sion o i m i based on
iS
sii
sii si
s
s
d
g
g
gg
ghgh
ghgh ghg
g
gh
g
g
g
NB
COG
N
i
i and
,,
,,
1
1
1
1
CC OG
i and
sii si
S
s
s
ghgh ghg
g
gh
g
g
g
g
g
g
g
h
g
g
g
g
,,
,
1
1
1
1
(A3)
whe e
Sz
zS
ss
kks ks
s
kks k
s
s



OG OG sOGOG
,,,
,,
,
,,
d
d
g
g
1
1ss ,
and
b
b
ks
D
ks ks
PX
s
,
,,
g
b
b
1
Inpu demands o goods p oduced by sec o s depend on he sizes o o he
sec o s k and he inpu -ou pu linkage be ween k and s. Meanwhile, we assume
ha he inal consump ion expendi u es XF,s a e ixed. So he inpu ma ke
clea ing condi ion is
XC
zP
ks
D
ks
in
ii
k
k
i
nk D
nk
k
kk
,,
,
,
hn
n
g
h
h
n
m
j
jn
g
hn
m
j
j
n
jj
jhgg
g
gg
11
yyy
g
g
g
g
g
g
h
n
g
k
Xnk
1,, (A4)
and
XC
z
ks
D
ks
in
ii
k
k
i
nk
k
k
,,
,
b
nn
bb b
n b
b
b
g
g
u
b
b
b
b
b
g
g
u
b
b
b
gghgh
h
h
hOG
111 1
1
b
bbb
n
bb
nh
b
b
g
g
g
hb
b
hh
gghgh
h
k
k
k
PX
D
nk nk
D
ks
i
ii
k
k
,,
,
hNC bb
g
g
u
b
b
b
b
b
g
g
u
b
b
b
nh
b
b
g
bb
bbb
hhh
hh
kk
k
C
zPX
i
kD
kk
OG
OG OG
,
,,
11
gg
g
hb
b
b
g
g
u
b
b
b
b
b
g
g
u
b
b
b
nm
bb b
gghgh
h
h
D
ks
i
ii
k
k
i
k
kk
k
C
z
,
,
hh
NB C
OG
111 1
b
bbb
nh
b
b
g
g
g
hh
k
k
PX
D
kk
OG OG
,,
(A5)
Quan i ying he impac s o local con en equi emen s 243
Combining he condi ions shown, we can de ine he wo equilib ia wi hou
and wi h he LCR.
P oposi ion 1 (Equilib ia wi hou and wi h LCR). Gi en exogenous a iables
φi, zk,s,
aDi
k,, a
Fi
k,, bL,i,
PF
k
, wk,
β
D
ks
,, XF,k, and λi, and pa ame e θ and σ s, he
equilib ium wi hou LCR is a ec o o p ice indexes {
PD
ks
,} ha sa is ies equa-
ions (A1) and (A4) o all k and s. The equilib ium wi h LCR is a ec o
o p ice indexes {
PD
ks
,
′
} ha sa is ies equa ion (A2), (A3), and (A5) o all k
and s.
A.2 Equilib ium in ela i e changes
We in es iga e he impac s o he LCR imposi ion by o mula ing he equilib-
ium in ela i e changes. The ela i e change o p ice index is:
PmP
D
ks
i
i
ks D
s
ii
s
s
hh
n
n
,,
g
g
g
yy
u
u
u
i
o
l
l
k
j
kk
k
m
11
1
1
1
1
u
yy
u
ggh
hNB
ss
s
s
s
i
i
ks D
s
ii
i
i
i
k
mP
m
by
u
u
u
i
o
l
l
b
j
kk
n
k
j
m
m
h
h
C
NC
,
h
n
n
gghn
yy
u
1
1
1
1
1
,, ,,
sD
s
ii
Pk
s
h
n
n
gghm
yy
u
1
1
1
1
1
kk
k
y
u
u
u
i
o
l
lyOG
PmP
D
ks
i
i
ks D
s
ii
s
s
hh
j
j
,,
u
g
gg
h
hhh
h
j
j
j
j
g
g
gg
g
11
1
1
1
g
gg
gg
gg
NB
sNC
U
ghgh
gh
hh
h
j
j
j
j
gh
ghh
g
g
gg
j
g
g
1
1
1
1
1
1
g
gg
g
ggk
s
s
s
i
i
ks D
s
ii
i
mP k
gC
OG
,
,,
h
j
j
FF
PmP
D
ks
i
i
ks D
s
ii
s
ss
,,
.
.
,,
hh
h
hh
hh
hhhh
h
h
h
j
j
h
h
hh
h
h
11
1
1
1
h
hh
hh
hh
NB NC
hh
h
hh
h
h
h
j
j
h
hhh
h
h
hh
h
h
h
h
1
1
1
1
1
1
1
h
hh
h
hhhj
s
s
s
i
i
ks Dii
i
i
mP
hh
C
s
,
,
.
.
ss
mkF
i
ks
*
h
hh
h
h
h
h
hh
hh
hhh
,
,,
(A6)
whe e
mY
X
i
ks
i
ks
ks
,,,
/= is he ma ke sha e o i m i in sec o k’s o al inpu
pu chase om sec o s, o he ma ke sha e o i m i in he inal consump-
ion demand. The change in ou pu p ices o non-binding i ms a ises om he
gene al equilib ium e ec ha a ec s he domes ic composi e inpu p ices. Fo
ins ance, non-binding i ms educe domes ic inpu usages when he domes ic
244 Lili Yan Ing and Rui Zhang
composi e inpu p ices inc ease. In addi ion o he gene al equilib ium e ec ,
he compliance cos penal ies
h
i
j
di ec ly in la e he ou pu p ices o he comply-
ing i ms. Meanwhile, non-complie s a e cha ged an ad alo em non-compliance
ee o τ when selling o he ups eam OG sec o , which also inc eases hei p ices.
Acco ding o he Cobb-Douglas o mula ion, he ela i e change in he
cos o domes ic composi e inpu is
PP
DnD
nD
sn

ss
gg
g
gg
g
g
g,,
.
b
(A7)
Looking in o he change in he local con en
λ

i
o i m i and assuming ha
o eign composi e inpu cos P
F
k is no a ec ed by he LCR, we no ice ha
h
hh
hh
hh
h
h
,
.
.
i
iD
k
i
ii D
k
P
P
h
h
h
hh
h
hhh
hh
h
h
h
h
g
g
g
g
h
h
hh
h
hh
hh
h
h
1
1
1
1
1
1
11
1
1
h
hh
h
h
h
h
g
g
g
ghh
hh
h
h
hh
h
h
ii
, (A8)
which depends on
PD
k

gi en λi, γi, and θ. Hence, we can ge
i
i
i
'=

. Wi h
i
'
in hand, we can also calcula e
k
i
':
g
g
ggg
g
gg
g
g
i
ii
g
gg
g
g g
;
;
[
g g
g
g
;
;
[
]
]
]
11
1
11
1
() . (A9)
The change in domes ic inpu sha e wi hin i m i’s local con en is
m
mm
b
b
h
m
m
i
D
k
iD
k
i
P
P
b
b
b
b
b
gh
()
h
h
1
1
1
. (A10)
So we can ew i e he ela i e change in i m-le el uni cos as
C
Pi
P
i
D
k
ii
kk
iD
k
ii
m
m
m
,
,
,
,
b
bn
nn
mnmn
mnmn
n
n
1
1
1
1
1
1
1
1
,i NC NB
,,i C
ik
b
b
b
n
n
m
n
nn
(A11)
Quan i ying he impac s o local con en equi emen s 245
The o al expendi u e on domes ic inpu in he LCR equilib ium can be w i -
en as:
XYXC
P
ks ks
D
ks k
k
in
ii i
nk iD
nk
k
kk
,,
,,
,



gg
gh
hh
gg
jj
jj
jgg jj
11
hOG
11
111
X
CP XY
nk
i
ii i
kiD
k
s
k
kk




,
,,
h
j
k
k
mg
g
hh gg
j
ngg
jjj
hNC
OG OG OG,,
,,,
,
k
i
ii i
kiD
kk
kk
kk
YCPXm
b
b
b
gn
hh gg
j
hh
NB C
OG OG OG
gg jj



11 (A12)
which helps o de ine
X
ks

,gi en o he a iables. The ollowing p oposi ion
desc ibes he ela i e change o he equilib ium caused by he LCR imposi ion.
P oposi ion 2 (Equilib ium in ela i e changes). Gi en endogenous a iables
V
ii i
ks ks
YX
,, ,
,,
{}
, exogenous a iables {XF,s}, policy a iables {τ, λi}, and
pa ame e s
b B
,,,s
D
ks
{}
, a ela i e change o he equilib ium caused by he
LCR is a ec o o p ice index changes
PD
ks

, ha sa is ies (A3), (A6), (A7),
(A8), (A9), (A10), (A11), and (A12).
Once we calib a e and ob ain he alues o {λi, γi,
Y
i
ks
,, Xk,s}, {XF,s}, {τ, λi},
and {θ, σ s,
β
D
ks
,}, we can e alua e he e ec s o imposing he MEMR LCR on
i ms and he economy.

Index
ADF G oup Inc. . Uni ed S a es o B azil 2, 4, 22, 34–37, 63, 118, 174
Ame ica 162–163 B azil, Russia, India, China, and Sou h
Ag eemen on Subsidies and A ica (BRICS) 174
Coun e ailing Measu es (SCM BREXIT 4, 7
Ag eemen ) 10, 63, 67, 150, Bu kina Faso 54
153–154, 164
Aha o Linkages P og am 73 Canada 2, 6, 58–59, 60, 71, 72, 88, 91,
Al a Ley Na ional Mining P og am 60 102, 112, 118, 158, 160–161,
Ame ican Au omobile Labeling Ac 174; seealso No h Ame ican
(AALA) 91, 96, 102–103 F ee T ade Ag eemen (NAFTA)
Anglo Ame ican Co po a ion 73 Ca gill, Inc. . Mexico 159
An o agas a Region Mining Clus e Chile 2, 6, 59–60, 61
(CMRA) 60 China: au omobile indus y 112,
A che Daniels Midland Company 116–119; domes ic con en in
(ADM) . Mexico 159–160 agg ega e expo s 132–134;
A gen ina 2, 5, 22, 30–31, 64, 174 domes ic con en in expo s
ASEAN–Aus alia–New Zealand by i m owne ship 134–135;
F ee T ade A ea (AANZFTA) domes ic con en in expo s by
Ag eemen 155 indus ies 135–142; domes ic
ASEAN–Japan FTA 165 con en in expo s by ading
Associa ion o Sou heas Asian Na ions pa ne s 135; domes ic con en
(ASEAN) 155 in expo s/p ocessing expo s
Aus alia 2, 6, 58–59, 60, 62, 66, 72, 174 128–131; domes ic con en in
au omobile indus y: A gen ina 22, p oduc ion 126–131; impac s
30–31; China 112, 116–119; o LCR policies 8–9, 114–142;
impo subs i u ion 2; Indonesia indus ial policy ini ia i es 4,
4–5, 23, 37–39, 64; egional 115–126; in eg a ed ci cui
pa s use 91–97; ules o o igin 4, indus y 119–123; Made in
6, 7, 87–112; heo e ical model China 2025 4, 8, 116, 123;
o pa s sou cing 97–111; Uni ed medical supplies 123–126; mining
S a es 87–112, 119 indus y 53, 61, 62, 63, 64, 68,
Au o Pac 88 69, 71; pha maceu ical indus y
123–126; elecommunica ions
backwa d linkages 5, 10, 16, 49, 51, indus y 123; ade ag eemen s
54–60, 178, 212 10; use o LCRs 174
bes e o s clauses 51, 54, 70 China – Audio isual dispu e 163
bila e al in es men ea ies (BITs) 10, 155 China – Au o Pa s dispu e 161
BMW 118–119 compliance 1, 7, 10, 49, 55, 57, 58, 90,
Bo swana 65, 69–70 97, 164
Index 247
compliance cos s 11, 99–102, 111, 150, 152–153, 155–157, 164,
213–214, 229, 235, 244 165; seealso No h Ame ican
compliance decision 11, 101–104, 106, F ee T ade Ag eemen (NAFTA)
212–216, 220, 225, 226, 228, Fundación Chile 60
230–233, 234, 238
comp ehensi e economic pa ne ship Gene al Ag eemen on Ta i s and T ade
ag eemen s (CEPAs) 10, (GATT) 9, 10, 63, 67, 150,
155–157, 164 152–153
Consejo Mine o 60 Gene al Ag eemen on T ade in Se ices
Con empo a y Ampe ex Technology (GATS) 9–10, 150, 153
Co., Limi ed 64 Ge many 4, 18, 118
con en p o ec ion see local con en Ghana 6, 54, 55–56, 68, 73
equi emen s (LCRs) global alue chains (GVCs) 54, 59, 60,
Co n P oduc s In e na ional, Inc. (CPI) 63, 71, 114
. Mexico 159–160 go e nmen p ocu emen 14, 20, 21,
22, 36, 39, 41, 115–116, 123,
De Bee s Diamond Conso ium 65 126, 145, 146–148, 149, 152,
de ac o measu es 159–160 154, 163, 178, 198, 215
demand-side policies 5, 50–51, 54–60,
66, 68, 72 Huawei 114
Democ a ic Republic o Congo 54 Huayou 64
Diamond T ading Company (DTC Hyundai 64
Bo swana) 65
di e ence-in-di e ences (DID) es ima o impo licensing p ocedu es 114–115
175, 208–211 incen i es-based policies 5–6, 15, 18, 19,
domes ic employmen equi emen s 49, 28, 51–53, 54, 58–60, 67, 71, 88,
51, 87 89, 97, 102, 115–116, 146, 147,
domes ic ma ke obliga ions 15, 16, 17, 154, 164, 165, 178, 198, 212
18, 63, 214 India 4, 18, 118, 174
domes ic p ocessing equi emen s 63, Indigenous Land Use Ag eemen s 59
146 Indonesia: au omobile indus y 4–5, 23,
domes ic alue added (DVA) 8–9, 115, 37–39, 64; Ben eng P og am 9,
127–128, 130, 132–134, 145, 174; Dele ion P og am 9,
145–146, 164, 175 145, 174; e ec s o LCR policy
on ade lows 174–198; ene gy
ene gy sec o 17, 81–86, 146, 147–148, sec o 147–148; impac s o LCRs
153 on manu ac u ing i ms/sec o s
equilib ium 241–245 212–238; impac s o LCRs on
Eu opean Union and Uni ed Kingdom p ices and wel a e 235–237;
T ade and Coope a ion Ac impac s o LCRs on sales,
(TCA) 7, 112 alue added, and employmen
Eu opean Union (EU) 63, 71, 90–91 233–235; Inc eased Use o
expo bans 53, 61–64, 146–147 Domes ic P oduc ion P og am 9,
expo licensing equi emen s 53, 63 145; LCR compliance decisions
220–238; LCR o ups eam OG
Finland 60 sec o 215–220; lessons om
Fo d Mo o Company 64 WTO and ISDS cases 161–163;
o eign di ec in es men (FDI) 54 mining indus y 61–64; Minis y
o wa d linkages 6, 60–65 o Ene gy and Mine al Resou ces
F ance 118 Regula ion No. 15/2013 11;
ee ade 98 mode n e ail 149; Na ional
ee ade ag eemen s (FTAs) 7, 9–10, Ca P og am 9, 145, 174;
11, 12, 63, 67, 89, 112, 146, pha maceu ical indus y 149;
248 Index
elecommunica ions indus y Mine al and Coal Mining Law 62
148; ade ag eemen s 9–11, mining indus y: Aus alia 58–59;
150–161, 164–165; ade and Bo swana 65; Canada 58–59;
indus ial policies on LCRs Chile 59–60, 61; China 53, 61,
146–149, 164 62, 63, 68, 69, 71; economic
Indonesia–Japan Economic Pa ne ship impac s o LCRs 5–6, 68–71;
Ag eemen 164–165 Ghana 6, 54, 55–56, 68, 73;
In la ion Reduc ion Ac 2 Indonesia 61–64; ins i u ional
in eg a ed ci cui (IC) indus y 119–123 amewo ks/coo dina ion
in e na ional in es men ag eemen s 66–68; policy implica ions o
(IIAs) 146, 155, 158, 164 LCRs 71–74; Sou h A ica
in e en ions 18–20 23, 39–40, 56–57, 69, 73;
in es men ea ies 10, 68, 155–157, 160 Tanzania 6, 54, 57–58, 64, 66,
In es o -S a e Dispu e Se lemen 73; ypology o LCRs a ec ing
(ISDS) 10, 157–161, 165 50–53; Zambia 58
I eland 91 Mining Skills Council (CCM) 60
Mobil In es men s Canada Inc. and
Japan 10, 60, 71, 118 Mu phy Oil Co po a ion .
Canada 158, 159
Kazakhs an 67, 69 mode n e ail 149
Kenya 17, 54 Mozambique 54
labo 18, 51 Namibia 54
LG Ene gy Solu ion 64 na ional ea men obliga ion 67,
local con en equi emen s (LCRs): 160–161, 162, 164
al e na i e model 192–193; Newmon –In e na ional Finance
China 114–142; de ini ion 14, Co po a ion 73
50, 192–193; de imen al e ec s Nige ia 54
16–17; economic impac s 68–71; non-compliance ees 11, 213, 215–216,
e ec s on expo s 187–192; 220, 222, 223, 225–226, 230,
e ec s on impo s 182–186; 232, 233, 235, 236
e ec s on ade 174–198; No h Ame ican F ee T ade Ag eemen
Indonesia 145–165, 174–198, (NAFTA) 6, 7, 87, 88, 89, 91,
212–238; li e a u e e iew 97, 99–100, 102, 106, 108, 112,
176–178, 197; measu ing 158, 159–160, 162, 214
impac s 53–54; mining indus y
5–6, 23, 39–40, 48–74; ope a ions 18
moni o ing 35, 43, 58, 66, 73, O ganisa ion o Economic Co-ope a ion
149; policy implica ions 71–74; and De elopmen (OECD)
ecen implemen a ion 18–20, 21, 145
145, 174; scope/con en 50; owne ship equi emen s 68–69
heo y/p ac ice 1–3,
15–18, 145 pe o mance equi emen s p o isions
localiza ion ba ie s 14, 20 158–160
pha maceu ical indus y 123, 149
Malaysia 62 p esc ip i e bene icia ion equi emen s
Mali 54 53, 54, 59, 69, 70, 72–73
medical supplies 5, 23, 39, 123–126 P oduc ion De elopmen Co po a ion
Me ill & Ring Fo es y LP . Canada 160 (CORFO) 59–60
METRO 4, 5, 21 PT Aneka Tambang 64
Mexico 2, 92, 102, 112, 118, 159; PT Indus i Ba e ai Indonesia (IBC) 64
seealso No h Ame ican F ee public–p i a e pa ne ships 51,
T ade Ag eemen (NAFTA) 59–60, 70
Index 249
QMB New Ene gy Ma e ials 64 ade ag eemen s 1–2, 10, 145–165,
179–198
Regional Comp ehensi e Economic T ade-Rela ed In es men Measu es
Pa ne ship (RCEP) Ag eemen (TRIMS) 9, 10, 67–68, 146,
154, 155 150–151, 155, 164
egional con en equi emen (RCR) 87, Tsingshan 64
89, 90, 91, 97, 101, 103–106,
109, 111 Uni ed Kingdom 4, 7, 18, 91
egional in es men ag eemen s (RIAs) Uni ed S a es: au omobile indus y 4,
155–157 87–112, 118; mining indus y
Regional T ade A ea (RTA) 88 60, 71; ade ag eemen s 10;
epo ing equi emen s 55, 66, 73 ade wa 114; use o LCRs
equi emen -based policies 51, 53–58, 174; seealso No h Ame ican
60–65, 68 F ee T ade Ag eemen
esou ce alloca ion e ec s 17, 176, 214 (NAFTA)
obus ness check 193–197 Uni ed S a es Buy Ame ica Ac 5, 16,
ules o o igin (RoOs) 1–2, 3, 4, 6, 7, 23, 40–41
87–91, 92, 95, 97, 98–99, 101, Uni ed S a es-Mexico-Canada T ade
104, 108, 177, 214 Ag eemen (USMCA) 1, 4, 6–7,
Russia 118 87, 89–90, 102, 104–106, 109,
112, 214
Saudi A abia 5, 23, 39 US – Renewable Ene gy dispu e 161
S.D. Mye s . Canada 160
sou cing 16, 18, 21, 49, 51, 67, 69, 70, Valo Mine o 60
73, 87–101, 112, 142, 149, 176, alue-added ac i i ies 1, 3, 6, 15, 49,
198, 213, 214, 220, 221–223, 56, 58, 68–69, 70, 87, 116, 119,
226, 229, 232, 238 123, 135, 176, 212, 213, 215,
Sou h A ica 5, 6, 23, 39–40, 54, 56–57, 230, 233–234, 238
69, 73 alue-added expo s 128
Sou h Ko ea 10, 64, 118 alue-added sha es 8–9, 132–134
Subsidies and Coun e ailing Measu es alue-added ax (VAT) 22
(SCM) 10 Venezuela 64
supplie de elopmen p og ams (SDPs) Volkswagen G oup China 64
51, 66, 74
supply-side policies 5–6, 50–51, 56, 66, Wes Ja a 66
68, 70, 72 Wo ld T ade O ganiza ion (WTO) 4,
Sweden 60 8, 9, 10, 63, 67–68, 71, 81–86,
116, 146, 150, 154, 175
Tanzania 6, 54, 57–58, 64, 66, 73
elecommunica ions indus y 4, 22, Zambia 6, 54, 58
34–37, 123, 148 Zimele En e p ise P og am 73