No2025-08–May2025

WorkingPaper

CEpii

CarbonBiasofTariffs:

AreFossilfuelstheCulprits?

CeciliaBellora,LionelFontagné,ChristopheGouel&YoussefSalib

Highlights

Thestudyfindsthatthemagnitudeofthe"carbonbias"intariffs–wherecarbon-intensivegoodsfacelowertradebarriers–issignificantlysmallerthanpreviouslyestimated,largelybecausehightariffsinagriculturecounterbalancetheunder-taxationoffossilfuelswhenaccountingforallgreenhousegases.

Fossilfuelsstandoutasthemainsourceofbiasbecauseoftheirconsistentlylowtariffs,yetthiseffect

almostdisappearswhenthemodelincludesnaturalresourceconstraintsonfossilfuelextraction.

Innon-fossil-producingcountries,significantdomesticfueltaxesalreadyactmuchliketariffs,reversingtheapparentbiasofundervaluedfossilfuels.

Giventhesefindings,policyreformsaimedatharmonizingprotectionacrosssectorswouldhavemodest,ifnotnegative,globalclimatebenefits.

CeciliaBellora,LionelFontagné,ChristopheGouel,&Youssef

RESEARCHANDEXPERTISE

ONTHEWORLDECONOMY

CEPIIWorkingPaperCarbonBiasofTariffs:AreFossilFuelstheCulprits?

Abstract

Thispaperrevisitstheexistenceofacarbonbiasintradepolicies,whereemissions-intensivesectorsreceivelowertradeprotectionthancleanersectors.Usingastylizedgeneralequilibriummodelthataccountsforgreenhousegasemissions,weconfirmthepresenceofacarbonbiasbutfindittobesignificantlysmallerthanpreviouslyestimated.Ouranalysisrevealsthatthisbiasisprimarilydrivenbylowtariffsonfossilfuels,particularlycrudeoil.Incorporatingthefinitenatureoffossilfuelresourcesintothemodelreducestheresponsivenessoffossilfuelproductiontotariffchanges,effectivelyneutralizingthecarbonbias.Furthermore,whenaccountingfordomesticconsumptiontaxesonfossilfuelsinnon-producingcountries–whichactasdefactotariffs–thebiasshiftstowardapro-environmentalstance.Thesefindingsunderscoretheimportanceofintegratingenergymarkets'specificitiesanddomesticdistortionsintotrademodelstobetteraccountfortheimpactoftradepoliciesontheenvironment.

Keywords

FossilFuels,GreenhouseGases,InternationalTrade,Tariffs.

JEL

F13,F18,Q40,Q56.

WorkingPaper

CEpii

?CEPII,PARIS,2025

ISSN2970-491X

EdItoRIAldIRECtoR:AntoInEBou?t

CEPIIWorkingPaper

Contributingtoresearchininternationaleconomics

May2025

Centred’étudesprospectives

etd’informationsinternationales

20,avenuedeSégurTSA10726

75334ParisCedex07

contact@cepii.fr

www.cepii.fr

–@CEPII_ParisPresscontact:presse@cepii.fr

CEPII(Centred’étudesProspectivesetd’InformationsInternationales)isaFrenchinstitutededicatedtoproducingindependent,policy-orientedeconomicresearchhelpfultounderstandtheinternationaleconomicenvironmentandchallengesintheareasoftradepolicy,competitiveness,macroeconomics,internationalfinanceandgrowth.

VISuAldESIgnAndPRoduCtIon:

lAuREBoIVIn

Tosubscribeto

TheCEPIINewsletter:

www.cepii.fr/KeepInformed

Allrightsreserved.Opinionsexpressedinthispublicationarethoseoftheauthor(s)alone.

RESEARCHANDEXPERTISE

ONTHEWORLDECONOMY

CEPIIWorkingPaperCarbonbiasoftariffs:Arefossilfuelstheculprits?

3

Carbonbiasoftariffs:Arefossilfuelstheculprits?

1

CeciliaBellora,

*

LionelFontagné,

t

ChristopheGouel,

?

andYoussefSalib

§

1.Introduction

Theimpactoftradeongreenhousegas(GHG)emissionsistypicallyanalyzedthroughscale,technique,andcompositioneffects(

CopelandandTaylor

,

1994

;

GrossmanandKrueger

,

1994

;

Copelandetal.

,

2022

).Amongthese,compositioneffectswhichdescribehowtraderedistributesresourcesbetweenlow-andhigh-emissionindustriesareparticularlyrelevantforclimatepolicy.Theextenttowhichtradepoliciesinfluencethisallocationdependsonthelevelandsectoralvariationoftradeprotectioninimportingcountries.Inaworldwithuniformcarbonpricing,

2

tariffstructureswouldnotaffectGHGemissions.However,inreality,bothcarbonpricesandtariffstructuresvaryacrosssectorsandcountries,potentiallyamplifyingormitigatingemissionsbyincentivizingtradeincleanerordirtiergoods.Onekeyfeatureoftradepolicythatmaycontributetohigheremissionsiswhat

Shapiro

(

2021

)callsthe“environmentalbiasoftradepolicies”whereemissions-intensivesectorsfacelowertradeprotection.Thisbiascanbeattributedtotariffescalation(

AntràsandChor

,

2022

),atariffstructureinwhichupstreamgoodstypicallymoreemission-intensivearetaxedlessthandownstreamproductstoprotectdomesticvalue-addedindustries(

Corden

,

1966

).

ThispaperinvestigateswhethercurrenttradepoliciesleadtohigherGHGemissionscomparedtoascenariowheretariffstructuresareharmonizedacrosssectors(whereeachimporterappliesthesametradepolicyacrosssectors,thoughnotnecessarilyacrosspartners).Analyzingthisrequiresunderstandinghowtariffsinfluencetradecomposition,particularlyforemission-intensiveindustriessuchasfossilfuels,brownindustries,andagriculture.Toanswerthisquestionmeaningfully,however,twokeycomplexitiesmustbeconsidered.First,tradepoliciesdonotoperateinisolation.Domesticpolicies,suchastaxesonfossilfuelconsumption,cancounteractorreinforcetheeffectsoftariffstructures.Forinstance,highdomesticfueltaxesinimportingcountriesmayoffsettheemissionsimpactoflowfossilfueltariffs.Second,GHGemissionsextendbeyond

1TheauthorsaregratefultoLolaBlandinandFran?oisChimitsforexcellentresearchassistance,andtoHousseinGuimbardforhelpwithtariffdata.ThisworkhasbenefitedfromthesupportoftheAgenceNationaledelaRecherchethroughtheprogramInvestissementsd'AvenirANR-17-EURE-0001.

*CEPII:

cecilia.bellora@cepii

tPSE:

lionel.fontagne@psemail.eu

?INRAEandCEPII:

christophe.gouel@inrae.fr

§PSEandEcoledesPonts:

youssef.salib@psemail.eu

2Inthispaper,weusethetermcarbonlooselytodesignateallgreenhousegases,notjustthosethatarecarbon-based.

CEPIIWorkingPaperCarbonbiasoftariffs:Arefossilfuelstheculprits?

4

CO2.Whilecarbondioxideisthemoststudied,methaneandnitrousoxide—particularlyfromagriculture—arealsomajorcontributorstoglobalwarming.Anymeaningfulassessmentoftrade-relatedcarbonbiasmustaccountforthefullrangeofGHGsandthedistinctemissionsprofilesofdifferentsectors.

Theconjunctionoftwounrelatedsourcesofheterogeneity—emissionintensityandborderprotection—makesitdifficulttopredicttheoveralldirectionofapotentialbiasintradepolicy.Similarly,domestictaxationandadditionalGHGscaneitheramplify,counteractorreversetheintuitiverelationshipbetweentariffstructuresandtheirimpactonGHGemissions.Toaddressthesecomplexities,ageneralequilibriummodeloftheworldeconomyisrequired,onethatincorporatesGHGemissionsandinter-regionalinput-outputrelationships.Weemployavariantofthestandardquantitativetrademodeldevelopedby

CaliendoandParro

(

2015

),whichisalsoadoptedby

Shapiro

(

2021

).WeextendthismodeltoincludeGHGemissions,consideringbothCO2emissionsfromfossilfuelcombustion,asanalyzedby

Shapiro

,andotherGHGs.Additionally,ourmodeltakesintoaccountthatfossilfuelproductionrequiressector-specificfactors,ashighlightedby

BaqaeeandFarhi

(

2024

).Itfurtherincorporatesdomestictaxationoffossilfuelsinimportingcountries,integratingbothborderpolicies(tariffs)andbehind-the-borderfiscalmeasures(taxes),whichtogetherinfluencetheenvironmentaloutcomesoftrade.

Themodeliscalibratedusingseveraldatasources:trade,input-output,andGHGemissionsdatafromEXIOBASE;bilateralappliedtariffdatafromMarketAccessMap(MAcMap-HS6);fossilfuelproductiondatafromtheInternationalEnergyAgency;andNetEffectiveCarbonRatesfromtheOECD.Thesediversedatasetsallowustocapturethecomplexinter-sectoralandinternationalrelationshipsthatinfluenceGHGemissions.Althoughouranalysisfocuseson2019,ourdataspantheperiodfrom2007to2019,whichweuseforrobustnesschecks.ThedataindicatethatemissionintensityishighestinfossilextractionandbrownindustriesforCO2,andinfossilextractionandagricultureforotherGHGs.Regardingthepatternofappliedtariffs,thelowaverageprotectionforfossilfuelsandthehighprotectionforagriculturesuggestsignificanteffectsarelikelywhenharmonizingtariffsacrosssectors.

Usingourmodeltosimulatetheharmonizationoftariffsacrosssectors,weassesstheoverallimpactonglobalGHGemissions,consideringbothCO2andnon-CO2GHGs.Ourresultsconfirmtheexistenceofacarbonbiasintariffs:implementingauniformtariffacrosssectorswouldreduceglobalGHGemissions,indicatingthatcurrenttariffstructuresfavorhigh-emissiongoods.However,wefindthatthemagnitudeofthisbiasissmallerthanpreviouslyreportedandislargelydrivenbylowtariffsonfossilfuels,particularlycrudeoil.Meanwhile,othercarbon-intensiveindustrieshaveanegligibleorpositiveimpact.Tobetterunderstandthecontributionoffossilfuelstothisbias,weexploretwoalternativeextensions.First,byextendingthequantitativemodeltoconsiderthelimitedavailabilityofnaturalresourcesrequiredforfossilfuelextraction,wefindthebiasshiftsclosetozero.Second,wearguethatdomestictaxesonfossilfuelsinnon-producingcountriesareequivalenttotariffs.Whenthesetaxes—which,onaverage,arequitehighinfossilfuel-importingcountries—areaccountedfor,thebiasreverses.Overall,the

CEPIIWorkingPaperCarbonbiasoftariffs:Arefossilfuelstheculprits?

5

smallsizeandhighsensitivityofthecarbonbiassuggestthatharmonizingtariffsacrosssectorsmaynotbeapriorityforclimatepolicy,unliketaxingfossilfuelconsumption(orclosingothertaxloopholeswithenvironmentalrepercussions,ashighlightedby

Iovinoetal.

,

2023

).

Thispaperbuildsonagrowingbodyofresearchexamininghowtradepoliciesimpactenvironmentaloutcomes,especiallyGHGemissions.Whileouranalysisprimarilyfocusesonthecarbonbiasoftradepolicies,thisissuefitswithinabroaderliteratureontherelationshipbetweentradepolicyandsector-specificemissions.Oneareaofsignificantattentionistheinteractionbetweentradepoliciesandagriculturalemissions.Thisbodyofworkprovidesimportantinsightsintohowtradeprotectionandsubsidiesaffectemissionsinagriculture,asectorresponsibleforapproximatelyone-thirdofglobalGHGemissions(

Crippaetal.

,

2021

).Moststudiesonagricultureemphasizethesector'shighborderprotectionandsignificantnon-CO2emissions,highlightingitscrucialroleinshapingtrade-relatedemissions(

Labordeetal.

,

2021

;

Guerreroetal.

,

2022

).Ourpapercontributestothisbroaderliteraturebydemonstratingthattradepoliciesappliedtotheagriculturalsectorarethesecond-largestdriveroftheenvironmentalbiasintradepolicies,withfossilfuelextractionbeingthelargest.Althoughbothsectorsareemission-intensive,theagriculturalsectortendstobehighlyprotected,whereasthefossilfuelssectorhasminimalprotection.Thiscontrastingprotectionstructureresultsinoppositechangesinemissionswhentariffsareharmonizedacrosssectors:anincreaseforagriculturalproductsandadecreaseforfossilfuels.

Anotherrelevantstrandofliteratureexplorestheoptimaltaxationoffossilfuels,particularlyinthecontextofinternationaltrade.Sincethe1960s,studieshaveexaminedtariffsasatooltocapturerentsfromfossilfuelproducers,ofteninthefaceofimperfectcompetition(

Johnson

,

1968

;

Dixit

,

1984

;

Karp

,

1984

;

JonesandTakemori

,

1989

).Morerecently,

Rubio

(

2011

)extendedthislineofresearchtoaccountforthefinitenatureoffossilresources.ThisdiscussionhasgainedrenewedattentioninlightofEUsanctionsonRussia.Rent-extractingtariffsonfossilfuelimportshavebeenpromotedasawaytoreducerelianceonRussianenergywhilemitigatingtheeconomicimpactontheEU(

Gros

,

2022a

,

b

;

Ockenfelsetal.

,

2022

).Additionally,

Tahvonen

(

1995

)andothershaveexploredhowtariffscanreducefossilfuelconsumptionwhileaddressingenvironmentalexternalities.Thus,tariffsonfossilfuelsresemblecarbontaxesintheirpotentialtoinfluencebothrentextractionandenvironmentaloutcomesaninsightweuseinthispaper.

Closertoourstudy,

Shapiro

(

2021

),

KlotzandSharma

(

2023

),and

MoreiraandDolabella

(

2024

)investigatetheexistenceofanenvironmentalbiasintradepolicies.

Shapiro

(

2021

)addressesthisquestionusingtwoapproaches.Thefirstapproachemploysstatisticalmethodstoidentifystylizedfactsabouttherelationshipbetweentradebarriersandemissionintensity,whilethesecondusesmodelingtoquantifytheemissions-relatedimplicationsofremovingtheidentifiedbias.Inhiseconometricanalysis,

Shapiro

(

2021

)findsthatin2007,tradeprotectionwaslowerfordirtiergoods,implyingthatharmonizingtariffscouldreduceCO2emissionsfromfossilfuelcombustion.

MoreiraandDolabella

(

2024

)applyasimilarstatisticalapproach,focusingonLatinAmericaandtheCaribbean.Theyextend

Shapiro

'seconometricanalysisbyincludingallGHGs,

CEPIIWorkingPaperCarbonbiasoftariffs:Arefossilfuelstheculprits?

6

aswellastheagricultureandminingsectors.Theirfindingsindicatethatthebiasishighlyheterogeneousacrosscountries,partlybecausetheyaccountfornon-fossilfuelemissions.Ontheotherhand,

KlotzandSharma

(

2023

)adoptamodelinganalysisfocusingspecificallyonCO2emissionsfromfossilfuelcombustion.Theyfindtheeffectofthetariffbiasonemissionstobesixtimessmallerthanthatreportedby

Shapiro

(

2021

).However,duetodifferencesinmodelingapproachesanddata,theoriginofthesediscrepanciesremainsunclear.Inthispaper,wedifferfrom

Shapiro

byusingupdatedprotectionandinput-outputdatafrom2019insteadof2007,andbyconsideringabroadersetofGHGs,alongwithamoredetailedregionalandsectoralaggregation.Westrivetominimizethesedeparturestopreciselyidentifywhichchoicesaccountforanydifferencesinresults(seeAppendix

B

).

Importantly,wefocussolelyonthecarbonbiasoftariffs,excludingnon-tariffmeasures(NTMs)fromouranalysis.AlthoughNTMscanrepresentsignificanttradebarriers,analyzingtheirpotentialenvironmentalbiasischallenging.First,theheterogeneityofNTMsacrosssectorsmakesitdifficulttocomparetheiradvaloremequivalentsmeaningfully.Differenttypesofmeasures—suchaspesticidemaximumresiduelevelsandautomotiveindustrystandards—operateunderfundamentallydistinctregulatorylogicsthatresiststraightforwardharmonization.Second,evenifNTMsrestricttradebyimposingstandardsthatmaybetoostringentforsomeproducers,itdoesnotnecessarilyimplyareductioninwelfare,particularlyifthesestandardspreventtheexportofpotentiallyharmfulproducts(

DisdierandMarette

,

2010

).Therefore,werefrainfromconsideringthesebarriersinouranalysis,asthecross-sectoralstandardizationofNTMscouldintroduceunintendedriskstopublichealth.

Therestofthepaperproceedsasfollows.Section

2

introducesthedataandpresentskeystylizedfactsontradeprotectionandGHGemissions.Section

3

outlinesthequantitativetrademodelusedfortheanalysis.Section

4

presentsthesimulationresults,highlightingthecontributionofdifferentsectorstothecarbonbias.Section

5

explorestwomodelextensionstorefinetheroleoffossilfuels.Finally,section

6

concludes.

2.Dataandstylizedfacts

2.1.Data

Togeneratethestylizedfactsinthenextsectionandtocalibratethemodelspresentedinsubsequentsections,weusethreemaindatasources.TradedataareobtainedfromtheEXIOBASEworldinput-outputtableversion3.8.2(

Stadleretal.

,

2018

),whichrepresentstheglobaleconomywith43countries,5rest-of-the-worldaggregates,163industries,and7finalusesectors.Dataareavailableforallyearsfrom1995to2021,withprojectionsorprovisionalestimatesforyearsafter2015.Forcomputationalpurposes,weaggregateEXIOBASEinto23regionsand47industries(seeAppendix

D

).

3

3Toensureexactreplicationof

Shapiro

’s(

2021

)resultsinAppendix

B

,wealsousethesameaggregationschemewith10regionsand21industries.

CEPIIWorkingPaperCarbonbiasoftariffs:Arefossilfuelstheculprits?

7

Additionally,allemissiondatafromsourcesotherthanfossilfuelcombustion(e.g.,otherCO2emissions

4

andnon-CO2emissions

5

)comefromEXIOBASEsatellitedata,whichexcludeemissionsassociatedwithland-usechanges.Non-CO2emissionsareconvertedtoCO2-equivalents(CO2eq)usingglobalwarmingpotentialswitha100-yeartimehorizonfromtheIPCCSixthAssessmentReport(

2023

).

TariffdataissourcedfromtheMarketAccessMapHS6(MAcMap-HS6)databasedevelopedatCEPII(

Guimbardetal.

,

2012

)basedonrawdataprovidedbyITC(UNCTAD-WTO).MAcMap-HS6providesadvaloremtariffsforbilateraltradeflowsforvirtuallyallcountriesattheHarmonizedSystem(HS)6-digitlevel.WeusetheversionofMAcMap-HS6thatisusedtoestimatetradeelasticitiesin

Fontagnéetal.

(

2022

).Inthisversion,theprotectionprovidedbytariffratequotasisrepresentedbytheoutsiderate.Weaveragethetariffs,usingtradeweights,toobtainadvaloremtariffequivalentsattheEXIOBASEcountryandsectorallevel.Sinceonlytheyears2007,2010,2013,2016,and2019areavailableinMAcMap-HS6,welinearlyinterpolatetoobtaintheotheryears.

OurthirdmainsourceofdataistheInternationalEnergyAgency(IEA).WeuseIEAfossilproductiondata(

IEA

,

2022

)tocalculatethequantitiesoffossilfuels(primarycoal,primaryoil,andnaturalgas)extractedineachcountry.UsingemissionfactorsfromtheEmissionFactorDatabase(EFDB)from

IPCC

(

2021

),wecalculatedtheCO2embeddedinfossilfuelproduction.FossilCO2emissionsarethustrackedfromthepointoftheirextraction,amethodologicalchoice(identicalto

Shapiro

,

2021

)thatwillbeexplainedinsection

3.3

.

Beyondthethreemaindatasourceslistedabove,weuseafewothers.Weusethesametradeelasticitiesasin

Shapiro

(

2021

).Sinceoursectoralaggregationimpliesmoresectorsthanin

Shapiro

(

2021

),wekeepthesametradeelasticitiesbutapplythemtothemoredetailedsectors.Intheextensionpresentedinsection

5.1

,weusedatafromGTAP(

Aguiaretal.

,

2022

)ontheshareofthenaturalresourcerentsintheproductioncostoffossilfuelextraction.Insection

5.2

forcomplementaryanalysis,weuseOECDNetEffectiveCarbonRates(

OECD

,

2022

)andBPStatisticalReviewofWorldEnergy(

BP

,

2020

).

2.2.Stylizedfacts

Inthissubsection,weprovidesomestylizedfacts.Theycannotreplacethemodel’sresults,astheyareverystylized,aggregatedattheworldlevel,andneglectequilibriumeffects.However,theyareusefulforgaininganintuitiveunderstandingofthepotentialenvironmentalbiasoftradepoliciesandtheassociatedimpactoftariffharmonization.Byharmonization,wemeanthateachcountryestablishes,foreachtradepartner,onlyonetariffvalueforallsectors(bytakingthetrade-weightedaverageofformervalues).Themathematicalformalismwillbeintroducedin

4Cementandlimeproductionprocesses,peatdecay,andwaste.

5Excludingtheemissionsassociatedwithland-usechanges,thecoverageofGHGemissionsiscomprehensive,encompassingCH4,HFCs,PFCs,N2O,andSF6.

CEPIIWorkingPaperCarbonbiasoftariffs:Arefossilfuelstheculprits?

8

section

4

.

Tosimplifythepresentation,wehavegroupedallsectorsintofive:Agriculture,FossilExtraction,BrownIndustries,Manufacturingn.e.s.,andServices.ThecompositionofthesefivesectorsinrelationtoEXIOBASEsectorsisgiveninAppendix

D

.Importantly,theFossilExtractionsectorcontainsonlysub-sectorsextractingfossilsfromthegroundandexcludessectorsrefiningfossils.RefiningsectorsareintheBrownIndustriescategory.Thisdistinctionisimportantbecausetariffsonrefinedfossilscanbehigh,whiletheyareusuallyveryloworevenabsentoncrudeoilandgas.Moreover,fossilsrefinedlocallydonotneednaturalresourcestobeproducedbutaredependentontheavailabilityofcrudefossils.

6

Followingatradepolicylogic,servicesaredefinedasthesectorsthatarenotcoveredbytheHarmonizedSystemoftheWorldCustomsOrganization,whichmeansthattheyincludesomebrownindustriessuchasconstructionandtransport.

7

2.2.1.Sectoraltradeexposureandtariffstructure

Understandingthedistributionoftradeexposureandtariffprotectionacrosssectorsiscrucialforassessingtheenvironmentalbiasintradepolicy.Sincetariffsinfluencesectoralresourceallocation,theycaneitherreinforceorcounteractemissionsincentivesembeddedintradeflows.Varyinglevelsoftradeexposureinteractwiththedifferenttariffstructuresappliedtoeachsector,creatingpotentialenvironmentalimplications.Inthissection,wecategorizesectorsbasedontheirtradeexposureandtariffprotectionlevels,settingthestageforanalyzingthepotentialemissionsimpactoftradepolicyreform.

TradeexposureandsectoralshareSectorsvarysignificantlyintheirexposuretointernationaltrade,shapingtheextenttowhichtariffstructuresinfluenceemissions.Forcoherentcomparisons,wecompareimportswithaggregatesupply,asimportsareoftenreusedindownstreamproducts.Wealsouseaggregatesupplytoaccountforsectorsthatareusedmainlyasintermediateconsumptionbutplayanimportantroleinvaluechains.Table

1

providesanoverviewofimportsasapercentageofaggregatesupplyacrossfivebroadsectoralcategories.FossilExtractionisthemosttrade-exposed,with47%ofitsproductionbeinginternationallytraded,makingitparticularlysensitivetotariffchanges.Manufacturingn.e.s.follows,with29%ofitssupplyimported,accountingfor39%oftotalimports.BrownIndustries,whichincludeemissions-intensivesectorssuchaschemicalsandmetals,exhibitasimilardegreeoftradeexposure,with25%oftheiraggregatesupplyimported.Incontrast,Agricultureremainsrelativelyshieldedfromtrade,withimportsmakinguponly13%ofitssupply.Whiletradeexposuredetermineshowmuchasectorisaffectedbytariffchanges,itsroleinshapingemissionsalsodependsoninitialtarifflevelsandemissionintensity,twopointsweaddressbelow.

6Thiswillbeespeciallyrelevantforamodelextensionanalyzedinsection

5.1

.

7WeclassifytheElectricitysectorinservices,despitetheexistenceofanHScode(271600)forthissector,becausethisheadingisoptionalandtradeinelectricityisgenerallycoveredbyspecialarrangementsthatdonotinvolvetariffs.

CEPIIWorkingPaperCarbonbiasoftariffs:Arefossilfuelstheculprits?

9

Table1–2019descriptivetradestatistics

Sector

Imports%insectoraggregatesupply

Sector%intotalaggregatesupply

Sector%in

totalimport

Average

tariff(%)

Agriculture

13.5

3.5

3.9

6.8

FossilExtraction

47.1

1.5

5.9

0.3

BrownIndustries

25.1

15.3

31.5

1.9

Manufacturingn.e.s.

28.6

16.7

39.1

3.0

Services

3.8

63.0

19.5

0

Sources:EXIOBASEandMAcMap-HS6.

TariffstructureThesevaryinglevelsoftradeexposureinteractwiththedifferenttariffstructuresappliedtoeachsector,creatingpotentialenvironmentalimplications.Overall,thetrade-weightedaveragetariffin2019is2.6%.Amongsectors,Agriculturefacesthehighestaveragetariff(6.8%),reflectinglongstandingprotectionistpolicies.Incontrast,FossilExtractionfacesvirtuallynotariffs(exceptforcoalinsomecountries),despitebeingtheultimatesourceofallCO2emissionsfromfossilcombustion.BrownIndustriesandManufacturingn.e.s.aresubjecttomoderatetariffs,closetotheglobalaverage,especiallyforManufacturingn.e.s.

2.2.2.Emissionintensityacrosssectors

Thefinalimportantparameterinthedecompositionoftheimpactoftariffpolicyonemissionsisthecarboncontentofthesectors.Sectorsarenotonlyveryheterogeneousintariffsbutalsoveryheterogeneousintheircarbonemissions.Becauseofourfocushereonemissionsassociatedwithproduction,thestatisticsdisplayedinthissubsectionexcludethoseemittedatthetimeoffinalconsumption.

WedistinguishCO2emissionsfromfossilfuelcombustionfromotherGHGemissions(CO2emissionsfromprocessesandemissionsofotherGHGs).IfweconsiderCO2emissionsfromfossilfuelcombustion,table

2

showsfirstthatFossilExtractionisthemostemissivesectorperEURofoutput,followedbyBrownIndustries.

8

Agriculture,othermanufacturingindustries,andServicesdisplaymuchlowerCO2emissionintensities.Second,Servicesareresponsibleformorethanhalfoftheemissions.Thisisbecauseofthelargeeconomicsizeofthisaggregate,whichincludes,inoursetup,emission-int