Beh inge , Jan; End es, Lukas; Ko sinnek, Maike
Wo king Pape
Cos pe cep ions and he suppo o ca bon p icing
IMK Wo king Pape , No. 226
P o ided in Coope a ion wi h:
Mac oeconomic Policy Ins i u e (IMK) a he Hans Boeckle Founda ion
Sugges ed Ci a ion: Beh inge , Jan; End es, Lukas; Ko sinnek, Maike (2025) : Cos pe cep ions and
he suppo o ca bon p icing, IMK Wo king Pape , No. 226, Hans-Böckle -S i ung, Ins i u ü
Mak oökonomie und Konjunk u o schung (IMK), Düsseldo
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/333495
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/legalcode
WORKING PAPER
No. 226 • Augus 2025 • Hans-Böckle -S i ung
COST PERCEPTIONS
AND THE SUPPORT
FOR CARBON PRICING
Jan Beh inge
1
, Lukas End es
2
, Maike Ko sinnek
3
ABSTRACT
We examine how pe cep ions abou he cos s o ca bon p icing a ec policy
accep ance. Using a ep esen a i e sample o he Ge man popula ion, we conduc
expe imen s ha p o ide andomly selec ed esponden s wi h pe sonalized in o ma ion
abou hei cos s a he cu en ca bon p ice o a highe u u e p ice. Pa icipan s end
o o e es ima e hei cu en cos s and inc ease hei ca bon p ice accep ance when
ecei ing cos in o ma ion. In con as , esponden s unde es ima e u u e cos s and
educe hei suppo once hey lea n abou ac ual cos s. This unde sco es he
impo ance o pe sonalized in o ma ion in os e ing cu en suppo o ca bon p icing,
while cau ioning agains po en ial backlash as p ices ise.
—————————
1 Mac oeconomic Policy Ins i u e (IMK). Email: jan-beh inge @boeckle .de
2 Mac oeconomic Policy Ins i u e (IMK). Email: lukas-end es@boeckle .de
3 Mac oeconomic Policy Ins i u e (IMK). Email: maike-ko sinnek@boeckle .de
Cos pe cep ions and he suppo o ca bon p icing
*
Jan Beh inge Lukas End es Maike Ko sinnek
Augus 25, 2025
Abs ac
We examine how pe cep ions abou he cos s o ca bon p icing a ec policy accep-
ance. Using a ep esen a i e sample o he Ge man popula ion, we conduc expe -
imen s ha p o ide andomly selec ed esponden s wi h pe sonalized in o ma ion
abou hei cos s a he cu en ca bon p ice o a highe u u e p ice. Pa icipan s
end o o e es ima e hei cu en cos s and inc ease hei ca bon p ice accep ance
when ecei ing cos in o ma ion. In con as , esponden s unde es ima e u u e
cos s and educe hei suppo once hey lea n abou ac ual cos s. This unde sco es
he impo ance o pe sonalized in o ma ion in os e ing cu en suppo o ca bon
p icing, while cau ioning agains po en ial backlash as p ices ise.
JEL Classi ica ion: D12, D83, H23, Q58
Keywo ds: Ca bon p icing, policy accep ance, pe cep ions, expe imen
*
Jan Beh inge , Mac oeconomic Policy Ins i u e (IMK), Geo g-Glock-S aße 18, 40474 D¨usseldo , Ge -
many. E-Mail: jan-beh inge @boeckle .de. Lukas End es, Mac oeconomic Policy Ins i u e (IMK)
and he Ins i u e o Socio-Economics, Uni e si y o Duisbu g-Essen, Lo ha s aße 65, 47057 Duis-
bu g, Ge many. E-Mail: lukas-end es@boeckle .de. Maike Ko sinnek, Mac oeconomic Policy Ins i u e
(IMK) and he Ins i u e o Economics, Uni e si y o Bambe g, Feldki chens . 21, 96045 Bambe g,
Ge many. E-Mail: maike-ko sinnek@boeckle .de. We hank Elena F anko o excellen esea ch as-
sis ance. This s udy was egis e ed in he Ame ican Economic Associa ion’s egis y o andomized
con olled ials unde ID AEARCTR-0012808.
1. In oduc ion
Limi ing global wa ming equi es e ec i e clima e mi iga ion policies. Howe e , public
opposi ion emains a majo obs acle o implemen ing ambi ious measu es. Despi e being
widely endo sed by economis s as a cos -e ec i e ool o educing emissions (Clima e
Leade ship Council, 2019), ca bon p icing has encoun e ed pa icula ly s ong esis ance.
In ea lie ins ances, public opposi ion - d i en by conce ns abou ising consume p ices -
has limi ed he scope o ca bon p icing schemes and e en led o hei e e sal (Douenne
and Fab e, 2022; C owley, 2021; Ande son e al., 2023). Inc easing public accep ance has
hus become a key challenge.
In Ge many, a ca bon p icing scheme o he building and anspo a ion sec o s was
in oduced in 2021 wi h ixed p ices, ye public accep ance emains low. Wi h he planned
in eg a ion o he na ional emissions ading sys em in o he Eu opean Union Emissions
T ading Sys em (EU ETS-II) by 2027, ca bon p ices a e expec ed o ise subs an ially due
o he ansi ion o ma ke -based p icing and a con inuously dec easing cap on emission
allowances. These an icipa ed p ice hikes will likely inc ease he inancial bu den on
p i a e households and isk in ensi ying conce ns abou pe sonal cos s. Unde s anding
indi idual cos pe cep ions and how hey shape public a i udes is he e o e essen ial
o designing e ec i e communica ion s a egies and policy measu es ha can ensu e he
long- e m iabili y o ca bon p icing.
This pape examines how belie s abou pe sonal cos s in luence he accep ance o ca bon
p icing. Speci ically, we add ess wo key ques ions: Fi s , how accu a e a e indi iduals’
pe cep ions o hei pe sonal cos s o ca bon p icing, and do hey co ec ly an icipa e
he inancial implica ions o u u e p ice hikes o hei own households? Second, does
p o iding pe sonalized in o ma ion abou he ac ual cos s o ca bon p icing a ec a i udes
owa d he policy?
To his end, we conduc ailo ed in o ma ion p o ision expe imen s o es o he causal
e ec o indi idual cos pe cep ions on ca bon p ice accep ance. Ou andomized expe -
imen s a e embedded in an online su ey on a la ge sample o 4,759 esponden s ha is
ep esen a i e o he adul Ge man popula ion. Da a collec ion ook place sho ly a e
1
an unan icipa ed p ice inc ease ha ecei ed widesp ead media co e age, making pe -
sonal cos s especially salien . Ou expe imen s p oceed as ollows: We i s elici people’s
accep ance o paying a ca bon p ice. Subsequen ly, esponden s a e andomly assigned o
one o wo ques ions abou hei pe cei ed addi ional cos s o ca bon p icing, based on
hei cu en ene gy consump ion, and hei unce ain y ega ding his es ima ion, ei he
o he cu en p ice o e45 o a p ojec ed p ice o e200 pe on o CO2. To gene a e
exogenous a ia ion in belie s, we p o ide andom subsamples o esponden s wi h pe -
sonalized in o ma ion abou hei ac ual cos s o ca bon p icing, which we calcula e based
on p e iously epo ed household cha ac e is ics, including de ailed in o ma ion on hei
ene gy use. Finally, we e-elici ca bon p ice accep ance o all esponden s.
We i s documen a se ies o s ylized ac s abou people’s a i udes owa d ca bon
p icing and hei cos pe cep ions: The majo i y o ou esponden s ejec s he policy (54
pe cen ) and accep ance is pa icula ly low among hose who pe cei e hei cos s o be
high. A he same ime, mos indi iduals a e no well in o med abou hei cu en and
p ojec ed u u e pe sonal cos s. A ound 64 pe cen o e es ima e wha hey cu en ly pay,
while app oxima ely 72 pe cen unde es ima e hei p ojec ed u u e cos s. On a e age,
ou esponden s o e es ima e hei cu en annual cos s by e206.4 and unde es ima e
u u e cos s a he highe p ice by e296.4.
The main inding o ou pape is ha pe sonalized in o ma ion abou he cos s o ca bon
p icing signi ican ly in luences accep ance o he policy. On a e age, indi iduals who
o e es ima e hei pe sonal cos s inc ease hei ca bon p ice accep ance when ecei ing
cos in o ma ion, while hose who unde es ima e hei cos s become less suppo i e. The
e ec is s onge o hose wi h la ge ini ial mispe cep ions and g ea e unce ain y abou
hei cos es ima es, which is consis en wi h he no ion ha he p o ided in o ma ion may
be mo e aluable o ex an e less in o med esponden s. The agg ega e implica ions o
policy accep ance di e ma kedly be ween he cu en and he p ojec ed p ice expe imen .
Since mos indi iduals o e es ima e hei cu en cos s, he ne e ec o pe sonalized cos
in o ma ion on accep ance is posi i e. In con as , because he majo i y unde es ima e
he inancial impac o he p ojec ed p ice, in o ma ion p o ision leads o a decline in
2
accep ance.
Impo an ly, we ind ha pe sonalized in o ma ion abou cu en cos s is pa icula ly
e ec i e in inc easing suppo among indi iduals who ini ially opposed ca bon p icing.
This highligh s he po en ial o build b oade public suppo by co ec ing cu en cos
mispe cep ions. O e all, ecei ing pe sonalized in o ma ion abou cu en cos s inc eases
he p obabili y o inding ca bon p icing accep able by 4.5 pe cen age poin s. In con as ,
in o ma ion abou p ojec ed u u e cos s educes suppo ac oss he boa d, ega dless o
indi iduals’ ini ial s ance. This esul s in a 9.6 pe cen age poin inc ease in he likelihood
o ejec ing ca bon p icing, sugges ing ha u u e in o ma ion shocks may ein o ce public
opposi ion.
We addi ionally iden i y demog aphic g oups ha d i e hese e ec s. While in o ma-
ion abou cu en cos s inc eases accep ance ac oss bo h go e nmen and opposi ion
suppo e s, in o ma ion abou p ojec ed u u e cos s disp opo iona ely educes suppo
among hose a ilia ed wi h he go e ning pa ies. This highligh s he isk ha u u e
cos shocks may e ode he poli ical ounda ion o ca bon p icing. O he han ha , he -
e ogenei y in ea men e ec s is la ges based on esponden s’ inancial si ua ion and
exposu e o ca bon p icing, which aligns wi h sys ema ic a ia ions in p e- ea men be-
lie s. The ea men inc eases accep ance pa icula ly among less a ec ed indi iduals who
s ongly o e es ima e hei cos s a cu en p ices, while u u e cos in o ma ion educes
suppo among he mo e a ec ed who subs an ially unde es ima e he inancial impac o
p ojec ed p ice inc eases.
Las ly, we e alua e he ex e nal alidi y o ou indings and demons a e hei o-
bus ness ega ding su ey- ela ed esponse biases, such as expe imen e demand, su ey
a igue, o dis us in he p o ided in o ma ion.
Ou s udy con ibu es o a g owing li e a u e on he de e minan s o a i udes owa d
clima e policy (D ews and Van den Be gh, 2016; Be gquis e al., 2022) and, mo e speci -
ically, suppo o ca bon p icing (Ca a ini, Ca alho and Fankhause , 2018; Maes e-
And ´es e al., 2019; Klene e al., 2018). Among o he ac o s, conce ns abou he speci ic
design ea u es, such as he e ec i eness in educing emissions, dis ibu ional ai ness, and
3
pe sonal cos s a e o en ci ed as shaping public accep ance o ca bon p icing (Maes e-
And ´es e al., 2019; Ca a ini e al., 2017; Ca a ini, Ca alho and Fankhause , 2018).1
The e is mixed e idence on he ( ela i e) impo ance o hese ac o s in explaining
public suppo o clima e policy. Some s udies emphasize inancial sel -in e es , showing
ha suppo co ela es wi h p oxies o indi idual a ec edness and exposu e o cos s (see,
o example, G oh and Ziegle , 2018; Somme e al., 2022). O he s a gue ha en i on-
men al and ai ness conce ns ma e mo e han pe sonal cos s (Kallbekken and Sælen,
2011; Be gquis e al., 2022). In Ge many, whe e ca bon p ice accep ance is ela i ely
low, c oss-coun y s udies on hypo he ical clima e policies highligh he p ominence o
pe sonal cos s and belie s abou household-le el impac s om ax-and-di idend schemes
as s ong co ela es o opposi ion (Dabla-No is e al., 2023; Dechezlep ˆe e e al., 2025).
These indings align wi h expe imen al e idence ha consis en ly shows how cos - ela ed
in o ma ion can causally in luence suppo o clima e policy (Schwa z e al., 2024; Dabla-
No is e al., 2023; Dechezlep ˆe e e al., 2025; Douenne and Fab e, 2022).
Mos p io expe imen al s udies on he suppo o ca bon p icing implici ly ely on
in o ma ion gaps o aising he salience o cos s. In con as , we explici ly documen and
accoun o indi idual cos mispe cep ions by p o iding pe sonalized cos in o ma ion o
manipula e belie s.2The e o e, ou app oach ela es closely o ecen wo k on hypo he ical
ca bon ax-and-di idend schemes ha examines how (inco ec ) indi idual belie s ha
he own household would be a ne inancial lose de e mines policy suppo (Douenne
and Fab e, 2022) and e idence on he ole o low public awa eness and unde es ima ion
o clima e eba e amoun s o he suppo o ca bon p icing (Mildenbe ge e al., 2022).
1A ela ed s and o he li e a u e addi ionally in es iga es he ole o e enue use and ea ma king o
ca bon p ice accep ance (Ba anzini and Ca a ini, 2017; Sælen and Kallbekken, 2011; Somme e al.,
2022; Kaes ne e al., 2023; Beise -McG a h and Be naue , 2019). Maes e-And ´es e al. (2019) and
Klene e al. (2018) p o ide e iews o he li e a u e on he ole o e enue use o policy suppo .
2Me hodologically, ou s udy is ela ed o he li e a u e ha in es iga es he d i e s o policy p e e -
ences by expe imen ally manipula ing belie s. Fo a comp ehensi e e iew o he li e a u e u ilizing
in o ma ion p o ision expe imen s, see Haaland e al. (2023). Ou app oach mos closely esembles
o he ailo ed in o ma ion p o ision expe imen s ha p o ide pa icipan s wi h cus omized in o ma ion
based on hei pe sonal cha ac e is ics (Ro h e al., 2022; Kuziemko e al., 2015; C uces e al., 2013;
Ka adja e al., 2017; H idbe g e al., 2023) o pee -g oup in o ma ion (Ca d e al., 2012; Zimme mann,
2020). Thema ically, we con ibu e o a b oade li e a u e ha s udies sel -in e es ed poli ical p e e -
ences (Haaland and Ro h, 2020; Kuziemko e al., 2015; Ka adja e al., 2017; S an che a, 2021; Fanghella
e al., 2023; Kaes ne e al., 2025).
4
P io wo k sugges s ha ini ial mispe cep ions may be co ec ed ollowing policy imple-
men a ion, po en ially inc easing public suppo o e ime (Konc e al., 2022).3Howe e ,
in line wi h o he s udies on eal-wo ld clima e policies (Mildenbe ge e al., 2022), we
show ha widesp ead mispe cep ions pe sis pos implemen a ion. Mo eo e , we ind ha
p e alen cos o e es ima ions a cu en p ices canno be ex apola ed o u u e p ice de-
elopmen s. Ins ead, he subs an ial unde es ima ion o cos s a highe p ojec ed p ices
sugges s ha p ospec i e in o ma ion shocks may, i any hing, s eng hen opposi ion o
ca bon p icing. This aligns wi h p io e idence ha highe p ice le els a e associa ed
wi h lowe suppo (Somme e al., 2023). Ou esul s indica e ha his nega i e e ec
may be u he exace ba ed by o e ly op imis ic belie s abou u u e inancial impac s. A
key inno a ion o ou s udy is his o wa d-looking pe spec i e: While exis ing esea ch
on cu en o hypo he ical policies consis en ly emphasizes ha cos - ela ed in o ma ion
can inc ease accep ance, we show ha such in o ma ion can also educe suppo - aising
conce ns abou he du abili y o clima e policy unde ising p ices.
The emainde o his pape is o ganized as ollows. In Sec ion 2, we desc ibe ou
su ey and sample as well as he design o ou in o ma ion p o ision expe imen s. Sec ion
3 documen s ou esponden s’ p e- ea men a i udes owa d ca bon p icing and hei
pe cep ions o pe sonal cos s. In Sec ion 4, we p esen he esul s on he causal e ec o
ailo ed cos in o ma ion on ca bon p ice accep ance, including a he e ogenei y analysis
o ou ea men e ec s, and se e al obus ness checks. Sec ion 5 concludes.
2. Da a
We use da a om a ep esen a i e online panel o he Ge man adul popula ion, collec ed
in collabo a ion wi h GapFish, a Ge man ma ke esea ch company. The su ey ook
place be ween 15 Janua y 2024 and 7 Feb ua y 2024, immedia ely a e he CO2 p ice in
Ge many was inc eased om e30 o e45 pe on o CO2. This p ice inc ease exceeded
3Fo policies beyond ca bon p icing, Schui ema e al. (2010) show ha accep ance o a conges ion cha ge
in S ockholm inc eased pos -implemen a ion as cos conce ns diminished. Simila ly, Ca a ini, Ba anzini
and Lali e (2018) ind ha accep ance o a ga bage ax in Swi ze land ose a e ini ial ai ness and
e ec i eness conce ns we e co ec ed.
5
he o iginally announced e40 and ecei ed widesp ead media a en ion.
We use a quo a sampling app oach o ensu e ha he sample ep esen s he adul
Ge man popula ion in e ms o age, gende , egion, and household income. Table A1
in he Appendix shows ha i closely esembles he Ge man mic ocensus, an o icial,
na ionally ep esen a i e da ase , ac oss key cha ac e is ics.4Addi ional in o ma ion on
he mic ocensus is p o ided in Appendix B.1. Fu he mo e, ou esponden s’ epo ed
pa y a ilia ions a e closely aligned wi h con empo aneous na ional polling da a. This
sugges s ha poli ical p e e ences a e well ep esen ed in ou sample, minimizing conce ns
abou poli ical bias in luencing a i udes owa d clima e policy.
In he ollowing, we ou line he s uc u e o he su ey, desc ibe he sample es ic ions,
and p esen i s main cha ac e is ics. We hen de ail he design o ou in o ma ion p o ision
expe imen s and explain how we measu e he key a iables used in he empi ical analysis.
2.1. Su ey and sample
The su ey begins wi h ques ions on esponden s’ age, gende , ede al s a e o esidence,
and household income. We hen collec in o ma ion o es ima e household-speci ic cos s
o ca bon p icing, equi ed o calcula e he ea men in o ma ion o ou su ey expe -
imen s. These a iables include household size, homeowne s a us (owne s. en e ),
li ing space, ene gy sou ces o space hea ing and ho wa e (i.e. elec ici y, gas, oil,
solid uels, o o he enewables), he numbe o gasoline o diesel- un ehicles and annual
mileage.
Ou su ey expe imen s s a by asking all esponden s abou hei accep ance o ca bon
p icing. Responden s a e hen andomly assigned o one o wo ques ions ha measu e
hei ex an e in o medness abou hei addi ional cos s o ca bon p icing, ei he o he
cu en p ice o e45 pe on o CO2 o a p ojec ed p ice o e200 pe on o CO2. Subse-
quen ly, hal o all esponden s a e andomly selec ed o ecei e pe sonalized in o ma ion
on hei ac ual cos s o ca bon p icing. Finally, we e-elici ca bon p ice accep ance o
4The sample closely ma ches he mic ocensus in e ms o age, gende , and egion. Howe e , ou espon-
den s’ household incomes a e somewha lowe on a e age.
6
4. The causal e ec o ailo ed cos in o ma ion
We now p esen esul s om he expe imen s, which p o ide esponden s wi h pe sonal-
ized in o ma ion abou hei household’s cu en and p ojec ed cos s o ca bon p icing.
We i s p e iew key p e- and pos - ea men a iables ele an o unde s anding accep-
ance e isions and hen o mally es ima e a linea model o ca bon p ice accep ance,
le e aging he exogenous a ia ion in cos pe cep ions gene a ed by ou ea men s. We
explo e a ious aspec s o ea men he e ogenei y and close wi h a discussion o he
obus ness o ou indings.
4.1. Desc ip i e e idence
Table 1 p esen s summa y s a is ics o key a iables ac oss ea men and con ol g oups
in he cu en p ice (Panel A) and p ojec ed p ice (Panel B) expe imen s. P io o he
in o ma ion ea men he e a e no s a is ically signi ican di e ences in mean ca bon p ice
accep ance and o he p e- ea men a iables. Pe cei ed cos s, ac ual cos s, pe cep ion
gaps, and he sha e o indi iduals o e es ima ing hei cos s a e consis en ac oss g oups,
con i ming success ul andomiza ion in bo h expe imen s.
The lowe pa o each panel epo s pos - ea men ou comes. In he cu en p ice
expe imen , ea ed esponden s exhibi signi ican ly highe pos - ea men accep ance
le els han he con ol g oup, wi h a mean di e ence o 0.25 (signi ican a he 1 pe cen
le el). The posi i e e ec o he in o ma ion ea men is consis en wi h he widesp ead
o e es ima ion o cu en cos s. The di e ence in mean accep ance e isions, which ac-
coun s o (insigni ican ) di e ences in p e- ea men accep ance, is sligh ly smalle bu
emains highly signi ican . No ably, he la ge absolu e accep ance e isions indica e sig-
ni ican upda ing beha io in bo h di ec ions along he accep ance scale. The las ow
in Panel A shows ha ea ed esponden s a e 17 pe cen age poin s mo e likely o e ise
hei accep ance compa ed o he con ol g oup – ye a la ge ac ion o esponden s in
bo h g oups does no e ise accep ance.
In con as , in he p ojec ed p ice expe imen , whe e esponden s p edominan ly un-
de es ima e hei u u e cos s, pos - ea men accep ance le els o ca bon p icing a e
13
signi ican ly lowe among ea ed esponden s ela i e o he con ol g oup, wi h a mean
di e ence o 0.34 (signi ican a he 1 pe cen le el). T ea ed esponden s dec ease hei
accep ance le el by 0.29 mo e han hose in he con ol g oup on a e age (signi ican a
he 1 pe cen le el), when accoun ing o p e-exis ing di e ences. The absolu e accep ance
e isions a e la ge , indica ing subs an ial upda ing beha io in bo h di ec ions. Finally,
he sha e o non- e ise s is 18 pe cen age poin s lowe in he ea men g oup.
O e all, hese esul s sugges a causal e ec o ou ea men s on ca bon p ice accep-
ance. The di ec ion o he a e age accep ance e ision is consis en wi h he sign o
he a e age pe cep ion gap. In he cu en p ice expe imen , pe sonalized in o ma ion
mainly co ec s cos o e es ima ion and boos s accep ance on a e age, whe eas in o ma-
ion abou u u e cos s educes accep ance due o cos unde es ima ion in he p ojec ed
p ice expe imen . We p o ide a mo e de ailed analysis o he ea men e ec s below.
4.2. Reg ession e idence
4.2.1. Empi ical app oach
To o mally iden i y he causal e ec o pe sonalized cos in o ma ion on ca bon p ice
accep ance, we es ima e he ollowing eg ession model using OLS:
accpos
i=α0+βTi+α1accp io
i+δXi+ϵi(1)
The dependen a iable accpos
icon inuously measu es he pos - ea men accep ance o
esponden i, anging om 1 ( e y unaccep able) o 5 ( e y accep able). Tiis a dummy
a iable aking he alue 1 o esponden s who ecei e he in o ma ion ea men and 0
o he wise. We con ol o an indica o o p e- ea men accep ance accp io
i. This helps
isola e ea men e ec s om p e-exis ing di e ences in accep ance, despi e he measu e
no necessa ily being in e pe sonally compa able. We also include a se o indi idual-le el
con ols Xi o imp o e he p ecision o ou es ima es and o accoun o mino imbalances
14
be ween ea men and con ol g oups.15 ϵiis an indi idual-speci ic e o e m. The
coe icien o in e es , β, iden i ies he a e age change in ca bon p ice accep ance among
ea ed esponden s ela i e o he con ol g oup.
Howe e , he model in Equa ion (1) incomple ely cha ac e izes how he in o ma ion
ea men s a ec policy accep ance, as he a e age ea men e ec may mask he e o-
genei y ega ding he di ec ion and magni ude o cos mispe cep ions. Genuine belie
upda ing sugges s ha he e ec size is la ge o esponden s wi h less accu a e p io s,
as he alue o he signal should inc ease wi h ex an e biasedness. Thus, we expand ou
model by in e ac ing he ea men dummy wi h esponden s’ pe cep ion gap:
accpos
i=α0+βTi+θ1∆ϕi+γ(Ti×∆ϕi) + α1accp io
i+δXi+ϵi(2)
The pe cep ion gap ∆ϕiis he di e ence be ween es ima ed and ac ual cos s in uni s o
e100. Hence, posi i e alues indica e an o e es ima ion o cos s and ice e sa. βnow
iden i ies he ea men e ec o pa icipan s wi h no bias in cos pe cep ions, whe eas γ
cap u es how he ea men e ec scales wi h he size o he pe cep ion gap. We expec
he es ima es o γ o be posi i e i esponden s change hei policy accep ance as a esul
o upda ing hei cos pe cep ions in he di ec ion o he p o ided signal.
Las ly, we explo e whe he pa icipan s’ esponsi eness o he in o ma ion ea men
a ies wi h hei baseline unce ain y abou he cos s o ca bon p icing. P io o he
in o ma ion ea men , we measu e unce ain y on a i e-poin Like scale in he second
s ep o he expe imen s, wi h la ge alues indica ing highe unce ain y. Responden s
wi h a highe cos unce ain y should place mo e weigh on he p o ided signal, leading o
la ge e isions in policy accep ance. We es his by adding in e ac ions wi h esponden s’
15Speci ically, we include dummies o being male, li ing in Eas Ge many, holding a high school diploma,
li ing in a high income household, homeowne ship, whe he he esponden ’s household uses ossil uels
o hea ing (including ho wa e ), owns any mo o ehicles, and an indica o a iable o he esponden ’s
poli ical pa y p e e ence. We include a quad a ic polynomial in he esponden ’s age, and con inuously
con ol o household size.
15
unce ain y abou hei cos s o ca bon p icing o ou model:
accpos
i=α0+βTi+θ1∆ϕi+γ(Ti×∆ϕi) + θ2Ui+βU(Ti×Ui)
+θ3(∆ϕi×Ui) + γU(Ti×∆ϕi×Ui) + α1accp io
i+δXi+ϵi
(3)
whe e Uiis a con inuous measu e o baseline unce ain y o esponden i. No ice ha
βnow measu es he a e age ea men e ec o esponden s wi h a ze o pe cep ion
gap and low unce ain y, whe eas γis an es ima e o he ea men -induced change in
accep ance wi h espec o pe cep ion gap size o high ce ain y esponden s. βUcap u es
he change in ea men e ec wi h inc easing unce ain y a a ze o pe cep ion gap. The
coe icien o p ima y in e es γUgi es us he change in accep ance wi h espec o he
pe cep ion gap and unce ain y o a ea ed esponden . We expec γU o be posi i e, i
he in o ma i eness o he p o ided signal is inc easing wi h baseline unce ain y.
4.2.2. Main esul s
The esul s om he main eg ession analyses a e p esen ed in Table 2. We begin by
discussing indings om he cu en p ice expe imen in Columns 1 o 3. Column 1
shows esul s om he baseline speci ica ion ha eg esses pos - ea men accep ance
on a ea men dummy, p e- ea men accep ance, and a se o indi idual-le el con ols
(Equa ion 1).16 In he cu en p ice expe imen , he in o ma ion ea men inc eases
esponden s’ accep ance le el o ca bon p icing by 0.161 on a e age (signi ican a he 1
pe cen le el). The posi i e es ima e o βis consis en wi h he widesp ead o e es ima ion
o cu en cos s.
Suppo ing e idence e eals subs an ial he e ogenei y in ea men e ec s based on he
sign o he pe cep ion gap (Appendix Table A7). Responden s who o e es ima e hei
cos s (including hose wi h a ze o pe cep ion gap) inc ease hei accep ance le el by 0.365
on a e age, while hose who unde es ima e cos s educe hei accep ance le el by 0.228
(Column 1). These pa e ns highligh ha he e ec o in o ma ion p o ision depends
16Fo comple eness, we ep oduce ou baseline analysis om Table 2 wi hou any addi ional con ols in
Table A6 o he Appendix, whe e we ob ain i ually unchanged es ima es.
16
c i ically on he di ec ion o an indi idual’s bias.
In Column 2 in Table 2, we examine whe he he size o he ea men e ec inc eases
wi h he magni ude o cos mispe cep ions by in e ac ing he ea men dummy wi h he
pe cep ion gap (Equa ion 2). The es ima ed coe icien on he in e ac ion e m (γ) is
posi i e and s a is ically signi ican a he 1 pe cen le el, indica ing ha he e ec o
ou ea men signi ican ly inc eases in he size o a esponden ’s bias. Speci ically, o
each e100 inc ease in he pe cep ion gap, accep ance inc eases by an addi ional 0.033, on
op o a baseline ea men e ec o 0.086 (β). La ge accep ance e isions among mo e
biased esponden s suppo he no ion o meaning ul upda ing o cos pe cep ions.17
In e es ingly, he es ima e o βsugges s ha he ea men inc eases he accep ance
o esponden s wi h accu a e p io belie s.18 This esul is consis en wi h beha io al
e idence sugges ing ha indi iduals espond posi i ely o a i ma ion. Fo ins ance, loss
a e sion (T e sky and Kahneman, 1991) unde sco es he asymme ic emo ional impac
o pe cei ed gains e sus losses, which may ein o ce posi i e esponses o alida ed ex-
pec a ions. Rela edly, mo i a ed easoning implies ha con i ming indi iduals’ belie s o
pe cep ions – such as hei cos assessmen s – can elici a o able eac ions (B´enabou and
Ti ole, 2016).
Column 3 in Table 2 ex ends he analysis o include esponden s’ ex an e unce ain y
abou he pe cei ed cos s o ca bon p icing (Equa ion 3). We ind ha ea men espon-
si eness inc eases wi h bo h highe unce ain y and la ge mispe cep ions, as indica ed by
a posi i e and signi ican iple in e ac ion coe icien (γU).19 The signi ican es ima e o
17The ele ance o he pe cep ion gap may depend on how i ela es o a esponden ’s ex-an e cos
pe cep ion. Tha is, o he same absolu e pe cep ion gap, he s eng h o he signal could a y ac oss
indi iduals wi h di e en baseline cos pe cep ions. To accoun o his, Appendix Table A8 e-es ima es
Equa ion (2) using a ela i e pe cep ion gap, de ined as he log di e ence be ween pe cei ed and ac ual
cos s. Resul s a e consis en o bo h expe imen s.
18We u he explo e his inding by es ima ing he e ogeneous ea men e ec s o esponden s who
o e es ima e, unde es ima e, o accu a ely es ima e hei cos s o ca bon p icing based on di e en
pe cep ion accu acy h esholds. To his end, we apply wo di e en de ini ions o unbiasedness. In
Column 2 in Table A7 we classi y esponden s wi h a pe cep ion gap o up o 20 pe cen o hei ac ual
cos s as unbiased. In Column 3 we allow o a de ia ion o e25, i.e. an absolu e pe cep ion gap o
up o e25, o a esponden ’s pe cep ion o be de ined as accu a e. The es ima ed ma ginal e ec s o
unbiased esponden s a e simila in magni ude o ou main speci ica ion ye noisily es ima ed (Table A7,
Columns 2 and 3).
19To pu ou esul s in o pe spec i e, a esponden o e es ima ing cos s by e300 and epo ing high
unce ain y (unce ain y = 4) is p edic ed o inc ease accep ance by 0.29 poin s in esponse o he
ea men (0.001 + 0.003 ×3 + 0.028 ×4 + 0.014 ×3×4).
17
γUcon i ms heo e ical p edic ions abou belie upda ing as he unde lying causal mech-
anism.
He e, we ind no ea men e ec on accep ance among high-ce ain y esponden s –
ega dless o whe he hei cos pe cep ions a e biased – e lec ed in insigni ican es i-
ma es o bo h βand γ. Hence, he p e iously obse ed posi i e ea men e ec a ze o
pe cep ion gap (es ima e o βin Column 2) disappea s once accoun ing o unce ain y.
Ins ead, he posi i e – hough imp ecisely es ima ed – coe icien on he in e ac ion wi h
unce ain y (βU) sugges s ha he ea men e ec among unbiased esponden s is d i en
by unce ain esponden s.
We conduc he same analyses o he p ojec ed p ice expe imen , p esen ed in Columns
4 o 6 in Table 2. Column 4 p o ides esul s om ou baseline speci ica ion (Equa ion
1), showing ha pe sonalized in o ma ion abou u u e cos s o ca bon p icing leads o
a dec ease in accep ance le els by 0.307 on a e age (signi ican a he 1 pe cen le el).
This esul e lec s ha esponden s p edominan ly unde es ima e hei u u e cos s and
he eby con as s ou indings om he cu en p ice expe imen . Column 4 in Appendix
Table A7 con i ms he p e iously documen ed asymme y in ea men e ec s based on
he sign o he pe cep ion gap. Among hose unde es ima ing hei cos s, we ind an
a e age educ ion in accep ance le els o 0.471, whe eas he much smalle numbe o
o e es ima o s inc ease hei s by 0.103.
Column 5 in Table 2 shows ha he ea men e ec inc eases wi h he ex en o espon-
den s’ mispe cep ions, as measu ed by he pe cep ion gap (Equa ion 2). The in o ma ion
ea men in he p ojec ed p ice expe imen dec eases accep ance by 0.229 a a ze o pe -
cep ion gap (β), and changes by an addi ional 0.026 o e e y e100 di e ence in he
pe cep ion gap (γ; bo h signi ican a he 1 pe cen le el). These esul s a e analogous
in pa e n, o he ampli ying e ec o cos mispe cep ions in he cu en cos expe imen .
Howe e , he analysis in Appendix Table A7 shows ha when classi ying esponden s by
disc e e pe cep ion accu acy h esholds, he es ima ed ea men e ec o unbiased in-
di iduals is s a is ically insigni ican and quan i a i ely close o ze o. This sugges s ha
he signi ican nega i e coe icien a a ze o pe cep ion gap in Column 5 in Table 2 may
18
be an a i ac o he linea i y assump ion and should be in e p e ed wi h cau ion.20
Column 6 in Table 2 inco po a es esponden s’ ex an e unce ain y (Equa ion 3). We
ind ha he e ec o co ec ing mispe cep ions abou u u e cos s is d i en by hose
wi h highe unce ain y. The es ima ed coe icien on he iple in e ac ion e m (γU) is
posi i e and s a is ically signi ican a he 5 pe cen le el, whe eas he pai wise in e ac ion
be ween ea men and pe cep ion gap (γ) is no signi ican ly di e en om ze o.21 This
esul again mi o s he mechanisms in he cu en cos expe imen .
O e all, we ind ha esponden s sys ema ically e ise hei accep ance o ca bon p ic-
ing in esponse o pe sonalized in o ma ion abou ei he hei cu en o u u e cos s.
T ea men esponses a e consis en ac oss bo h expe imen s and align wi h heo e ical
expec a ions: They a e la ge among hose wi h g ea e mispe cep ions and highe ex
an e unce ain y abou hei cos s. Howe e , due o di e ing dis ibu ions o cos mis-
pe cep ions ac oss he wo expe imen s, a e age ea men e ec s di e ge. In he cu en
p ice expe imen , whe e mos esponden s o e es ima e hei ac ual cos s, he in o ma-
ion ea men leads o a ne inc ease in accep ance. By con as , in he p ojec ed p ice
expe imen , whe e unde es ima ion o u u e cos s is mo e widesp ead, he pe sonalized
in o ma ion esul s in a ne dec ease in accep ance. Ou indings highligh how indi idual
upda ing o policy p e e ences in esponse o pe sonalized in o ma ion depends c ucially
on he di ec ion and magni ude o mispe cep ions, as well as indi iduals’ unce ain y.
Thus, while pe sonalized in o ma ion-in e en ions can shi public suppo , he di ec ion
ul ima ely depends on he in o ma ional backg ound o he popula ion.
20In Columns 5 and 6 in Table A7 in he Appendix we es ima e he e ogeneous ea men e ec s o
esponden s who o e es ima e, unde es ima e, o accu a ely es ima e hei u u e cos s o ca bon p icing.
In Column 5, we allow o a ela i e e o o up o 20 pe cen in pe cei ed cos s and in Column 6 o
an absolu e e o o up o e100 o a esponden o be classi ied as unbiased. Bo h app oaches yield
ea men e ec s o unbiased esponden s ha a e no s a is ically di e en om ze o. This con as s
wi h he esul s in Column 5 o Table 2, whe e he signi ican nega i e e ec a a ze o pe cep ion gap
(β= -0.229) likely s ems om he linea i y assump ion o he speci ica ion. Speci ically, i a ises om
ex apola ing a linea ela ionship ac oss he ull ange o pe cep ion gaps, while he speci ica ions in
Table A7 ely on disc e e classi ica ions. Acco dingly, he in e ac ion e ec in Column 5 in Table 2
should be in e p e ed wi h cau ion, pa icula ly a ound he ze o pe cep ion gap egion.
21While he esul s in Column 6 o Table 2 should be in e p e ed wi h cau ion due o he discussed
limi a ions o he linea model speci ica ion, he es ima es p edic ha a esponden unde es ima ing
u u e cos s by e300 and epo ing high unce ain y (unce ain y = 4) dec eases accep ance by 0.334
poin s in esponse o he ea men (-0.187 + 0.009 ×(-3) + (-0.009) ×4 + 0.007 ×(-3) ×4).
19
4.2.3. He e ogenei y by p e- ea men accep ance
The p e ious analysis explo es he a e age ea men e ec s and he unde lying mecha-
nisms. F om a poli ical pe spec i e, he e ec i eness o he in o ma ion in e en ion also
hinges on he abili y o coun e ac o e en e e se indi iduals’ s ance owa d ca bon p ic-
ing. Fo example, a highe a e age accep ance may s em om suppo e s becoming mo e
posi i e owa d ca bon p icing o , mo e meaning ully, om indi iduals ini ially opposed
o he policy lessening hei opposi ion o e en u ning suppo i e. Only he la e has
he capaci y o shi public opinion owa d majo i y suppo , whe eas he o me would
lea e a dicho omous dis ibu ion o p e e ences unchanged.
The e o e, we e alua e he in e en ion’s po en ial o build b oade suppo o ca bon
p icing by examining whe he he size o he ea men e ec di e s based on ex-an e
a i ude owa d ca bon p icing. To his end, we in e ac he ea men dummy wi h
esponden s’ p e- ea men accep ance (Appendix Table A9) and plo he ma ginal ea -
men e ec s in Figu e 4.22 In he cu en p ice expe imen , ma ginal ea men e ec s a e
posi i e and s a is ically signi ican among esponden s who ound ca bon p icing e y
unaccep able o a he unaccep able p io o he ea men (Figu e 4a). In u n, he
es ima ed e ec s o su ey pa icipan s wi h an ex-an e neu al o posi i e s ance a e
no signi ican ly di e en om ze o.23 Hence, he posi i e ea men e ec is d i en by
hose wi h a low baseline accep ance, unde sco ing he poli ical e ec i eness o ailo ed
in o ma ion p o ision abou cu en cos s.
To be e unde s and he implica ions o his esul o mo ing owa d majo i y sup-
22In Table A10 o he Appendix, we add ess he conce n ha he e ogenei y in ea men e ec s po en-
ially esul s om co ela ions be ween p e- ea men accep ance and o he esponden cha ac e is ics.
Following Haaland and Ro h (2020), we decompose he o al a ia ion in p e- ea men accep ance in o
a componen explained by obse ed esponden cha ac e is ics and an un ela ed esidual componen .
This is done by eg essing p e- ea men accep ance on he same se o con ol a iables used in ou main
analyses. We hen examine ea men e ec he e ogenei y sepa a ely o each componen . Consis en
wi h he main esul s om bo h expe imen s, we ind signi ican he e ogenei y in ea men e ec s wi h
espec o he esidual a ia ion (Columns 2 and 5), sugges ing ha p e- ea men accep ance is inde-
penden ly meaning ul. In con as , we do no obse e ea men he e ogenei y based on he p edic ed
componen linked o esponden cha ac e is ics (Columns 3 and 6).
23We p e iously iden i ied pe cei ed cos s as a s ong p edic o o ca bon p ice accep ance (Appendix
Table A5). Panel A in Appendix Table A12 shows ha pe cei ed cu en cos s a e g ea ly in la ed
o esponden s wi h lowe baseline accep ance le els, despi e simila ac ual cos s ac oss g oups. Con-
sequen ly, he he e ogenei y o ou esul s is d i en by la ge pe cep ion gaps (and a la ge sha e o
o e es ima o s) among esponden s wi h a low p e- ea men accep ance.
20
po , we addi ionally es ima e ea men -induced changes in he p obabili y o selec ing
each accep ance le el using an o de ed p obi model. The ma ginal e ec s in Panel A in
Appendix Table A11 indica e ha he likelihood o conside ca bon p icing e y unac-
cep able d ops by 5.1 pe cen age poin s o ea ed esponden s. We ind a co esponding
inc ease in he p obabili ies o ind ca bon p icing accep able (4.5 pe cen age poin s)
o exp ess neu ali y (0.6 pe cen age poin s). Thus, p o iding ailo ed in o ma ion on
cu en cos s can meaning ully b oaden public suppo o ca bon p icing.
An analogous analysis o he p ojec ed p ice expe imen yields a less posi i e ou look.
He e, we ind la ge and s a is ically signi ican educ ions in accep ance ac oss almos
all le els o p e- ea men accep ance, excep o hose leas a o able owa d ca bon
p icing a baseline (Figu e 4b).24 Ma ginal e ec s om he o de ed p obi model in Panel
B o Table A11 show ha he p obabili y o inding ca bon p icing e y o somewha
unaccep able inc eases by 9.6 pe cen age poin s in esponse o he ea men , while he
likelihoods o neu ali y and policy accep ance decline by 1.7 and 7.9 pe cen age poin s,
espec i ely. Hence, ou esul s sugges ha in o ma ion shocks ega ding he u u e cos s
o ca bon p icing may ein o ce opposi ion o he policy.
In sum, pe sonalized in o ma ion on cu en cos s can build b oade suppo by shi ing
iews among hose mos c i ical o ca bon p icing, ye in o ma ion abou u u e cos s ends
o educe suppo ac oss he boa d. Mo eo e , he ad e se e ec s om in o ming abou
u u e cos s exceed he inc ease in ca bon p ice accep ance om pe sonalized in o ma ion
on cu en cos s.
4.2.4. O he he e ogenei y
Iden i ying he popula ions ha d i e a e age ea men e ec s can in o m mo e a ge ed
and cos -e ec i e communica ion s a egies (Allco , 2011). To be e unde s and which
g oups a e mos esponsi e, we in e ac he ea men indica o wi h s anda d sociode-
mog aphic cha ac e is ics (dummies o age, educa ion, gende , household size, egion),
24This inding is consis en wi h limi ed di e ences in mean pe cei ed cos s, ac ual cos s, and pe cep-
ion gaps by p e- ea men accep ance le els (Appendix Table A12, Panel B). Apa om hose wi h
he lowes baseline accep ance, who hold mo e accu a e belie s, all o he su ey pa icipan s end o
unde es ima e hei u u e cos s o a simila deg ee.
21
poli ical a ilia ion ( uling pa y s. opposi ion), and indica o s o a esponden ’s inan-
cial si ua ion and exposu e o ca bon p icing (income, homeowne ship, households’ ca bon
p ice exposu e in he building and anspo sec o s).25,26
When examining he e ogenei y along s anda d sociodemog aphic dimensions, we ind
limi ed sys ema ic a ia ion. Mos in e ac ion e ec s a e imp ecisely es ima ed and close
o ze o. In he cu en p ice expe imen (Panel A, Table 3), men show a signi ican ly
smalle inc ease in accep ance (Column 1), while esponden s li ing in Eas Ge many
appea sligh ly mo e esponsi e (Column 3). Howe e , he la e e ec is no s a is-
ically signi ican . In he p ojec ed p ice expe imen (Panel B), olde indi iduals and
suppo e s o he go e ning coali ion become signi ican ly mo e opposed ollowing ea -
men (Columns 2 and 6). O e all, hese pa e ns align wi h di e ences in p e- ea men
belie s, such as mean pe cep ion gaps o he sha e o esponden s o e es ima ing cos s
(Appendix Table A13 and Table A14).
The di e en ial esponse by poli ical a ilia ion is pa icula ly no able. While suppo -
e s o bo h go e nmen and opposi ion pa ies espond posi i ely o in o ma ion abou
cu en cos s, go e nmen suppo e s exhibi a s onge nega i e eac ion when exposed
o p ojec ed u u e cos s. This di e gence unde sco es he poli ical isk ha u u e cos
shocks may disp opo iona ely educe suppo among hose mos likely o back ca bon
p icing poli ically, he eby weakening i s suppo base.
Fu he mo e, we consis en ly obse e p onounced he e ogenei y based on inancial cha -
ac e is ics and hose di ec ly ela ed o ca bon p icing exposu e. Columns 7 o 10 in Panel
A o Table 3 e eal ha he posi i e ea men e ec in he cu en p ice expe imen is
concen a ed among lowe -income indi iduals, en e s and hose no elian on ossil u-
els o hea ing and anspo a ion. Appendix Table A13 shows ha hese esponden s
ace signi ican ly lowe ac ual cos s han hei coun e pa s, whe eas cos pe cep ions a e
mo e aligned ac oss g oups. This implies a signi ican ly highe p e alence o cos o e -
25We include homeowne s a us as a measu e o inancial exposu e, as en e s ace subs an ially lowe
cos s o ca bon p icing due o egula ions ha impose a sha e o he cos s on hei landlo ds.
26The uling coali ion comp ises he Social Democ a ic Pa y (SPD), he G een Pa y (B¨undnis 90/Die
G ¨unen), and he libe al F ee Democ a ic Pa y (FDP), which in oduced he ca bon p ice in 2021 and
was in powe when we conduc ed he su ey.
22
axes in a pos -Pa is wo ld: a e millions o nays ine i able?’, En i onmen al and Re-
sou ce Economics 68, 97–128.
Ca a ini, S., Ca alho, M. and Fankhause , S. (2018), ‘O e coming public esis ance o
ca bon axes’, Wiley In e disciplina y Re iews: Clima e Change 9(5), e531.
Ca d, D., Mas, A., Mo e i, E. and Saez, E. (2012), ‘Inequali y a wo k: The e ec o
pee sala ies on job sa is ac ion’, Ame ican Economic Re iew 102(6), 2981–3003.
Clima e Leade ship Council (2019), ‘Economis s’ S a emen on Ca bon Di idends’.
C owley, K. (2021), ‘Figh ing he u u e: The poli ics o clima e policy ailu e in Aus alia
(2015–2020)’, Wiley In e disciplina y Re iews: Clima e Change 12(5), e725.
C uces, G., Pe ez-T uglia, R. and Te az, M. (2013), ‘Biased pe cep ions o income dis i-
bu ion and p e e ences o edis ibu ion: E idence om a su ey expe imen ’, Jou nal
o Public Economics 98, 100–112.
Dabla-No is, M. E., Helbling, M. T., Khalid, S., Khan, H., Magis e i, G., Sollaci, A. and
S ini asan, M. K. (2023), Public pe cep ions o clima e mi iga ion policies: E idence
om c oss-coun y su eys, S a Discussion No e 2023/002, In e na ional Mone a y
Fund, Washing on, DC.
Dechezlep ˆe e, A., Fab e, A., K use, T., Plan e ose, B., Sanchez Chico, A. and
S an che a, S. (2025), ‘Figh ing clima e change: In e na ional a i udes owa d clima e
policies’, Ame ican Economic Re iew 115(4), 1258–1300.
Douenne, T. and Fab e, A. (2022), ‘Yellow es s, pessimis ic belie s, and ca bon ax
a e sion’, Ame ican Economic Jou nal: Economic Policy 14(1), 81–110.
D ews, S. and Van den Be gh, J. C. (2016), ‘Wha explains public suppo o clima e
policies? A e iew o empi ical and expe imen al s udies’, Clima e Policy 16(7), 855–
876.
29
Fanghella, V., Fau e, C., Gue lein, M.-C. and Schleich, J. (2023), ‘Wha ’s in i o me?
Sel -in e es and p e e ences o dis ibu ion o cos s and bene i s o ene gy e iciency
policies’, Ecological Economics 204, 107659.
Gen zkow, M. and Shapi o, J. M. (2006), ‘Media bias and epu a ion’, Jou nal o Poli ical
Economy 114(2), 280–316.
G oh, E. D. and Ziegle , A. (2018), ‘On sel -in e es ed p e e ences o bu den sha ing ules:
An econome ic analysis o he cos s o ene gy policy measu es’, Ene gy Economics
74, 417–426.
Haaland, I. and Ro h, C. (2020), ‘Labo ma ke conce ns and suppo o immig a ion’,
Jou nal o Public Economics 191, 104256.
Haaland, I., Ro h, C. and Wohl a , J. (2023), ‘Designing in o ma ion p o ision expe i-
men s’, Jou nal o Economic Li e a u e 61(1), 3–40.
H idbe g, K. B., K eine , C. T. and S an che a, S. (2023), ‘Social posi ions and ai ness
iews on inequali y’, Re iew o Economic S udies 90(6), 3083–3118.
Jacksohn, A., G ¨osche, P., Rehdanz, K. and Sch ¨ode , C. (2019), ‘D i e s o enewable
echnology adop ion in he household sec o ’, Ene gy Economics 81, 216–226.
Kaes ne , K., Pahle, M., Schwa z, A., Somme , S. and S ¨unzi, A. (2023), Expe s’ con-
jec u es, people’s s a emen s and ue p e e ences: The case o ca bon p ice suppo ,
USAEE Wo king Pape 23-591, USAEE.
Kaes ne , K., Somme , S., Be neise , J., Henge , R. and Obe s , C. (2025), ‘Cos sha ing
mechanisms o ca bon p icing: Wha d i es suppo in he housing sec o ?’, Ene gy
Economics 142, 108134.
Kalkuhl, M., Kellne , M., Be gmann, T. and R¨u en, K. (2023), CO2-Bep eisung zu
E eichung de Klimaneu ali ¨a im Ve keh s- und Geb¨audesek o : In es i ionsan eize
und Ve eilungswi kungen, Technical epo , Me ca o Resea ch Ins i u e on Global
Commons and Clima e Change, Be lin.
30
Kallbekken, S. and Sælen, H. (2011), ‘Public accep ance o en i onmen al axes: Sel -
in e es , en i onmen al and dis ibu ional conce ns’, Ene gy Policy 39(5), 2966–2973.
Ka adja, M., Molle s om, J. and Seim, D. (2017), ‘Riche (and holie ) han hou? The
e ec o ela i e income imp o emen s on demand o edis ibu ion’, Re iew o Eco-
nomics and S a is ics 99(2), 201–212.
Klene , D., Ma auch, L., Combe , E., Edenho e , O., Hepbu n, C., Ra a y, R. and
S e n, N. (2018), ‘Making ca bon p icing wo k o ci izens’, Na u e Clima e Change
8(8), 669–677.
Konc, T., D ews, S., Sa in, I. and Van Den Be gh, J. C. (2022), ‘Co-dynamics o clima e
policy s ingency and public suppo ’, Global En i onmen al Change 74, 102528.
Kuziemko, I., No on, M. I., Saez, E. and S an che a, S. (2015), ‘How elas ic a e p e -
e ences o edis ibu ion? E idence om andomized su ey expe imen s’, Ame ican
Economic Re iew 105(4), 1478–1508.
Maes e-And ´es, S., D ews, S. and Van Den Be gh, J. (2019), ‘Pe cei ed ai ness and
public accep abili y o ca bon p icing: a e iew o he li e a u e’, Clima e Policy
19(9), 1186–1204.
Mildenbe ge , M., Lachapelle, E., Ha ison, K. and S adelmann-S e en, I. (2022), ‘Lim-
i ed impac s o ca bon ax eba e p og ammes on public suppo o ca bon p icing’,
Na u e Clima e Change 12(2), 141–147.
Ne ini, F. F., Keppo, I. and S achan, N. (2017), ‘Myopic decision making in ene gy
sys em deca bonisa ion pa hways. A UK case s udy’, Ene gy S a egy Re iews 17, 19–
26.
Ro h, C., Se ele, S. and Wohl a , J. (2022), ‘Risk exposu e and acquisi ion o mac oe-
conomic in o ma ion’, Ame ican Economic Re iew: Insigh s 4(1), 34–53.
Ro h, C. and Wohl a , J. (2020), ‘How do expec a ions abou he mac oeconomy a ec
31
pe sonal expec a ions and beha io ?’, Re iew o Economics and S a is ics 102(4), 731–
748.
Sælen, H. and Kallbekken, S. (2011), ‘A choice expe imen on uel axa ion and ea ma k-
ing in No way’, Ecological Economics 70(11), 2181–2190.
Schui ema, G., S eg, L. and Fo wa d, S. (2010), ‘Explaining di e ences in accep abili y
be o e and accep ance a e he implemen a ion o a conges ion cha ge in S ockholm’,
T anspo a ion Resea ch Pa A: Policy and P ac ice 44(2), 99–109.
Schwa z, A., S ¨unzi, A., Kaes ne , K., Pahle, M. and Somme , S. (2024), Tailo ed in o -
ma ion and he public suppo o ca bon p icing, Wo king pape .
Schyns, B. and Paul, T. (2002), ‘Deu sche Sel -Moni o ing Skala’, A ailable a h ps:
//zis.gesis.o g/DoiId/zis55 (2025/06/11).
Snyde , M. (1974), ‘Sel -moni o ing o exp essi e beha io .’, Jou nal o Pe sonali y and
Social Psychology 30(4), 526–537.
Somme , S., Konc, T. and D ews, S. (2023), How esilien is public suppo o ca bon
p icing? Longi udinal e idence om Ge many, Ruh Economic Pape s 1017, RWI -
Leibniz-Ins i u ¨u Wi scha s o schung.
Somme , S., Ma auch, L. and Pahle, M. (2022), ‘Suppo ing ca bon axes: The ole o
ai ness’, Ecological Economics 195, 107359.
S an che a, S. (2021), ‘Unde s anding ax policy: How do people eason?’, The Qua e ly
Jou nal o Economics 136(4), 2309–2369.
T e sky, A. and Kahneman, D. (1991), ‘Loss a e sion in iskless choice: A e e ence-
dependen model’, The Qua e ly Jou nal o Economics 106(4), 1039–1061.
Wasi, N. and Ca son, R. T. (2013), ‘The in luence o eba e p og ams on he demand o
wa e hea e s: The case o New Sou h Wales’, Ene gy Economics 40, 645–656.
Zimme mann, F. (2020), ‘The dynamics o mo i a ed belie s’, Ame ican Economic Re iew
110(2), 337–363.
32
In o ques ions: gende , age, ede al s a e, household income
Ques ions on households’ hea ing and d i ing habi s
Q: P io accep ance o ca bon p icing
Q: Cos s o ca bon p icing a
p ice o AC45/ CO2 + unce ain y
Q: Cos s o ca bon p icing a
p ice o AC200/ CO2 + unce ain y
T ea men :
Ac ual cos s No ea men T ea men :
Ac ual cos s No ea men
Q: Pos e io accep ance o ca bon p icing
O he ques ions: socio-demog aphics, poli ical pa y p e e ences, su ey eedback
p≈0.5 p ≈0.5
p≈0.5 p ≈0.5 p ≈0.5 p ≈0.5
No es: The igu e p o ides an o e iew o he s uc u e o he su ey expe imen s.
Figu e 1: S uc u e o he su ey expe imen s
0.0 0.1 0.2 0.3 0.4
F ac ion
Ve y accep able
Ra he accep able
Nei he no
Ra he unaccep able
Ve y unaccep able
No es: The igu e shows he p e- ea men accep ance o ca bon p icing o all esponden s.
Figu e 2: P e- ea men accep ance o ca bon p icing
33
0.00
0.05
0.10
0.15
0.20
0.25
F ac ion
−4000 −3000 −2000 −1000 0 1000 2000 3000 4000 5000
Pe cep ion gap
(a) Cu en p ice
0.00
0.05
0.10
0.15
0.20
0.25
F ac ion
−4000 −3000 −2000 −1000 0 1000 2000 3000 4000 5000
Pe cep ion gap
(b) P ojec ed p ice
No es: The igu e shows he dis ibu ions o he esponden s’ cos pe cep ion gaps (in bins o e100) o
he cu en p ice o e45 pe on o CO2 (a) and he p ojec ed p ice o e200 pe on o CO2 (b). The
pe cep ion gap is de ined as he di e ence be ween pe cei ed and ac ual cos s. Posi i e (nega i e) alues
indica e o e es ima ion (unde es ima ion) o cos s.
Figu e 3: Dis ibu ion o cos pe cep ion gaps
−1.0
−0.5
0.0
0.5
Ma ginal e ec
1 2 3 4 5
P e− ea men accep ance
(a) Cu en p ice
−1.0
−0.5
0.0
0.5
Ma ginal e ec
1 2 3 4 5
P e− ea men accep ance
(b) P ojec ed p ice
No es: The igu e shows ma ginal ea men e ec s on esponden s’ pos - ea men accep ance o ca bon
p icing by esponden s’ p e- ea men accep ance (1= e y unaccep able o 5= e y accep able). The
igu e displays he ma ginal ea men e ec s wi h 95 pe cen con idence in e als. The co esponding
es ima es a e p esen ed in Appendix Table A9.
Figu e 4: He e ogenei y by p e- ea men accep ance
34
Table 1: A e age accep ance and cos pe cep ions by ea men
Panel A: 45 e/ CO2 Con ol (1) T ea ed (2) Di . (1)-(2)
P e- ea men accep ance 2.40 2.51 -0.11
Pe cei ed cos s 392.29 413.53 -21.24
Ac ual cos s 199.99 193.00 6.99
Pe cep ion gap 192.30 220.53 -28.23
O e es ima ion 0.64 0.65 -0.01
Pos - ea men accep ance 2.37 2.62 -0.25***
Accep ance e ision -0.03 0.11 -0.14***
Abs. accep ance e ision 0.22 0.47 -0.25***
Non- e ise 0.81 0.64 0.17***
Obse a ions 814 816 1630
Panel B: 200 e/ CO2 Con ol (1) T ea ed (2) Di . (1)-(2)
P e- ea men accep ance 2.53 2.48 0.05
Pe cei ed cos s 559.36 576.28 -16.91
Ac ual cos s 864.11 864.65 -0.55
Pe cep ion gap -304.74 -288.38 -16.37
O e es ima ion 0.28 0.29 -0.01
Pos - ea men accep ance 2.43 2.08 0.34***
Accep ance e ision -0.10 -0.39 0.29***
Abs. accep ance e ision 0.26 0.54 -0.29***
Non- e ise 0.80 0.61 0.18***
Obse a ions 795 829 1624
No es: The able p esen s summa y s a is ics o key a iables ac oss ea men and
con ol g oups in he cu en p ice (Panel A) and p ojec ed p ice (Panel B) expe -
imen s. Columns 1 and 2 show means o he con ol and ea men g oups, and
Column 3 shows he di e ence in means be ween he wo g oups. O e es ima ion in-
cludes a small numbe o esponden s wi h a ze o pe cep ion gap. * p<0.1, ** p<0.05,
*** p<0.01
35
Table 2: T ea men e ec s on ca bon p ice accep ance
45 e/ CO2 200 e/ CO2
(1) (2) (3) (4) (5) (6)
T 0.161*** 0.086** 0.001 -0.307*** -0.229*** -0.187**
(0.034) (0.037) (0.099) (0.035) (0.036) (0.090)
Pe cep ion gap -0.004 0.009 -0.006*** -0.001
(0.004) (0.007) (0.002) (0.005)
T×Pe cep ion gap 0.033*** 0.003 0.026*** 0.009
(0.008) (0.016) (0.004) (0.009)
Unce ain y -0.019 -0.026
(0.018) (0.019)
T×Unc. 0.028 -0.009
(0.032) (0.030)
Pe cep ion gap ×Unc. -0.006** -0.002
(0.003) (0.002)
T×Pe cep ion gap ×Unc. 0.014** 0.007**
(0.006) (0.003)
Con ols Yes Yes Yes Yes Yes Yes
R-squa ed 0.741 0.747 0.749 0.712 0.724 0.725
Obse a ions 1630 1630 1620 1624 1624 1609
No es: The able p esen s es ima ion esul s om OLS eg essions. Resul s o he cu en p ice expe imen a e shown
in Columns 1 o 3 and esul s o he p ojec ed p ice expe imen a e shown in Columns 4 o 6. The dependen a iable
is he pos - ea men accep ance o ca bon p icing, measu ed on a i e-poin Like scale (1= e y unaccep able o
5= e y accep able). T is a dummy a iable indica ing ha a esponden ecei ed pe sonalized in o ma ion abou he
cos s o ca bon p icing. The pe cep ion gap is de ined as he di e ence be ween pe cei ed and ac ual cos s (di ided
by 100). Unce ain y ega ding pe cei ed cos s is measu ed on a i e-poin Like scale (0= e y unce ain o 4= e y
ce ain). All eg essions include he p e- ea men accep ance o ca bon p icing and he se o con ols desc ibed in
Appendix Table A5. Robus s anda d e o s a e in pa en heses. * p<0.1, ** p<0.05, *** p<0.01
36
Table 3: He e ogenei y by esponden cha ac e is ics
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
Panel A: High Ruling High Fossil Mo o
45 e/ CO2 Male Aged 45+ Eas school Single pa y income Homeowne hea ing ehicles
T 0.262*** 0.124** 0.138*** 0.178*** 0.151*** 0.129*** 0.233*** 0.254*** 0.412*** 0.438***
(0.052) (0.057) (0.036) (0.049) (0.040) (0.045) (0.042) (0.047) (0.079) (0.087)
T×C -0.191*** 0.057 0.154 -0.034 0.037 0.033 -0.250*** -0.221*** -0.318*** -0.341***
(0.068) (0.068) (0.108) (0.067) (0.067) (0.078) (0.071) (0.067) (0.087) (0.094)
Con ols Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
R-squa ed 0.742 0.741 0.741 0.741 0.741 0.765 0.742 0.742 0.743 0.743
Obse a ions 1630 1630 1630 1630 1630 1334 1630 1630 1630 1630
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
Panel B: High Ruling High Fossil Mo o
200 e/ CO2 Male Aged 45+ Eas school Single pa y income Homeowne hea ing ehicles
T -0.261*** -0.229*** -0.312*** -0.294*** -0.325*** -0.220*** -0.265*** -0.201*** -0.115 -0.083
(0.053) (0.057) (0.039) (0.049) (0.041) (0.042) (0.042) (0.046) (0.077) (0.092)
T×C -0.086 -0.121* 0.035 -0.026 0.056 -0.212** -0.142* -0.269*** -0.241*** -0.274***
(0.069) (0.067) (0.085) (0.069) (0.067) (0.085) (0.075) (0.069) (0.087) (0.099)
Con ols Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
R-squa ed 0.713 0.713 0.712 0.712 0.712 0.729 0.713 0.715 0.714 0.714
Obse a ions 1624 1624 1624 1624 1624 1344 1624 1624 1624 1624
No es: The able p esen s es ima ion esul s om OLS eg essions. Resul s o he cu en p ice expe imen a e shown in Panel A and esul s o he p ojec ed
p ice expe imen a e shown in Panel B. The dependen a iable is he pos - ea men accep ance o ca bon p icing, measu ed on a i e-poin Like scale (1= e y
unaccep able o 5= e y accep able). T is a dummy a iable indica ing ha a esponden ecei ed pe sonalized in o ma ion on he cos s o ca bon p icing. C is a
dummy a iable ep esen ing he ollowing indi idual-le el cha ac e is ics: Male (1=yes, 0=no), Aged 45+ (1=aged 45 o o e , 0=o he wise), Eas (1=li ing in Eas
Ge many, 0=li ing in Wes Ge many), High school (1=yes, 0=no), Single (1=Single household, 0=o he wise), Ruling pa y (1=suppo e s o SPD, B90/Die G ¨unen
o FDP, 0=suppo e s o CDU/CSU, Linke, A D, o he pa ies), High income (1=household income o e4,000 o mo e, 0=less han e4,000), Homeowne (1=yes,
0=no), Fossil hea ing (1=yes, 0=no), Mo o ehicles (1=yes, 0=no). All eg essions include he p e- ea men accep ance o ca bon p icing and he se o con ols
desc ibed in Appendix Table A5. Robus s anda d e o s a e in pa en heses. * p<0.1, ** p<0.05, *** p<0.01
37
Online appendix: Cos pe cep ions and he suppo o ca bon
p icing
Jan Beh inge , Lukas End es, Maike Ko sinnek
Summa y o he appendix
In Sec ion A we show summa y s a is ics, balance es s, and addi ional es ima ion esul s.
Table A1 compa es summa y s a is ics o he ini ial sample wi h he Ge man mic ocensus
o key demog aphic a iables. Table A2 con as s he ini ial sample o ou su ey wi h
he inal sample on which we es ima e ou main esul s. Table A3 and Table A4 p o ide
e idence o co a ia e balance in he ea men and con ol g oups o ou expe imen s. Ta-
ble A5 examines de e minan s o p e- ea men accep ance. Table A6 eplica es ou main
eg ession esul s wi hou addi ional con ols. Table A7 shows di e en ial ea men e -
ec s o esponden s who o e es ima e, unde es ima e, o accu a ely es ima e hei cos s
o ca bon p icing. Table A8 shows how ea men e ec s a y wi h ela i e pe cep ion
gaps. Table A9 examines he e ogeneous ea men e ec s by p e- ea men accep ance.
Table A10 shows ha he es ima ed he e ogeneous e ec s by p e- ea men accep ance
a e obus o using a esidual componen o p e- ea men accep ance. Table A11 p o ides
o de ed p obi es ima es o ou main ea men e ec s. Table A12 shows cos pe cep ions
by p e- ea men accep ance. Table A13 and Table A14 show cos pe cep ions by e-
sponden cha ac e is ics. Table A15 e alua es he ex e nal alidi y o he main esul s by
eweigh ing ou sample o ep esen he gene al popula ion and demons a es he obus -
ness o ou main esul s ega ding su ey ela ed esponse biases such as expe imen e
demand e ec s, dis us in he p o ided in o ma ion, and su ey a igue.
Sec ion B.1 p o ides backg ound in o ma ion on he 2022 mic ocensus. In Sec ion B.2
we desc ibe in de ail he me hod o he calcula ion o esponden s’ CO2 cos s. In Sec ion
B.3 we show English ansla ions o ou main su ey ques ions. Sec ion B.4 p o ides
sc eensho s o he main expe imen ques ions om he online su ey.
38
Table A9: He e ogenei y by p e- ea men accep ance
(1) (2)
45 e/ CO2 200 e/ CO2
T 0.253*** 0.042
(0.054) (0.037)
T×Accep ance=2 0.015 -0.511***
(0.102) (0.078)
T×Accep ance=3 -0.178* -0.371***
(0.099) (0.099)
T×Accep ance=4 -0.227** -0.694***
(0.091) (0.105)
T×Accep ance=5 -0.252** -0.323***
(0.100) (0.125)
Con ols Yes Yes
R-squa ed 0.742 0.722
Obse a ions 1630 1624
No es: The able p esen s es ima ion esul s om OLS e-
g essions. Resul s o he cu en p ice expe imen a e shown
in Column 1 and esul s o he p ojec ed p ice expe imen
a e shown in Column 2. The dependen a iable is he pos -
ea men accep ance o ca bon p icing, measu ed on a i e-
poin Like scale (1= e y unaccep able o 5= e y accep able).
T is a dummy a iable indica ing ha a esponden ecei ed
pe sonalized in o ma ion abou he cos s o ca bon p icing. All
eg essions include he se o con ols desc ibed in Appendix
Table A5. Robus s anda d e o s a e in pa en heses. * p<0.1,
** p<0.05, *** p<0.01
45
Table A10: He e ogenei y by esidual p e- ea men accep ance
45 e/ CO2 200 e/ CO2
(1) (2) (3) (4) (5) (6)
T 0.346*** 0.160*** 0.255 0.066 -0.299*** -0.209
(0.067) (0.034) (0.199) (0.058) (0.035) (0.190)
Accep ance 0.878*** 0.865***
(0.017) (0.018)
T×Accep ance -0.076*** -0.146***
(0.023) (0.026)
Res. accep ance 0.899*** 0.867***
(0.020) (0.020)
T×Res. accep ance -0.115*** -0.153***
(0.032) (0.032)
P ed. accep ance 1.188*** 0.960***
(0.206) (0.236)
T×P ed. accep ance -0.007 -0.040
(0.080) (0.077)
Con ols Yes Yes Yes Yes Yes Yes
R-squa ed 0.742 0.743 0.250 0.711 0.710 0.246
Obse a ions 1630 1630 1630 1624 1624 1624
No es: The able p esen s es ima ion esul s om OLS eg essions. Resul s o he cu en p ice expe imen a e
shown in Columns 1 o 3 and esul s o he p ojec ed p ice expe imen a e shown in Columns 4 o 6. In Columns
1 and 4, we include he p e- ea men accep ance o ca bon p icing. Fo he o he eg essions, we decompose he
o al a ia ion in p e- ea men accep ance o ca bon p icing in o a componen p edic ed by he se o con ol
a iables we use h oughou he pape , and a esidual componen ha is no explained by hese a iables. In
Columns 2 and 5, we include he esidual componen o p e- ea men accep ance. In Columns 3 and 6, we include
he p edic ed componen o p e- ea men accep ance. The dependen a iable is he pos - ea men accep ance o
ca bon p icing, measu ed on a i e-poin Like scale (1= e y unaccep able o 5= e y accep able). T is a dummy
a iable indica ing ha a esponden ecei ed pe sonalized in o ma ion abou he cos s o ca bon p icing. All
eg essions include he se o con ols desc ibed in Appendix Table A5. Robus s anda d e o s a e in pa en heses.
* p<0.1, ** p<0.05, *** p<0.01
46
Table A11: O de ed p obi es ima ions o ea men e ec s
(1) (2) (3) (4) (5)
Panel A: 45 e/ CO2 Acc.=1 Acc.=2 Acc.=3 Acc.=4 Acc.=5
T -0.051*** 0.000 0.006*** 0.028*** 0.017***
(0.011) (0.002) (0.002) (0.006) (0.004)
Con ols Yes Yes Yes Yes Yes
Pseudo R-squa ed 0.415 0.415 0.415 0.415 0.415
Obse a ions 1630 1630 1630 1630 1630
(1) (2) (3) (4) (5)
Panel B: 200 e/ CO2 Acc.=1 Acc.=2 Acc.=3 Acc.=4 Acc.=5
T 0.090*** 0.006 -0.017*** -0.055*** -0.024***
(0.011) (0.004) (0.003) (0.007) (0.004)
Con ols Yes Yes Yes Yes Yes
Pseudo R-squa ed 0.410 0.410 0.410 0.410 0.410
Obse a ions 1624 1624 1624 1624 1624
No es: The able p esen s a e age ma ginal ea men e ec s om o de ed p obi eg essions.
Resul s o he cu en p ice expe imen a e shown in Panel A and esul s o he p ojec ed p ice
expe imen a e shown in Panel B. The dependen a iable is he pos - ea men accep ance o ca bon
p icing, measu ed on a i e-poin Like scale (1= e y unaccep able o 5= e y accep able). T is a
dummy a iable indica ing ha a esponden ecei ed pe sonalized in o ma ion abou he cos s o
ca bon p icing. All eg essions include he p e- ea men accep ance o ca bon p icing and he se
o con ols desc ibed in Appendix Table A5. Robus s anda d e o s a e in pa en heses. * p<0.1,
** p<0.05, *** p<0.01
Table A12: Cos pe cep ions by p e- ea men accep ance
(1) (2) (3) (4) (5)
Panel A: 45 e/ CO2 Acc.=1 Acc.=2 Acc.=3 Acc.=4 Acc.=5
Pe cei ed cos s 519.441 402.866 335.878 324.316 250.654
Ac ual cos s 216.931 197.086 175.599 191.394 169.149
Pe cep ion gap 302.510 205.780 160.279 132.921 81.505
O e es ima ion 0.713 0.653 0.631 0.572 0.579
Obse a ions 540 343 320 320 107
(1) (2) (3) (4) (5)
Panel B: 200 e/ CO2 Acc.=1 Acc.=2 Acc.=3 Acc.=4 Acc.=5
Pe cei ed cos s 786.310 514.375 475.833 479.536 263.364
Ac ual cos s 949.527 881.252 791.325 834.672 717.996
Pe cep ion gap -163.217 -366.877 -315.492 -355.137 -454.632
O e es ima ion 0.349 0.266 0.274 0.243 0.200
Obse a ions 510 349 317 338 110
No es: The able p esen s summa y s a is ics o key a iables ac oss di e en le els o p e-
ea men accep ance in he cu en p ice (Panel A) and p ojec ed p ice (Panel B) expe imen s.
47
Table A13: Cos pe cep ions by esponden cha ac e is ics (cu en p ice expe imen )
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
No high High
Female Male Aged <45 Aged 45+ Wes Eas school school No single Single
Pe cei ed cos s 426.306 382.292* 436.879 384.215* 390.387 473.443** 396.723 409.229 450.504 279.670***
Ac ual cos s 177.805 212.975*** 195.151 197.229 197.977 188.127 182.214 211.014*** 230.443 108.545***
Pe cep ion gap 248.501 169.317*** 241.728 186.986** 192.410 285.316*** 214.508 198.215 220.062 171.125*
O e es ima ion 0.688 0.612*** 0.660 0.641 0.631 0.744*** 0.674 0.621** 0.620 0.720***
Obse a ions 764 866 579 1051 1384 246 822 808 1176 454
(11) (12) (13) (14) (15) (16) (17) (18) (19) (20)
Non- uling Ruling Low High No home- Home- No ossil Fossil No mo o Mo o
pa y pa y income income owne owne hea ing hea ing ehicles ehicles
Pe cei ed cos s 460.318 288.309*** 384.450 448.376** 369.389 448.836*** 360.917 413.952 214.275 445.822***
Ac ual cos s 213.217 176.362*** 166.348 270.663*** 130.297 287.122*** 101.489 221.437*** 49.449 229.929***
Pe cep ion gap 247.101 111.946*** 218.102 177.713 239.092 161.714*** 259.429 192.515** 164.826 215.893
O e es ima ion 0.663 0.589*** 0.687 0.552*** 0.743 0.517*** 0.820 0.603*** 0.808 0.611***
Obse a ions 884 450 1159 471 942 688 339 1291 302 1328
No es: The able p esen s summa y s a is ics o key a iables ac oss esponden cha ac e is ics in he cu en p ice expe imen . We conduc ed - es s o equali y o means
wi hin each demog aphic subg oup (e.g., emale s. male). * p<0.1, ** p<0.05, *** p<0.01
48
Table A14: Cos pe cep ions by esponden cha ac e is ics (p ojec ed p ice expe imen )
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
No high High
Female Male Aged <45 Aged 45+ Wes Eas school school No single Single
Pe cei ed cos s 574.255 562.479 699.578 493.911*** 549.400 671.177** 544.967 590.914 640.437 403.732***
Ac ual cos s 815.510 907.487*** 918.554 833.889** 862.943 872.400 774.185 954.146*** 1031.309 485.873***
Pe cep ion gap -241.255 -345.009** -218.976 -339.977** -313.542 -201.222* -229.218 -363.232*** -390.873 -82.141***
O e es ima ion 0.323 0.250*** 0.333 0.257*** 0.275 0.335* 0.295 0.274 0.254 0.354***
Obse a ions 761 863 585 1039 1376 248 810 814 1127 497
(11) (12) (13) (14) (15) (16) (17) (18) (19) (20)
Non- uling Ruling Low High No home- Home- No ossil Fossil No mo o Mo o
pa y pa y income income owne owne hea ing hea ing ehicles ehicles
Pe cei ed cos s 668.588 406.331*** 485.687 765.920*** 503.322 666.163*** 526.930 578.351 304.220 627.765***
Ac ual cos s 947.273 798.459*** 690.612 1282.249*** 580.431 1295.383*** 471.656 963.402*** 236.197 1006.726***
Pe cep ion gap -278.685 -392.127** -204.925 -516.329*** -77.110 -629.220*** 55.273 -385.052*** 68.023 -378.961***
O e es ima ion 0.296 0.229*** 0.303 0.241** 0.346 0.191*** 0.520 0.225*** 0.503 0.235***
Obse a ions 864 480 1147 477 979 645 327 1297 300 1324
No es: The able p esen s summa y s a is ics o key a iables ac oss esponden cha ac e is ics in he p ojec ed p ice expe imen . We conduc ed - es s o equali y o means
wi hin each demog aphic subg oup (e.g., emale s. male). * p<0.1, ** p<0.05, *** p<0.01
49
Table A15: Robus ness
(1) (2) (3) (4) (5) (6)
Panel A: Sel - Feedback Feedback
45 e/ CO2 Baseline Reweigh ing moni o ing Plausibili y in e es leng h
T 0.161*** 0.123*** 0.157*** 0.158*** 0.159*** 0.176***
(0.034) (0.037) (0.036) (0.037) (0.035) (0.043)
Con ols Yes Yes Yes Yes Yes Yes
R-squa ed 0.741 0.741 0.756 0.745 0.742 0.743
Obse a ions 1630 1630 1453 1519 1596 1062
(1) (2) (3) (4) (5) (6)
Panel B: Sel - Feedback Feedback
200 e/ CO2 Baseline Reweigh ing moni o ing Plausibili y in e es leng h
T -0.307*** -0.304*** -0.298*** -0.311*** -0.307*** -0.311***
(0.035) (0.038) (0.036) (0.036) (0.035) (0.044)
Con ols Yes Yes Yes Yes Yes Yes
R-squa ed 0.712 0.712 0.720 0.720 0.712 0.702
Obse a ions 1624 1624 1451 1497 1576 1058
No es: The able p esen s es ima ion esul s om OLS eg essions. Resul s o he cu en p ice expe imen a e
shown in Panel A and esul s o he p ojec ed p ice expe imen a e shown in Panel B. Columns 1 and 2 show he
esul s o he unweigh ed and eweigh ed sample and Columns 3 o 6 show he esul s o a ious subsamples.
In Column 3, we exclude esponden s wi h high le els o sel -moni o ing. Fo his pu pose, we cons uc an
index based on ou i ems om he Ge man e sion o he sel -moni o ing scale by Schyns and Paul (2002) and
exclude he op decile. In Column 4, we exclude esponden s ha ind he p o ided in o ma ion unplausible.
In Column 5, we exclude esponden s ha ind he su ey ( a he ) unin e es ing. In Column 6, we exclude
esponden s ha ind he su ey ( a he ) oo long. The dependen a iable is he pos - ea men accep ance o
ca bon p icing, measu ed on a i e-poin Like scale (1= e y unaccep able o 5= e y accep able). T is a dummy
a iable indica ing ha a esponden ecei ed pe sonalized in o ma ion abou he cos s o ca bon p icing. All
eg essions include he p e- ea men accep ance o ca bon p icing and he se o con ols desc ibed in Appendix
Table A5. Robus s anda d e o s a e in pa en heses. * p<0.1, ** p<0.05, *** p<0.01
50
B. Da a appendix
B.1. Addi ional in o ma ion on he 2022 mic ocensus
The mic ocensus is Ge many’s la ges annual gene al popula ion su ey, conduc ed by
he o icial s a is ical au ho i ies. I employs a s a i ied clus e sampling design in which
all membe s o households in andomly selec ed dis ic s a e legally equi ed o pa ici-
pa e. Wi h app oxima ely 810,000 esponden s, he su ey co e s oughly 1 pe cen o
Ge many’s o al popula ion. Fo ou analysis we use he mos ecen a ailable da ase
om 2022. Due o da a p o ec ion easons, he Scien i ic Use File consis s o a ep esen-
a i e 70 pe cen subsample, o alling 683,588 indi idual obse a ions. We es ic he
sample o esponden s aged 18 o 75, esiding a hei main esidence o ma ch ou own
su ey. This lea es us wi h 488,363 indi idual obse a ions in ou inal da ase . A de-
ailed documen a ion o he 2022 mic ocensus is p o ided by he Fede al S a is ical O ice
(2023).
B.2. Cos calcula ion
To illus a e he cos calcula ion me hod, conside he ollowing example: A amily o
ou en s a 120 m2apa men wi h an oil hea ing sys em o bo h space hea ing and
ho wa e supply. The amily owns wo ca s - one wi h a gasoline engine and one wi h a
diesel engine - and has d i en 5,000 kilome e s in he pas wel e mon hs. To de e mine
o al addi ional annual cos s (H6), we i s calcula e se e al auxilia y a iables (H1-H5).
We s a by es ima ing he amily’s emissions om space hea ing and ho wa e supply
(H1). To app oxima e he amily’s ene gy use o space hea ing, we mul iply he size o
hei dwelling by he a e age ene gy consump ion pe squa e me e , as p o ided by he
En i onmen al-Economic Accoun s o he Fede al S a is ical O ice (2022). We ob ain he
amily’s ene gy use o ho wa e consump ion by mul iplying he household size by he
a e age ho wa e consump ion pe pe son also sou ced om he En i onmen al-Economic
Accoun s. Bo h alues a e hen mul iplied by an ene gy-sou ce-speci ic emission coe icien
o oil o ansla e consump ion in o CO2 emissions in ons. The amily’s o al emissions
51
om hea ing and ho wa e use a e he e o e gi en by:
H1 = 120 m2×(133 ×0.000266) CO2
m2+ (4 ×1,280 ×0.000266) CO2= 5.60728 CO2
Ge man law (CO2 Cos Alloca ion Ac ) dic a es ha he cos s o ca bon p icing o
emissions om hea consump ion in en al housing ha e o be sha ed be ween landlo ds
and enan s. The sha e o ca bon cos s bo ne by he enan is de e mined based on he
dwelling’s ene gy e iciency, measu ed in e ms o CO2 emissions pe squa e me e o li ing
space. To adjus enan s’ cos s acco dingly, we i s de e mine he ene gy e iciency o
each enan ’s dwelling (H2) by di iding o al emissions om space hea ing and ho wa e
supply by he dwelling’s size:
H2 = 5.60728 CO2
120m2= 0.046727 CO2
m2
As s a ed by he CO2 Cos Alloca ion Ac , he enan ’s sha e o CO2 cos s (H3) is
de e mined acco ding o he ollowing scale:
H3 =
1,i H2 <0.012
0.9,i 0.012 ≤H2 <0.017
0.8,i 0.017 ≤H2 <0.022
0.7,i 0.022 ≤H2 <0.027
0.6,i 0.027 ≤H2 <0.032
0.5,i 0.032 ≤H2 <0.037
0.4,i 0.037 ≤H2 <0.042
0.3,i 0.042 ≤H2 <0.047
0.2,i 0.047 ≤H2 <0.052
0.05,i H2 ≥0.052
Because o he low ene gy e iciency o ou exempla y amily’s housing, hey only bea
he cos s o 30 pe cen o o al emissions om space hea ing and ho wa e supply.
The e o e, we calcula e he CO2 emissions ha he household has o e ec i ely pay o
52
(H4) by mul iplying household emissions by 0.3 (H3):
H4 = 0.3×5.60728 CO2= 1.682184 CO2
Nex , we calcula e he amily’s ca bon emission om anspo a ion (H5). We begin by
es ima ing he household’s gasoline and diesel consump ion in li e s by appo ioning o al
mileage ac oss gasoline and diesel ehicles based on he household’s lee and applying
a e age diesel and gasoline consump ion pe kilome e om he “T anspo in Figu es
2022/2023” epo by he Fede al Minis y o Digi al and T anspo (2022). We mul iply
hese a e ages by uel-speci ic CO2 emissions pe li e , which a e 0.00265 o diesel and
0.00237 o gasoline. Fo he amily owning one diesel and one gasoline ca , he o mula
o o al anspo emissions o a a elled dis ance o 5000 kilome e s is:
H5 = 5000 km ×(0.07 L
km ×0.00265 CO2
L×1
2+ 0.077 L
km ×0.00237 CO2
L×1
2)
= 0.919975 CO2
Finally, we compu e o al addi ional cu en cos s o ca bon p icing (H6) by agg ega ing
emissions om household hea ing and anspo a ion and mul iplying o al emissions by
he cu en ca bon p ice o e45/ CO2:28
H6 = (0.919975 CO2+ 1.682184 CO2)×
e45
CO2
=e117.097155
Thus, he amily has cu en addi ional annual cos s o ca bon p icing o e117 ( ounded
o he nea es eu o).
B.3. Main expe imen : Su ey ques ions
Q1. Age
How old a e you?
Age in yea s
28Al e na i ely, we calcula e p ojec ed u u e cos s by mul iplying o al emissions by he p ojec ed ca bon
p ice o e200/ CO2. Ou example amily would ha e u u e cos s o e520.
53
Q2. Gende
Please en e you gende :
[ ] Male
[ ] Female
[ ] Di e se
Q3. S a e
In which ede al s a e do you li e?
[ ] Baden-W¨u embe g
[ ] Ba a ia
[ ] Be lin
[ ] B andenbu g
[ ] B emen
[ ] Hambu g
[ ] Hesse
[ ] Mecklenbu g-Wes e n Pome ania
[ ] Lowe Saxony
[ ] No h Rhine-Wes phalia
[ ] Rhineland-Pala ina e
[ ] Saa land
[ ] Saxony
[ ] Saxony-Anhal
[ ] Schleswig-Hols ein
[ ] Thu ingia
[ ] I do no li e in Ge many
Q4. Household income
Wha is you household’s cu en o al mon hly ne income?
This e e s o he o al amoun om wages, sala ies, income om sel -employmen , e-
i emen pensions o ci il se ice pensions, each a e deduc ing axes and social secu i y
con ibu ions. Please also include income om public assis ance, en al o lease income,
housing bene i s, child bene i s, and any o he sou ces o income.
A household is de ined as people who li e oge he and sha e inances, ha is, hey co e
daily li ing expenses oge he and do no accoun o hei pu chases sepa a ely.
I you don’ know he exac amoun , please p o ide an es ima e.
[ ] Less han e500
[ ] e500 o less han e1,000
[ ] e1,000 o less han e1,500
[ ] e1,500 o less han e2,000
[ ] e2,000 o less han e2,500
[ ] e2,500 o less han e3,000
[ ] e3,000 o less han e3,500
[ ] e3,500 o less han e4,000
[ ] e4,000 o less han e4,500
54
61
62
B.4.2. P ojec ed p ice expe imen
63
64
Re e ences Appendix
Fede al Minis y o Digi al and T anspo (2022), ‘Ve keh in Zahlen 2022/2023’.
Fede al S a is ical O ice (2022), ‘Umwel ¨okonomische Gesam echnungen - P i a e
Haushal e und Umwel - Be ich szei aum 2000-2020’.
Fede al S a is ical O ice (2023), ‘Quali ¨a sbe ich Mik ozensus 2022’.
65
Imp in
Publishe
Mac oeconomic Policy Ins i u e (IMK) o Hans-Böckle -Founda ion, Geo g-Glock-S . 18,
40474 Düsseldo , Ge many, phone +49 211 7778-312, email imk-publika ionen@boeckle .de
IMK Wo king Pape is an i egula online publica ion se ies a ailable a :
h ps://www.imk-boeckle .de/de/imk-wo king-pape -15378.h m
The iews exp essed in his pape do no necessa ily e lec hose o he IMK o he Hans-Böckle -Founda ion.
ISSN 1861-2199
This publica ion is licensed unde he C ea i e commons license:
A ibu ion 4.0 In e na ional (CC BY).
P o ided ha he au ho 's name is acknowledged, his license pe mi s he edi ing, ep oduc ion and dis ibu ion o he ma e ial in
any o ma o medium o any pu pose, including comme cial use.
The comple e license ex can be ound he e: h ps://c ea i ecommons.o g/licenses/by/4.0/legalcode
The e ms o he C ea i e Commons License apply o o iginal ma e ial only. The e-use o ma e ial om o he sou ces (ma ked wi h
sou ce) such as g aphs, ables, pho os and ex s may equi e u he pe mission om he copy igh holde .
