Ricci, Ma ia; Pé ez Domínguez, Ignacio; H is o , Jo dan; Vandyck, Toon; Van
Hou en, S ijn
Wo king Pape
P icing GHG emissions in ag icul u e: Accoun ing o ade
and ai ness o e ec i e clima e policy
JRC Wo king Pape s on Taxa ion and S uc u al Re o ms, No. 10/2024
P o ided in Coope a ion wi h:
Join Resea ch Cen e (JRC), Eu opean Commission
Sugges ed Ci a ion: Ricci, Ma ia; Pé ez Domínguez, Ignacio; H is o , Jo dan; Vandyck, Toon; Van
Hou en, S ijn (2024) : P icing GHG emissions in ag icul u e: Accoun ing o ade and ai ness
o e ec i e clima e policy, JRC Wo king Pape s on Taxa ion and S uc u al Re o ms, No. 10/2024,
Eu opean Commission, Join Resea ch Cen e (JRC), Se ille
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/311131
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by/4.0/
M. Ricci, I. Pe ez Dominguez, J. H is o , T. Vandyck and S. Van Hou en
P icing GHG emissions in ag icul u e:
accoun ing o ade and ai ness o e ec i e clima e policy
JRC Wo king Pape s on Taxa ion and S uc u al Re o ms No 10/2024
JRC139782
Se ille: Eu opean Commission, 2024
© Eu opean Union, 2024
The euse policy o he Eu opean Commission documen s is implemen ed by he Commission Decision 2011/833/EU o 12 Decembe 2011
on he euse o Commission documen s (OJ L 330, 14.12.2011, p. 39). Unless o he wise no ed, he euse o his documen is au ho ised
unde he C ea i e Commons A ibu ion 4.0 In e na ional (CC BY 4.0) licence (h ps://c ea i ecommons.o g/licenses/by/4.0/). This means ha
euse is allowed p o ided app op ia e c edi is gi en and any changes a e indica ed.
Fo any use o ep oduc ion o pho os o o he ma e ial ha is no owned by he Eu opean Union pe mission mus be sough di ec ly om
he copy igh holde s.
How o ci e his epo : Eu opean Commission, Join Resea ch Cen e, Ricci, M., Pe ez Dominguez, I., H is o , J., Vandyck, T. and an Hou en,
S., P icing GHG emissions in ag icul u e: accoun ing o ade and ai ness o e ec i e clima e policy, Eu opean Commission, Se ille, 2024,
JRC139782.
This documen is a publica ion by he Join Resea ch Cen e (JRC), he Eu opean Commission’s science and knowledge
se ice. I aims o p o ide e idence
-based scien i ic suppo o he Eu opean policymaking p ocess. The con en s o his
publica ion do no necessa ily e lec he posi ion o
opinion o he Eu opean Commission. Nei he he Eu opean
Commission no any pe son ac ing on behal o he Commission is esponsible o he use ha migh be made o his
publica ion. Fo in o ma ion on he me hodology and quali y unde lying he da a us
ed in his publica ion o which he
sou ce is nei he Eu os a no o he Commission se ices, use s should con ac he e e enced sou ce. The designa ions
employed and he p esen a ion o ma e ial on he maps do no imply he exp ession o any opinion wha
soe e on he
pa o he Eu opean Union conce ning he legal s a us o any coun y, e i o y, ci y o a ea o o i s au ho i ies, o
conce ning he delimi a ion o i s on ie s o bounda ies.
Con ac in o ma ion
Name:
Ma ia Ricci
Email:
Ma ia. [email protected] opa.eu
EU Science Hub
h ps://
join - esea ch-cen e.ec.eu opa.eu
Execu i e Summa y
The ag icul u al sec o emains la gely excluded om g eenhouse gas (GHG) p icing, despi e being
a majo sou ce o emissions. To align ag icul u e wi h clima e goals, his pape examines wo
p icing mechanisms: supply-side GHG axes on p oduc ion and demand-side axes on
consump ion. We assess hei espec i e impac s on emissions, EU p oduce s' compe i i eness
and dis ibu ional ou comes.
Analysis and Key Findings:
1. En i onmen al Impac and Compe i i eness’ Conce ns: A supply-side GHG ax in
ag icul u e pu s EU a me s a a compe i i e disad an age and esul s in o e 40% ca bon
leakage, as non-EU p oduce s eplace Eu opean ones. On he o he hand, a consump ion-
based GHG ax le els he playing ield, applying equally o EU and o eign p oduc s,
boos ing EU expo s o g eene p oduc s, and gene a ing posi i e spillo e e ec s.
2. Case S udy on VAT Re o m: We he e o e explo e a eal-wo ld e o m based on
adjus ing Value-Added Taxes (VAT) on ood o e lec clima e ex e nali ies. Ou indings
indica e ha , while his e o m can achie e emissions educ ions, i is eg essi e wi hou
complemen a y measu es. Households a he lowe end o he income dis ibu ion a e
disp opo iona ely a ec ed due o hei highe ood expendi u e sha e.
3. Fai ness and Compensa ion Mechanisms: We p opose compensa ion mechanisms,
such as a eeba e sys em o equal-pe -capi a e enue ecycling, o add ess equi y conce ns.
Bo h app oaches imp o e wel a e o he majo i y, wi h a sligh bu den emaining on he
op 20-30% o mea consume s. These measu es also help p ese e equi y and social
accep abili y o he p oposed VAT e o ms.
4. Limi a ions and Policy Implica ions: The VAT-based app oach achie es a modes
educ ion in emissions, as i lacks di ec incen i es o a me s o inno a e and shi s ou pu
o expo ma ke s. Fo g ea e e ec i eness, complemen a y ade measu es and enhanced
suppo o a m-le el inno a ion a e necessa y.
O e all, ou analysis sugges s ha p ice-based measu es, pa icula ly a well-designed VAT e o m
wi h app op ia e compensa ion mechanisms, can help align ag icul u e wi h clima e goals while
add essing compe i i eness and equi y conce ns. Howe e , achie ing signi ican emission
educ ions equi es b oade policy suppo beyond me e p icing ins umen s.
1
P icing GHG emissions in ag icul u e:
accoun ing o ade and ai ness o e ec i e clima e policy
1
Ma ia Ricci*, Ignacio Pe ez Dominguez*, Jo dan H is o *,
Toon Vandyck*¤ and S ijn Van Hou en¤
*Eu opean Commission, Join Resea ch Cen e, Calle Inca Ga cilaso, Se ille, Spain.
¤Depa men o Economics, KU Leu en, Naamses aa 69, 3000 Leu en, Belgium.
Co esponding au ho : Ma ia Ricci (Ma ia[email p o ec ed]opa.eu)
Abs ac
Al hough ag icul u e is an impo an sou ce o g eenhouse gas emissions, he sec o emains ou
o scope o g eenhouse gas (GHG) p icing policies. To align he u u e ood sys em wi h he
ansi ion o ne ze o emissions, wo key ques ions a ise: To wha ex en can ax policies help
achie ing his ansi ion in a ai and e ec i e way? And, would i be p e e able o le y a GHG
ax on he p oduc ion o he consump ion side? We employ an EU ag o-economic model o
compa e p oduc ion and consump ion-side GHG axes and o quan i y hei en i onmen al
impac . We ind ha supply-side p icing in ag icul u e displays leakage a es o o e 40% and
lea es EU p oduce s in a si ua ion o compe i i e disad an age; on he o he hand, demand-side
measu es le el he playing ield in he Single Ma ke and gene a e posi i e leakage as hey boos
he expo o (g eene ) EU p oduce s. Focussing on ou coun ies – Spain, F ance, Romania
and Poland – we he e o e conside a eal-wo ld e o m based on adjus ing Value-Added Taxes
o e lec clima e change ex e nali ies. Using mic osimula ion echniques and household-le el
da a we show ha - while his e o m can gene a e educ ions in emissions - is eg essi e wi hou
complemen a y measu es. Feeba e and equal-pe -capi a e enue ecycling add ess equi y
conce ns and p oduce wel a e gains o he majo i y o he popula ion, while he op 20-30% o
mea consume s expe iences wel a e losses. O e all, indings sugges ha p ice-based measu es
can help align ag icul u e wi h clima e goals bu ade and equi y aspec s should be e lec ed in
policy design.
Keywo ds: clima e policies, ood sys em, VAT e o m, ai ness.
JEL classi ica ion: D6, H2, Q1, Q5
1
This pape was de eloped wi hin he AMEDI p ojec s o he Di ec o a e Gene al o Employmen and Social
A ai s o he Eu opean Commission. Au ho s a e indeb ed wi h Ba Capéau, And é Decos e and So ia Maie o
insigh ul discussions in he ea ly s ages o his wo k ha helped shaping i . Au ho s a e u he mo e hank ul o he
pa icipan s o he 2022 EUROMOD Resea ch Con e ence in P ague who p o ided nume ous and help ul
commen s o imp o e ou analysis. Finally, we kindly acknowledge Da id Klene and An onio F. Amo es o a
ho ough e iew o his wo k and hei aluable sugges ions. All e o s a e ou own.
2
1. In oduc ion
Food sys ems we e esponsible o abou one- hi d o global an h opogenic g eenhouse gas
emissions (GHG) in 2015 (C ippa e al., 2021). In he EU, cu en -policy p ojec ions sugges ha
he sha e o non-CO2 GHG emissions om ag icul u e is oughly o double om jus unde
10% in 2010 o nea ly 20% by mid-cen u y (EC, 2021), as cu en policies imply apid
deca boniza ion o o he sec o s. An ambi ious and e ec i e clima e policy, he e o e, canno
o e look he emissions ela ed o ood. Ye , p ice signals o ood p oduc s in he EU cu en ly
do no e lec co esponding en i onmen al damages, and do no help s ee ing consume s
owa ds die a y pa e ns wi h low g eenhouse gas emissions in ensi y. A he same ime, die a y
choices ha exace ba e clima e change, such as highe consump ion o mea , dai y and o he
p ocessed commodi ies, a e associa ed wi h de imen al consequences o public heal h
(Sp ingmann e al., 2016), de o es a ion, nu ien pollu ion and ecosys em loss and could
indi ec ly cause p ema u e dea hs (Sp ingmann e al., 2023; Himics e al., 2022; Funke e al.
2022).
While he bulk o EU g eenhouse gas emissions is co e ed by he EU emission ading sys em
(including ag eed ex ensions o buildings and oad anspo ), p ice mechanisms emain la gely
absen in ag icul u e, which is co e ed unde he emission a ge s se by he E o Sha ing
Regula ion (ESR). One way o add ess his issue is o le e age p icing o accoun o
en i onmen al damages ela ed o ood consump ion. This idea is no no el, as i has been
discussed (e.g. in he EU Fa m o Fo k s a egy Eu opean Commission, 2020), s udied in policy
(EC, 2023) and academic con ex s (Klene e al., 2023), ecommended by he Eu opean
Scien i ic Ad iso y Boa d (ESABCC, 2024) and ecen ly p oposed o implemen a ion in
Denma k. One o he ad an ages o p ice-based policy measu es, as a ma ke -based ins umen ,
is in o ma ional e iciency: he p ice signal di ec ly ep esen s ex e nali ies, educing he need o
consume s and p oduce s o engage in ime-consuming assessmen s o sus ainabili y c i e ia o
compa ison o ecolabels. As such, p icing ins umen s can o m a s ong componen o a
b oade policy package (Sp ingmann e al., 2018).
Howe e , se e al conce ns s em om he adop ion o p ice measu es. One is he
compe i i eness o he EU ag icul u e sec o . In he absence o a la ge-scale in e na ional
coo dina ion o a ca bon bo de adjus men mechanism co e ing ag icul u e, g eenhouse gas
emission axes on EU p oduce s (i.e. supply-side measu es) isk lea ing hem in a si ua ion o
compe i i e disad an age is-à- is he es o he wo ld. A ela ed aspec is ca bon leakage a ising
as om o eign p oduce s eplacing (g eene ) EU p oduce s in sa is ying domes ic and
in e na ional demand when EU p oduce s a e subjec o GHG emission axes while o eign ones
a e no . In he p esence o ca bon leakage, domes ic e o s o cu b emissions a e pa ly o se by
inc eases in o eign emissions. Finally, clima e-based ood p icing aises conce ns abou ai ness
as well. Food, in gene al, is a necessi y, and p ice inc eases could isk he accessibili y and
a o dabili y o ood, pa icula ly o households a he bo om end o he income dis ibu ion,
who ypically spend a compa ably highe sha e o hei budge on ood p oduc s (Klene e al.,
2023, Maie and Ricci 2023). In addi ion o e hical conce ns o he si ua ion o low-income
households, pe cei ed un ai ness o ax e o ms may hampe hei accep abili y (see Klene e
al., 2018 and Maes e-And es e al. 2019 o a discussion).
3
P ice measu es a ge ed o he demand-side based on consump ion axes migh alle ia e
compe i i eness and ca bon leakage conce ns as hey apply o all p oduc s consumed
domes ically, ega dless o hei o igin. Al hough demand-side measu es may ace challenges
simila o supply-side GHG p icing in e ms o dis ibu ional impac s, ea lie wo k has illus a ed
ha ai ness conce ns can be add essed h ough e enue ecycling (Rausch e al., 2011).
Pa icula ly, i has been a gued ha ‘ eeba es’ (blend o ee and eba e) migh be a cos -e ec i e
ins umen ha enhance accep abili y (Pa y, 2021). Feeba es ep esen a e enue-neu al change
in axes, combining highe ax a es o en i onmen ally ha m ul p oduc s wi h educed ax a es
(o subsidies) o goods o sec o s wi h low emission in ensi y. While eeba e scena ios o ood
p oduc s a e eme ging in he li e a u e (e.g. Sp ingmann e al., 2017), esea ch on he
dis ibu ional impac s o ood p icing e o ms emains sca ce (see Klene , Funke and Cai 2023).
In his pape , we unde ake a wo- old analysis. Fi s ly, we p o ide a compa a i e assessmen o
equi alen demand-side and supply-side GHG p icing quan i ying hei impac on demand,
supply and emissions. We analyse he ex en o which subs i u ion e ec s ia in e na ional ade
as well as emission leakage shape he en i onmen al impac o di e en e o ms as well as hei
e ec s on he compe i i eness o he EU ag icul u al sec o . Secondly, we p esen an assessmen
o a Value Added Tax (VAT) e o m designed o e lec he g eenhouse gas emissions. We
ega d his e o m ha le e ages exis ing VAT a es as a eal-wo ld, implemen able,
co esponden o a ull-blown consump ion-side ax ha ully e lec s GHG emissions
associa ed wi h each ood p oduc . We explo e he dis ibu ional consequences wi hin and ac oss
household income g oups and conside compensa ion mechanisms o o se po en ial eg essi e
impac s.
Ou key indings indica e ha he ha m o he EU ag icul u al sec o ’s compe i i eness and
ca bon leakage esul ing om supply-side p ice measu es would be signi ican unless hese
measu es a e widely adop ed in e na ionally. In pa icula , we es ima e ha an EU-only supply-
side GHG ax o EUR 100 pe onne o ca bon dioxide equi alen ( CO2e) would cause
domes ic p oduc ion o all by abou one- hi d mo e han domes ic consump ion as o eign
p oduce s eplace EU ones. Mo eo e , gi en o eign p oduce s a e gene ally “less g een”, his
would come wi h a ca bon leakage amoun ing o 42% o he emissions educ ion achie ed in he
EU - hence se e ely us a ing i s clima e e o . By con as , ou p oposed VAT ax e o m can
econcile compe i i eness, clima e and equi y conce ns when combined wi h complemen a y
measu es. Indeed, since VAT equally applies o EU and o eign p oduce s selling in he EU, his
e o m would le el he play ield in he Single Ma ke and gene a e posi i e ca bon leakage. F om
an equi y s andpoin , while his measu e is eg essi e, a eeba e o equal-pe -capi a cash ans e
scheme can add ess equi y conce ns. Indeed, hese compensa ion measu es would p oduce
wel a e gains o he majo i y o he popula ion, while he op 20-30% o mea consume s
expe ience small wel a e losses. Howe e , ou esul s u he indica e ha VAT e o ms would
b ing a limi ed 2-3% educ ion in GHG emissions. Tha is because p oduce s shi ou pu o
expo ma ke s and he policy does no o e di ec incen i es o a me s o inno a e.
Acco dingly, complemen a y ade measu es should be conside ed o enhance he en i onmen al
e ec i eness o demand-side p icing.
This analysis goes beyond exis ing li e a u e by aking eal-wo ld policy ins umen s (VAT) as he
poin o depa u e, by conside ing ade impac s, and by p o iding an in-dep h iew o
4
he e ogeneous impac s ac oss di e en households and coun ies. We do so by ep esen ing he
di e si y in exis ing ax sys ems and (emission in ensi ies o ) die a y pa e ns in a b oad
geog aphical scope o he EU, including Spain, F ance, Poland and Romania. These coun ies a e
among he mos popula ed and highes mea consume s Membe S a es while p esen ing a
signi ican deg ee o he e ogenei y in e ms o GDP pe capi a. Mo eo e , hey cu en ly ea u e
compa ably low VAT a es on GHG-in ensi e ood p oduc s (as necessi y goods), he e o e he
en i onmen al p icing o ood p oduc s is likely o cause dis ibu ional conce ns due o
consump ion pa e ns ac oss income g oups. Impo an ly, ou analysis also conside s he
ho izon al equi y aspec . This is key o in o m abou he scope o ad ocacy g oups o mobilize
consume s ha a e homogeneous along non-income dimensions, such as mea -based die a y
pa e ns.
The es o he pape is s uc u ed as ollows. Sec ion 2 in oduces he ma e ial and me hods,
p esen ing he models we employ in ou analysis, he da a unde pinning hem as well as he
cons uc ion o he policy scena ios we conside . Sec ion 3 illus a es he esul s o ou analysis
by discussing he simula ed impac o hese policies on p ices, p oduc ion and emissions.
Fu he mo e, i p o ides an ex ensi e discussion o he dis ibu ional implica ions and policy
design op ions o add ess equi y conce ns. Sec ion 4 p o ides some conclusions.
2. Ma e ial and me hods
Add essing ou esea ch ques ion equi es de ailed modelling o he ag icul u al sec o in each
coun y and i s in e connec ions wi h he es o he wo ld o analyse he impac o any e o m
on domes ic p oduc ion, ade, and emissions. Mo eo e , we need mic oda a on consump ion o
explo e he he e ogeneous e ec s o ou p oposed e o m ac oss households and coun ies.
Finally, we need a ho ough and in eg a ed modelling o he ax and bene i sys ems o e lec he
exis ing consump ion ax policy and o model i s e o m.
In his sec ion, we discuss he me hodology and da a unde pinning ou analysis. In Sec ion 2.1,
we desc ibe he ag o-economic model, CAPRI, which allows us o model he impac o supply-
side and demand-side measu es on he ag icul u al sec o and he emissions i gene a es, c ucially
accoun ing o he ole o in e na ional ade. Alongside his, we p esen ou mic osimula ion
sui e, EUROMOD, which allows us o model he edis ibu i e impac o ou p oposed VAT
e o m in he EU coun ies. In Sec ion 2.2, we o e a discussion on he di e ence be ween
supply-side and demand-side axes on emissions– which we will hen quan i y in ou subsequen
analysis. We hen a gue how ou p oposed VAT e o m can be ega ded as an implemen able,
eal-wo ld policy coun e pa o a ull-blow consump ion ax on GHG emissions on ag icul u al
p oduc s. We he e o e cons uc he ele an VAT e o m scena ios and compensa ion
mechanisms.
Gi en he nume ous shocks occu ing om 2020 o da e, om he COVID-19 pandemic o he
cos -o -li ing c isis, we op ed o model ou e o m scena ios wi h espec o he baseline yea o
2019. In his way, we can abs ac om he impac o hese shocks – and c ucially he policy
esponses o hem which included signi ican p ice measu es (see Amo es e al., 2023) – om
ou esul s.
5
2.1 Modelling amewo k
The CAPRI model
The CAPRI (Common Ag icul u al Policy Regionalised Impac ) modelling sys em is an ag o-
economic, pa ial equilib ium model, whe e he impac s o di e en ag icul u al, clima e change,
en i onmen al and ade policies on ag icul u al p oduc ion, a m p ices, and income, ade as
well as en i onmen al indica o s a e analysed in a compa a i e-s a ic amewo k (B i z and
Wi zke, 2014). The model comp ises wo in e ac ing modules, a supply module o Eu ope and a
global ma ke module whe e bila e al ade and p ices o ag icul u al commodi ies a e compu ed
when hey in e ac in an i e a i e p ocess un il equilib ium is eached. The supply model
conside s a p o i -maximiza ion o ep esen a i e egional a ms in he Eu opean Union and
candida e coun ies, aking in o accoun cons ain s ela ed o land a ailabili y, nu ien balances
o c opping and animal ac i i ies, and policy es ic ions. The ma ke module consis s o a
global mul i-commodi y model o abou 60 p ima y and p ocessed ag icul u al p oduc s,
co e ing 77 coun ies in 40 ading blocks. Bila e al ade lows and a ached p ices a e modelled
based on he A ming on assump ion o quali y di e en ia ion. The beha iou al unc ions in he
ma ke model ep esen supply and demand o p ima y ag icul u al and p ocessed commodi ies
(including human and eed consump ion, bio uel use, impo demand om mul ila e al ade
ela ions), ade balancing cons ain s, and ag icul u al ma ke policy ins umen s ( ha is, impo
a i s, a i a e quo as, p oduce and consume suppo es ima es).
Gi en he de ailed echnical in o ma ion on inpu s and ou pu s, CAPRI allows o compu e
indica o s o GHG emissions such as Ni ous oxide (N2O) and me hane (CH4), based on he
IPCC (2006) ie 2 me hodology. I also akes in o accoun de ailed echnical and managemen -
based GHG mi iga ion op ions o EU ag icul u e (Pé ez Dominquez e al., 2020). GHG
emissions in he ma ke model a e es ima ed based on Food and Ag icul u al O ganiza ion
(FAO) GHG emission in en o ies on a commodi y basis. GHG mi iga ion in non-Eu opean
coun ies is ep esen ed by a change in emission ac o s and a ma ching change in ou pu p ices
o e lec he inc ease in cos , de i ed om mi iga ion cos unc ions om he li e a u e (Lucas
e al., 2007). In e ms o he da abase, he Eu opean da a a e mos ly sou ced om Eu os a ,
while he in e na ional da a a e mos ly om he FAO Supply and U iliza ion Accoun s,
supplemen ed by opic- ela ed sou ces.
EUROMOD and he ex ension o Consump ion Taxes
To es ima e he edis ibu i e impac o he p oposed VAT e o m we use he ecen ex ension
o EUROMOD o Consump ion Taxes.
2
EUROMOD is he Eu opean Union ax-bene i
mic osimula ion model. I combines coun y-speci ic coded policy ules wi h households
mic oda a (mainly om he Eu opean Union S a is ics on Income and Li ing Condi ions
da abase, EU-SILC) o simula e ax liabili ies and cash bene i en i lemen s. EUROMOD
simula ions he e o e conside he ole played by each ax-bene i ins umen and hei
in e ac ions and gene a e disposable household income (i.e. income a e di ec axes and cash
bene i s).
12
All in all, unless supply-side p ice measu es a e widely adop ed in e na ionally, which is di icul
gi en he la ge policy coo dina ion hey equi e and he begga - hy-neighbou incen i es
in ol ed, hey would ha m he compe i i eness o he EU ag icul u al sec o . This loss in
compe i i eness is isible in he EU GHG ax scena io as he all in domes ic p oduc ion
signi ican ly exceeds he all in domes ic consump ion. On he o he hand, ou p oposed VAT
e o m, while p oducing a mo e limi ed educ ion in bee consump ion, main ains EU
p oduce ’s compe i i eness gene a ing a all in p oduc ion la gely in line wi h he one ha i
would p oduce in p esence o in e na ional coo dina ion.
GHG emission impac s
Figu e 3 depic s he educ ion o GHG emissions in he EU and he es o he wo ld gene a ed
by each o he e o ms.
4
Unde a Global GHG ax, cleane p oduc ion echnologies and a
educ ion in ou pu o high-emi ing ood p oduc s d i es down GHG emission in EU by nea ly
20 M CO2e (i.e. million me ic onnes o Ca bon Dioxide equi alen s). Ins ead, unde an EU
GHG ax, his educ ion amoun s o 35 M since, as explained ea lie , he EU GHG ax does no
apply o non-EU p oduce s and he p oduc ion o he emission-in ensi e goods ou side o he
EU (whe e emission e iciency is lowe ) would lead o an emission leakage e ec as o eign
p oduce s eplace EU ones. We es ima e his ca bon leakage o amoun o 42% o he emissions
educ ion achie ed in he EU. Hence a signi ican pa o he EU emission educ ion e o
would be o se , limi ing he e ec i eness o his e o m.
Figu e 3. GHG emission educ ion (M CO2e, compa ed o baseline) o di e en policy designs
No es: absolu e changes in ag icul u al GHG emissions o he EU and Wo ld o di e en policy scena ios as
compu ed in CAPRI. GHG emissions om ag icul u al p oduc ion co e mos o me hane and ni ous oxide
ag icul u e emi ing sou ces. The global GHG ax is es ima ed o educe 913 M CO2e globally.
4
Fo de ailed esul s, see also Table 6 in he appendix o his pape .
13
On he o he hand, an inc ease in consump ion axa ion in he EU gene a es a la ge educ ion
o GHG emissions ab oad han a home and hence implies posi i e spillo e s o he es o he
wo ld. Two mechanisms a e a play. Fi s , educed domes ic demand leads EU p oduce s o shi
hei p oduc s o he expo ma ke , whe e p oduc s do no ace he consump ion ax. As EU
ood p oduc s end o ha e lowe emission in ensi y han hose p oduced ab oad, his educes
non-EU emissions. Second, as he consump ion ax also applies o p oduc s p oduced ou side
he EU bu consumed wi hin he EU, he ax lowe s he EU demand o o eign, high-emission
ood p oduc s such as mea and dai y. Despi e he posi i e spill o e e ec on emissions ou side
o he EU, he EU consump ion ax achie es 41% less o he educ ion in GHG emissions,
compa ed o he wo ldwide emissions educ ion o an EU-only GHG ax. The eason is ha a
consump ion-based ax does no p o ide di ec incen i es o a me s o educe GHG emissions,
as e o s o b ing down a m-speci ic emissions would no be e lec ed in he ax a e. As such,
his measu e b ings smalle p og ess owa ds EU g eenhouse gas emission educ ion goals, e.g.
as speci ied in he Fi - o -55 package. Fu he mo e, compa ed o he ine-g ained consump ion
ax, a coa se VAT e o m ha uses he exis ing a e s uc u e o VAT sys ems limi s he GHG
impac u he by abou one- hi d, sugges ing ha exis ing VAT s uc u es imply a s aigh jacke
ha hampe s he e ec i eness o VAT e o ms o clima e change mi iga ion.
In summa y, EU-only supply-side measu es would achie e la ge wo ldwide emission educ ions
han i s consump ion-side equi alen , o he la e does no o e di ec incen i es o a me s o
inno a e. Howe e , unila e al supply-side axes lea e EU p oduce s in a si ua ion o compe i i e
disad an age o he es o he wo ld equi ing hem o shoulde he whole bu den o emissions
educ ion. On op o i , pa o his emission educ ion would be us a ed by he ca bon
leakage om (mo e pollu ing) o eign p oduce s who would pa ly subs i u e he Eu opean ones.
Finally, we analyze he emission impac s in each coun y in Figu e 4.
5
Ac oss coun ies, he
educ ion in emissions unde he a ious ax measu es is ully consis en wi h ou discussion o
he EU as a whole. Unde he a ious scena ios, emission educ ion is also compa able in he
a ious coun ies– excep in Romania whe e i is signi ican ly smalle . Tha is because he access
o emission-sa ing echnologies in Romania is s ill lowe compa ed o olde EU Membe S a es
leading o lowe emission educ ions. Fo easons al eady discussed, e e ywhe e he
implemen able VAT e o m (i.e. he “EU VAT e o m”) is less e ec i e in cu bing emissions
bu s ill goes a long way in gene a ing he e ec s o a ull-blown consump ion ax on emissions.
5
Fo de ailed esul s, see also Table 5 in he appendix o his pape .
14
Figu e 4. Impac on GHG emissions in he EU and selec ed Membe S a es
No es: changes in ag icul u al GHG emissions o di e en coun ies and he EU as compu ed in CAPRI. Impac s
a e di e en ia ed by di e en ypes o ood commodi ies.
3.2 Equi y conce ns and compensa o y measu es
In absence o la ge-scale in e na ional coo dina ion, demand-side p ice measu es bes econcile
en i onmen al and EU compe i i eness ins ances. None heless, hese measu es a e open o
equi y conce ns. Indeed, ood is a necessi y good ha ep esen s a highe sha e o consump ion
in he baske o lowe -income households (see on his opic, e.g., Maie and Ricci, 2023). As he
ecen cos o li ing c isis has highligh ed (see Amo es e al., 2024), shocks o ood p ices ha e a
eg essi e impac ha disp opo iona ely a ec s poo e households who also ha e li le inancial
oom o shoulde inc eases in he cos o hei consump ion baske s. In his sec ion, we
he e o e explo e he dis ibu ional consequences o ou p oposed VAT e o m on e ical and
ho izon al equi y. We hen conside how ad e se dis ibu i e ou comes can be add essed by
app op ia ely designed compensa o y measu es.
The dis ibu ional implica ions o a VAT e o m
We unde ake ou dis ibu ional assessmen based on a household wel a e me ic measu ed in
e ms o compensa ing a ia ion a cons an quan i ies. In a nu shell, we measu e wel a e
a ia ion in e ms o he ex a eu os needed o buy he same consump ion baske a he
consume p ices in he e o m scena io. Fo a o maliza ion o his concep o wel a e, we e e
o Amo es e al. (2024). He e, i is wo h s essing ha such a wel a e analysis p o ides a i s -
o de app oxima ion o he wel a e e ec s which does no accoun o demand esponses o
changes in consume p ices.
-12.0
-9.0
-6.0
-3.0
0.0
EU Spain F ance Poland Romania
% change ela i e o
baseline Bee O he mea Dai y O he
Global GHG ax
EU GHG ax
EU consump ion ax
EU VAT e o m
EU consump ion ax
EU VAT e o m
Global GHG ax
EU GHG ax
15
We begin conside ing e ical equi y, by analysing he impac o ou p oposed VAT e o m on
household wel a e ac oss income deciles. Deciles a e cons uc ed o indi iduals based on
equi alen household disposable income. Fo each decile, we epo he median impac as well
he sp ead a ound ha median.
6
Resul s a e p esen ed in Figu e 5 wi h compensa ing a ia ion
exp essed as % o baseline disposable income.
7
The e, we can app ecia e ha he impac o ou
p oposed VAT e o m is eg essi e in all coun ies wi h he poo es households shoulde ing he
g ea es bu den. F ench households in he i s decile bea he la ges loss o up o 1.1% in
disposable income, while in all o he coun ies he impac is abou hal o i . To make sense o
hese di e ences, conside he dis ance be ween cu en VAT a es and hose implied by he
VAT e o m, in Table 1. In he case o F ance he a e o VAT implied by he e o m o e he
mos pollu ing ood i ems, such as bee and po k mea , is mo e han 4 imes he cu en one,
whe eas in Spain and in Romania he change is jus hal o i .
Howe e , wi hin income g oups, he die a y pa e ns o households a e likely o a y due o
di e ences in as es, household composi ion, habi s, e c. Acco dingly, ho izon al equi y is also a
po en ial sou ce o conce n. To dig in o his aspec , we p esen in Figu e 6 he wel a e impac by
deciles o bee consump ion as a sha e o o al ood expendi u e. This igu e cap u es he ex en
o he impac o his e o m anking households acco ding o hei exposi ion o i , as
exempli ied by hei consump ion o bee – one o he ood p oduc s whose axa ion is o
inc ease he mos – ela i e o how much hey spend on ood al oge he . By his o de ing,
households in he en h decile a e hose wi h he highes ela i e consump ion o bee and,
he e o e, hose who a e mo e likely o be opposed o he p oposed VAT e o m. Looking a he
igu e, as expec ed he magni ude o he impac inc eases ac oss deciles wi h wo s -a ec ed
households being F ench op mea consume s. Howe e , o he 90% o households in all
coun ies, he excess expendi u e gene a ed by he e o m does no exceed 1.3% o hei
disposable income
All in all, ou dis ibu ional analysis sugges s ha he impac o ou VAT e o m on households
is gene ally small. Al hough wi h a ew excep ions, in he e o m scena io households a e
gene ally ound wi h a wel a e loss co esponding o less han 1.3% o hei disposable income.
Howe e , despi e i s ela i e small magni ude, he inequali y-enhancing di ec ion o he e o m
emains a conce n o a numbe o easons. Fi s ly, because he poo es households a e o en
c edi -cons ained and consume mo e han hey ea n. As such, inc eases in p ices can
immedia ely a ec hei abili y o sus ain hei ma e ial s anda ds. Secondly, because he p esen
e o m canno gene a e, by i sel , all he necessa y educ ions in emissions needed o achie e he
EU clima e goals, meaning i needs o be pa o a b oade package o e o ms. In his con ex ,
ad e se dis ibu ional ou comes migh compound and become –as a whole– e y signi ican .
Add essing he eg essi i y o any in e en ion is he e o e necessa y. In he emaining o ou
discussion, we he e o e conside how he e enues aised by ou p oposed VAT e o m can be
used o econcile clima e and equi y conce ns hanks o app op ia ely designed compensa ion
measu es.
6
Fo his pu pose, we o de wi hin each decile household’s based on he wel a e impac hey su e unde each
e o m. We hen show wi hin each decile he 10 h, 25 h, 50 h, 75 h and 90 h pe cen ile o he a ia ion in wel a e.
7
Fo de ailed esul s, see Table 7 in he Appendix o his pape .
16
Figu e 5: Impac o VAT e o m on CO2e in ensi e ood p oduc s o e households along he income dis ibu ion.
No es: Deciles o indi iduals o de ed on baseline equi alized household disposable income. We show he sp ead o
he indi idual change in household wel a e ela i e o household disposable income o each decile. Household
income is kep ixed a he p e- e o m le el.
Figu e 6: Impac o VAT e o m on CO2e in ensi e ood p oduc s o e households based on hei mea consump ion.
No es: Deciles o indi iduals o de ed on he sha e o household bee consump ion in o al household ood
consump ion. We show he sp ead o he indi idual change in household wel a e ela i e o household disposable
income o each decile. Household income is kep ixed a he p e- e o m le el.
Spain
Poland
F ance
Romania
Spain
Poland
F ance
Romania
-1.2%
-0.9%
-0.6%
-0.3%
0.0%
1 2 3 4 5 6 7 8 9 10 TOT
p10 p25
p50 p75
p90
-1.2%
-0.9%
-0.6%
-0.3%
0.0%
1 2 3 4 5 6 7 8 9 10 TOT
-1.2%
-0.9%
-0.6%
-0.3%
0.0%
1 2 3 4 5 6 7 8 9 10 TOT
-1.2%
-0.9%
-0.6%
-0.3%
0.0%
1 2 3 4 5 6 7 8 9 10 TOT
-1.2%
-0.9%
-0.6%
-0.3%
0.0%
12345678910 TOT
p10 p25 p50
p75 p90
-1.2%
-0.9%
-0.6%
-0.3%
0.0%
12345678910 TOT
-1.2%
-0.9%
-0.6%
-0.3%
0.0%
12345678910 TOT
-1.2%
-0.9%
-0.6%
-0.3%
0.0%
12345678910 TOT
17
Re enue ecycling and compensa o y measu es
The dis ibu ional analysis indica es ha he impac o he VAT e o m g eening ood axa ion is
expec ed o be small. None heless, gi en i s eg essi e na u e and he ac ha households a he
bo om o he income dis ibu ion a e he mos a ec ed, we conside how he e enues om he
e o m can be ecycled o co ec his e ec . Fo his pu pose, we s udy wo e enue ecycling
schemes o add ess e ical equi y conce ns. The i s scheme, which we name “Feeba e”,
consis s in combining he inc eased VAT a es on high emission-in ensi e ood consump ion –
unde ou VAT e o m scena io - wi h a dec ease in VAT a es on low-emission ood p oduc s.
A second scheme, which we name “lump sum”, simply consis s in ans e ing back o
households e enues om inc eased VAT h ough lump sum ans e .
8
Bo h hese e o ms will
be e enue-neu al by cons uc ion. Thei impac s on e ical equi y a e p esen ed in Figu e 7,
which shows he e ec s ac oss he income dis ibu ion o bo h e enue ecycling op ions.
9
On
he le -hand side he eeba e scheme is conside ed, whe eas on he igh -hand side, we show he
lump sum ans e .
While esul s sligh ly a y ac oss he ou coun ies, we gene ally ind ha he eeba e sys em
la ens ou he losses ac oss income g oups o ze o, while he lump sum ans e p oduces a
small edis ibu i e e ec in a ou o he poo es households. Again, we explo e whe he hese
a e ages mask la ge he e ogenei y wi hin income g oups. Acco dingly, as in he p e ious
sec ion, we analyse he sp ead wi hin deciles o he impac o he e o ms. The he e ogenei y in
he impac on he wel a e o bo h e o ms, ela i e o disposable income, is la ge o he low
deciles. None heless, his emains close o ze o unde bo h compensa ion schemes. We conclude
ha bo h ypes o measu es can la gely add ess he small bu ad e se dis ibu ional ou come
gene a ed by ou p oposed VAT e o ms. In sum, equi y conce ns can be add essed whene e
his e o m is accompanied by app op ia ely designed compensa o y measu es.
As an icipa ed, he main pu pose o he compensa o y measu es is o add ess he wel a e o low-
income households and neu alise any eg essi e e ec o he VAT e o m. Howe e , he impac
on ho izon al equi y is also wo h explo ing o ga he a sense o he poli ical esis ance his
e o m migh encoun e . In, we show he impac o he VAT e o m on households o de ed by
ela i e mea consump ion in he p esence o bo h ypes o compensa o y measu es. The panels
on he le co espond o he eeba e scena io, while he panels on he igh co espond o he
lump sum compensa ion.
E en hough we ind in all coun ies ha hose ha mos bee consume, ela i e o o al ood
expendi u e, s ill expe ience ne losses, hei wel a e loss is dec eased because o he
compensa o y measu es. Fo example, in Spain, he median wel a e loss o he highes 10% o
bee consume s is limi ed o 0,2% o baseline disposable income in he eeba e scena io and
0,1% in he lump sum scena io, whe eas i was 0,3% wi hou compensa ion. None heless, he
ne e ec o he VAT e o m and bo h compensa ion measu es is s ill a edis ibu ion om
hose households consuming mo e bee owa ds households consuming less bee - wi h he op
20-30% o mea consume s expe iencing small wel a e losses. Fo his eason, he main
8
The lump sum ans e is designed o be gi en o each indi idual, including child en.
9
Fo de ailed esul s, see Table 8 in he Appendix o his pape .
18
opposi ion owa ds g eening he ood ax sys em migh be expec ed om hose households
consuming CO2e-in ensi e ood p oduc s, o he ad ocacy g oups ha ep esen hem.
Figu e 7: Impac o VAT e o m and compensa o y measu es o e households ac oss he income dis ibu ion.
No es: Deciles o indi iduals based on baseline equi alized household disposable income. We show he sp ead o
he indi idual change in household sa ings ela i e o baseline household disposable income o each decile.
Feeba e
Lump sum
Spain
F ance
Poland
Romania
-0.6%
-0.3%
0.0%
0.3%
0.6%
1 2 3 4 5 6 7 8 9 10 TOT
p10 p25
p50 p75
p90
-0.6%
-0.3%
0.0%
0.3%
0.6%
1 2 3 4 5 6 7 8 9 10 TOT
-0.6%
-0.3%
0.0%
0.3%
0.6%
1 2 3 4 5 6 7 8 9 10 TOT
-0.6%
-0.3%
0.0%
0.3%
0.6%
1 2 3 4 5 6 7 8 9 10 TOT
-0.6%
-0.3%
0.0%
0.3%
0.6%
1 2 3 4 5 6 7 8 9 10 TOT
-0.6%
-0.3%
0.0%
0.3%
0.6%
1 2 3 4 5 6 7 8 9 10 TOT
-0.6%
-0.3%
0.0%
0.3%
0.6%
1 2 3 4 5 6 7 8 9 10 TOT
-0.6%
-0.3%
0.0%
0.3%
0.6%
1 2 3 4 5 6 7 8 9 10 TOT
19
Figu e 8: Impac o VAT e o m and compensa o y measu es o e households based on hei bee consump ion.
No es: Deciles o indi iduals o de ed on he sha e o household bee consump ion in o al household ood
consump ion. We show he sp ead o he indi idual change in households’ wel a e ela i e o household disposable
income o each decile. Household income is kep ixed a he p e- e o m le el.
Feeba e
Lump sum
Spain
F ance
Poland
Romania
-0.9%
-0.6%
-0.3%
0.0%
0.3%
1 2 3 4 5 6 7 8 9 10 TOT
p10 p25
p50 p75
p90
-0.9%
-0.6%
-0.3%
0.0%
0.3%
1 2 3 4 5 6 7 8 9 10 TOT
-0.9%
-0.6%
-0.3%
0.0%
0.3%
1 2 3 4 5 6 7 8 9 10 TOT
-0.9%
-0.6%
-0.3%
0.0%
0.3%
1 2 3 4 5 6 7 8 9 10 TOT
-0.9%
-0.6%
-0.3%
0.0%
0.3%
1 2 3 4 5 6 7 8 9 10 TOT
-0.9%
-0.6%
-0.3%
0.0%
0.3%
1 2 3 4 5 6 7 8 9 10 TOT
-0.9%
-0.6%
-0.3%
0.0%
0.3%
1 2 3 4 5 6 7 8 9 10 TOT
-0.9%
-0.6%
-0.3%
0.0%
0.3%
1 2 3 4 5 6 7 8 9 10 TOT
20
4. Conclusions
EU clima e policy has made subs an ial ad ances in add essing emissions ela ed o ene gy use.
Reaching clima e neu ali y by 2050, howe e , calls o addi ional measu es o b ing down
emissions ela ed o ood consump ion. Wi h socio-poli ical challenges o EU-wide and supply-
side measu es, go e nmen s may u n o domes ic, demand-side policies o add ess emissions
om ag icul u e. We p o ide a model-based assessmen o eme ging ques ions on he
e ec i eness and equi y o such measu es, ela ed o ca bon leakage (in e na ional ade),
compe i i eness o EU a ms, and social accep abili y (dis ibu ional impac s) o e o ms,
espec i ely.
We ind ha a well-designed VAT e o m, ha le e ages he exis ing EU VAT sys em o e lec
he g eenhouse gas in ensi y o ood p oduc s, can gene a e educ ions in emissions in he EU
and in he es o he wo ld while p ese ing EU p oduce s' compe i i eness ab oad. This is in
con as o supply-side measu es, such as a ca bon ax on p oduce s, which can ha m he
compe i i eness o he EU ag icul u al sec o and lead o signi ican ca bon leakage when
adop ed unila e ally and wi hou complemen a y ade measu es. While being based on a eal-
wo ld ins umen , he VAT e o m s udied in his pape achie es wo- hi ds o he emission
educ ions o a heo e ical ine-g ained consump ion ax based on GHG con en . A he same
ime, demand-side p icing based on GHG con en alued a EUR 100 pe CO2e b ings only a 2-
3% educ ion in GHG emissions as p oduce s shi ou pu o expo ma ke s and he policy does
no o e di ec incen i es o a me s o inno a e. Achie ing la ge emission educ ions h ough
demand-side p icing would hus equi e amping up he ax a es.
None heless, his measu e isks aising equi y conce ns due o he high sha e o ood
consump ion in he budge o low-income households. Ou dis ibu i e analysis con i ms ha
he expec ed impac o his e o m – while small – is expec ed o be eg essi e. We, he e o e,
conside wo e enue ecycling schemes, a eeba e scena io, and a lump sum ans e scena io, o
add ess he eg essi e na u e o he VAT e o m. We ind ha bo h hese mechanisms can
la gely add ess he small bu ad e se dis ibu ional ou come gene a ed by ou p oposed VAT
e o ms, ensu ing ha equi y conce ns a e add essed whene e his e o m is accompanied by
app op ia ely designed compensa o y measu es. Ho izon al equi y conce ns emain, al hough
ou quan i ica ion sugges s ha hese a e somewha concen a ed in he op 30% o mea
consume s. These esul s indica e ha obs acles o implemen a ion based on gene al, income-
based ai ness pe cep ions could be easily o e come, whe eas opposi ion o e o m is mo e likely
o a ise when ad ocacy g oups a e able o mobilize consume s ha a e homogeneous along non-
income dimensions, such as mea -based die a y pa e ns.
Ou assessmen con ibu es o he li e a u e on he e ec i eness and accep abili y o clima e
policy, pa icula ly o he wo k on demand-side measu es, in e na ional ade and ca bon leakage,
and g eenhouse gas mi iga ion in ag icul u e and he ood sys em. Mo e b oadly, ou s udy
p o ides a model-based e alua ion o conc e e policy measu es ha can help guide ou economy
o a sa e and jus space o humani y (Rocks öm e al., 2023). As such, his pape p o ides an
illus a ion o ‘Doughnu Economics’ (Rawo h, 2017) in ac ion, by assessing he po en ial o
adjus ing exis ing policy in as uc u e (VAT axes) o limi global wa ming while conside ing
equi y.
21
Se e al aspec s go beyond he scope o his pape bu wa an u he a en ion in ollow-up
esea ch. Fi s , ou esul s indica e ha in e na ional ade plays an impo an ole in he
e ec i eness o policy measu es, bu he scena ios do no conside changes in ade policy. This
could be conside ed in u u e wo k o con ibu e o he discussion on how ade policy can
suppo en i onmen al goals (Jakob e al., 2022). Second, en i onmen al damages o he ood
sys em go beyond clima e change, as ag icul u e also con ibu es o ai and wa e pollu ion as
well as biodi e si y loss. Heal h impac s induced by he p oposed VAT e o m is also impo an
aspec o be conside ed in u u e wo k. B oadening he lens beyond clima e would in o m
policies ha aim o p og ess on mul iple sus ainable de elopmen goals. Thi d, dis ibu ional
impac s on he income side a e no conside ed he e. Ye , he po en ial impac s on a me s a e
c ucial o he accep abili y and equi y o policy e o ms and dese e u he s udy.
