ArticlePDF Available

Abstract and Figures

Our study examines public acceptability of the EU’s future mitigation targets. Using the discrete choice experiment, we elicit the preferences of about 4,098 respondents from the Czech Republic, Poland, and the United Kingdom for the GHG emission reduction policies that differ in four attributes: emission reduction target, burden sharing across the EU Member States, the distribution of costs within each country, and cost. The three specific reduction targets we analysed correspond to the EU 2050 Roadmap and deep decarbonisation policy (80% target), the climate-energy 2014 targets (40% target), and the status quo policy (20% target); each will result in a specific emission trajectory by 2050. Our results reveal stark differences between the three countries. Czechs would be on average willing to pay around EUR 13 per household and year for the 40% GHG emission reductions by 2030 or EUR 17 for 80% reduction target by 2050, and the citizens of the UK are willing to pay about EUR 40. Conversely, the mean willingness to pay of the Polish household for adopting more stringent targets is not statistically different from zero. The willingness to pay for adopting 40% and 80% targets are not statistically different in any of the examined countries. However, we found that the preferences in all three countries are highly heterogeneous. In addition, we provide an insight into the preferred characteristics of the future GHG emission reduction policies.
Content may be subject to copyright.
1
Public Acceptability of Climate Change Mitigation Policies:
A Discrete Choice Experiment
Milan Ščasný*, Iva Zvěřinová*, Mikolaj Czajkowski#, Eva Kyselá*, Katarzyna Zagórska#
* Charles University Prague, Environment Center, Czech Republic
# University of Warsaw, Department of Economics, Poland
Corresponding author: Dr. Milan Ščasný, Charles University Environment Center,
milan.scasny@czp.cuni.cz
1. Introduction
The Paris Agreement - adopted on December 12, 2015 - provides an international framework
for combating climate change, with the key objective of limiting the rise in global temperature
to below 2°C by 2100. All countries are required to share responsibility, but in different
proportions determined primarily by their historic level of responsibility for emissions and their
current level of development.
Policies to comply with the Paris Agreement face the challenge of policy infeasibility.
Decision-makers are usually reluctant to pursue policies that the public is not willing to endorse
or accept. Gaining a better understanding of the public acceptability of different policy
alternatives is thus a valuable tool for overcoming an important barrier to the successful
implementation of climate policy.
The present study contributes to the recent debate by examining the public acceptability of the
EU’s future mitigation policies, yielding future benefits and costs, together with several options
for burden sharing and cost distribution. Specifically, we elicit preferences and estimate
willingness to pay (WTP) of Czech, British, and Polish citizens for policies to reach three
different EU emission reduction targets with no further reductions 20% by 2020, 40% by
2030, and 80% by 2050 and several options to share the burden among the EU countries and
costs among citizens of each country.
2
The three GHG emission reduction targets present three different emission trajectories resulting
in three different global temperature increases and likely impacts by 2100. More specifically,
the status quo policy will result in a 20% reduction in EU emissions by 2020, as already
committed to by the European Commission (Decision 406/2009/EC). According to data
available at the time of our survey, total EU emissions were 19.8% below 1990 levels in 2013
and projected levels with existing measures were expected to be 24% below 1990 levels by
2020 (European Environmental Agency, 2015).
1
In case of the status quo policy, future
emissions remain more or less at their current levels. The 40% reduction by 2030 (on top of the
20% reduction by 2020) corresponds to the climate-energy integrated 40-27-27 target
(European Council, 2014) and submitted in March 2015 as the Intended Nationally Determined
Contributions (INDC) to the UNFCCC before the COP21 meeting in Paris (European
Commission [EC], 2015a). The 80% reduction target by 2050 corresponds to the EU’s
Roadmap for Moving to a Competitive Low-Carbon Economy in 2050 (EC, 2011), and to a
pathway to achieve deeper emission cuts by 2050. These two targets require the implementation
of new policies.
There are several rules to allocate GHG emission reduction effort and related costs. Current EU
policy requires the Member States to participate in this effort, balancing considerations of
fairness and solidarity (European Council 2014, EC 2016). Its burden sharing relies on a mix
of allocation principles; it is based upon proportionality linked to either verified historical
emissions (ETS sectors) or GDP per capita (non-ETS), with adjustments to reflect cost-
effectiveness in a fair and balanced manner, reflecting the principle of progressivity.
This study uses discrete choice experiments to provide an insight into the drivers of public
acceptability of the EU’s future mitigation policies, focusing specifically on preferences for
various burden sharing and cost allocation options. Stated preference data from nationally
representative samples of the Czech Republic, Poland and the UK are analysed.
Our results reveal stark differences among the three countries when preferences are analysed.
The Czech households would on average be willing to pay around EUR 13 for 40% or EUR 17
for 80% GHG emission reductions per month, while the UK households would be willing to
1
This trend has been later confirmed by more recent estimates, as emissions decreased to 23 % below the 1990
levels in 2014 and the projections until 2020 stayed unchanged (EC, 2015b).
3
pay EUR 46 for 40% and EUR 44 for 80% emission reductions. In contrast, the mean WTP of
the Polish households for adopting policies delivering greater emission reductions is not
statistically different from zero, with a preference for the status quo policy. Still, the preferences
in all countries are highly heterogeneous. In each of the three countries, we identify three
distinct classes of respondents who are against, modestly and strongly in favour of a more strict
GHG emission reduction targets. In addition, there is a clear general preference for the polluter-
pays principle at both national and EU levels, with the exception of Poland.
2. Literature review
In general, the public acceptability of policies is influenced by at least two types of factors and
their interactions
2
: individual characteristics and the characteristics of the policy. Looking
first at individual characteristics, social-psychological studies have identified the positive
effect of biospheric and pro-environmental values and awareness of the consequences of
climate change (e.g. T. Dietz, Dan, & Shwom, 2007; Leiserowitz, 2006; Poortinga, Spence,
Demski, & Pidgeon, 2012; Steg, Dreijerink, & Abrahamse, 2005), personal and social norms
(Cools et al., 2011; Harring & Jagers, 2013; Schade & Schlag, 2003), trust (Dietz et al., 2007;
Harring & Jagers, 2013; Sælen & Kallbekken, 2011) and actual knowledge about climate
change (T. Dietz et al., 2007; O’Connor, Bard, & Fisher, 1999). However, the evidence for the
influence of several socio-demographic characteristics is not conclusive (Zvěřinová, Ščasný,
& Kyselá, 2014).
Turning to the second type of factors, several policy characteristics have been studied as
potentially influential. Studies aimed at specific policy instruments, such as permits versus
taxes (Bristow, Wardman, Zanni, & Chintakayala, 2010) or forestry practices (Layton &
Levine, 2003). Ancillary benefits of the climate change mitigation policy which might be quite
substantial (Ščasný, Massetti, Melichar, & Carrara, 2015) may also play quite a decisive role
in respondents’ decision making.
Three main categories of policy characteristics can be distinguished as important influences on
public acceptability: perceived policy effectiveness, fairness linked either to distribution of
2
The effects of policy characteristics are interlinked with social-psychological variables mentioned above and
hence the individual characteristics of the respondent (Kim, Schmöcker, Fujii, & Noland, 2013).
4
costs or burden-sharing rules, and the use of revenues. In this study, we focus on effectiveness
and fairness as attributes.
People are more likely to vote for or be willing to pay for policies that they perceive as
effective, i.e. that a policy will achieve its objectives. Overall, the price that people are willing
to pay for improved environmental effectiveness is quite high. Dietz and Atkinson (2010) found
that people from Southwark, UK, are willing to pay EUR 255 or EUR 548 per year (2014
Purchasing Power Standard) for medium and high improvement in national climate-change
mitigation. Many people, however, do not know what consequences may be associated with a
temperature increase of 2 °C (Reynolds, Bostrom, Read, & Morgan, 2010). One way to deal
with this is to specify the effects in the questionnaire. Hence in our questionnaire, we provided
information about the effects of climate change if not mitigated and if mitigated as
corresponding to the three GHG emission reduction targets and the respective emission
trajectories.
Perceived fairness of the distribution of policy costs has been identified as another key factor
influencing WTP for climate policy (Berrens et al. 2004; Li et al. 2004; Wiser 2007; 2003).
While equity is based upon external ethical criteria, fairness is based on the preferences of
individuals and groups, and on the ways in which they perceive themselves in relation to others.
The fairest distributions are those in which there is little or no envy among parties (Fishburn &
Sarin, 1994).
Four allocation principles to determine a claimant’s share in decisions can be distinguished,
according to Young (1994). The leading principle in fairness is proportionality; next include
progressivity, parity, and priority. In environmental context, the allocation rule may be linked
to the allocation of property rights that include polluter-pay and beneficiary-pays principle (see
S. Dietz & Atkinson, 2010). Concern about the regressive impact of a policy may involve
introducing measures to mitigate social impact, following the ability-to-pay principle. Since a
wide variety of distributive principles can be applied, it is difficult to identify the commonalities
and put them to practical use (S. Dietz & Atkinson, 2010). Therefore, it is important to know
public preferences for the various allocation principles and/or the equity-efficiency trade-off.
The burden sharing rule among countries or regions may follow the same principles as
described above. In the vast literature on burden-sharing rules, the most common rules are
5
based on historical emissions or historical responsibility, current emissions (polluter-pays
principle), carbon intensity, income level (ability to pay), per capita emissions (right to emit),
or multi-criteria formulas (Brick & Visser, 2015; Carlsson et al., 2013; Ringius, Torvanger, &
Underdal, 2002).
Studies assessing burden-sharing at the country level either analyse and rank implications of
the respective burden-sharing rules, or discuss rational behaviours in the game theory context
(Cazorla & Toman, 2010; Ringius et al., 2002; Rose, Stevens, Edmonds, & Wise, 1998). In
the ranking by economic impact, developed countries are bearing higher costs under the
historical polluter-pays rule and are relatively better off under the carbon intensity approach,
and for developing countries historical polluter-pays and equal emissions per capita are the
least costly (Brick & Visser, 2015; Miketa & Schrattenholzer, 2006).
Some studies found that citizens prefer the polluter-pays principle (Bechtel & Scheve, 2013;
Schleich, Dütschke, Schwirplies, & Ziegler, 2016) and distribution based on ability to pay (S.
Dietz & Atkinson, 2010) and progressive, rather than regressive, taxation (Brännlund &
Persson, 2012; Cole & Brännlund, 2009). Bechtel and Scheve (2013) explored preferences for
cost distribution within international mitigation agreements and their results suggest a
preference for polluter-pays principle between French, German, UK, and US citizens over the
principle that only rich countries should pay. That rich countries should pay more than poor,
however, had a similar positive effect on willingness to pay, suggesting that the citizens of rich
and emission-intensive economies are ready to accept larger shares of the costs. An empirical
study conducted among students suggests that domestic distribution and payment vehicles are
more important than international cost distribution and burden sharing (Cai, Cameron, &
Gerdes, 2010). However, a study of the general public of the United States and China indicates
that people can still consider international burden sharing rules and presumably prefer options
which impose lower costs on their country (Carlsson et al., 2013).
Neither the study by Carlsson et al. (2013) nor the study by Bechtel and Scheve (2013) include
domestic distribution as an attribute, therefore it is not possible to evaluate which characteristics
are more important for respondents. There is no recent study on the public acceptability of
climate policy that examines the effects of both domestic cost distribution and burden sharing
among the EU Member States simultaneously as two distinct attributes in the choice
experiments. The present study aims to fill this research gap.
6
3. The survey and the study design
3.1. Survey method and data
Nationally representative samples of population aged 18 to 69 years in the Czech Republic, the
United Kingdom and Poland were surveyed in September and October 2015 within the EU
funded project CECILIA2050. The three countries were selected based on their different
political stances in the European climate policy debate and distinct national contexts for the
purpose of comparison.
3
Data were collected using web-based questionnaires administered either by interviewers
personally (CAPI) or by online access panels (CAWI).
4
The combination of the survey modes
was chosen in order to alleviate possible generalizability issues caused by lower internet
coverage rates in the Czech Republic and Poland (78% and 75% respectively of households in
2014; Eurostat, 2014). Since internet coverage reached 90% of British households in 2014
(ibid.), only CAWI questionnaires were distributed in the UK.
Web-based instruments were chosen, as they provide both the possibility for more complicated
experimental designs and randomizations. In the self-administered mode some respondents
may flip through the questions without properly reading them. Therefore, all cases with the
overall time shorter than 48% of median specific for a given CAWI subsample were excluded
from the final sample as speeders (see Supplemental online information, further SI Appendix).
In total, the interview was conducted with almost 5,500 respondents, including 680 respondents
interviewed in the last pilot round. After excluding speeders from the CAWI sample and
3
The Czech Republic belongs to the medium-sized EU countries, while population of the UK and Poland is the
third, or the sixth, respectively, largest among the 28 EU Member States. The UK economy produces the highest
per capita GDP among the three analysed and 10th highest among the EU28. Czech and Polish economies are less
productive, with per capita GDP ranked at 15th and 23rd. Poland and the Czech Republic belong to carbon-
intensive economies, having third and fourth highest volume of GHG emissions per GDP (1.0 and 0.8 Mg per one
billion EUR GDP). On the other hand, the UK economy belongs to one of the least carbon intensive countries
with only 0.3 Mg GHG per billion euros, which is the sixth best performance in the EU28.
4
CAWI and CAPI data yield the same results in the Czech Republic, however, two different survey modes result
in differences in Poland (see SI Appendix). We highlight that due to our sampling strategy, the two sub-samples
were not constructed in a way to allow analysing data separately.
7
deleting the pilot data, the cleaned sample includes in total 4,098 valid observations (see Table
1).
< Table 1 around here >
The country subsamples are representative of national populations aged 18 to 69 years in terms
of gender, age, region and education. The percentages in all categories are not statistically
different from the quota set based on national statistics. Median household monthly income for
the country samples correspond to national statistics as reported by Eurostat (see SI Appendix,
Table S1).
3.2. The instrument
The questionnaire was tested and developed throughout qualitative pre-survey and several pilot
rounds. The elaborated version was tested on a representative sample of the Czech adult
population (N=727) by CAWI mode in September 2014. The last pilot round was carried out
in summer 2015 in all of the surveyed countries (N=680).
The questionnaire is provided in Supplemental online material. It consisted of screening
questions, two discrete choice experiments, several social-psychological constructs,
respondents’ environmentally significant behaviours and socio-demographic characteristics.
Discrete choice experiments were accompanied by questions on perception of policy
characteristics included in the experiments. Respondents could familiarize themselves with
these characteristics before committing a decisive response in the choice experiments. While
the first choice experiment is described and analysed in this paper, the second choice
experiment targeted perceived effectiveness of the policy instruments and preferences for
revenue use with the 80% reduction target set (not analysed here).
3.3. The discrete choice experiment
In order to explore public preferences for climate policies and their characteristics, we designed
a choice experiment containing four attributes. The first three attributes described the policy:
emissions reduction target for the European Union (20% by 2020, 40% by 2040, 80% by 2050),
distribution of costs among the European Union countries, and cost distribution among the
8
citizens of the given country, see the design in Table 2, and an example of the choice card in
Figure 1.
< Table 2 around here >
< Figure 1 around here >
The three GHG emission reduction targets present three different emission trajectories and they
correspond to the EU’s Roadmap (by 80% by 2050), to the EU’s 2014 integrated 40-27-27
targets (by 40% by 2030).
5
The status quo situation represents the current policy mix most
certainly leading to 20% reduction by 2020 with both allocation rules based upon
proportionality principle linked to wealth (GDP), or household income, respectively.
We informed our respondents that after achieving the two less stringent targets, emission
volumes will remain stable or may even slightly increase at least until 2050. It implies that each
stricter target includes all less stringent targets that are presented in our design. As a
consequence of these emission trajectories, by 2100 global average temperature would increase
much more due to the 20% mitigation policy than due to the 80% reduction policy, and would
result in severe, moderate, or mild impacts, respectively (see Figure 2). In line with scientific
knowledge, both the average temperature increase and the impacts are presented as uncertain,
providing the temperature interval and a notion of likeliness.
< Figure 2 around here >
The remaining attribute was the increased monthly costs to the respondent’s household, which
allows us to estimate the WTP for each level of the policy attributes.
6
It was explained to
respondents that introducing fees for polluting technology and supporting new (lower emission)
technologies development would have to be financed mainly by businesses. These higher costs
5
Recall that preferences were elicited under the assumption that non-European countries ‘comply equivalently’.
6
Costs for a household are correlated with certain cost allocation rule, for instance, the proportional allocation
linked to income will increase relative burden of a rich household, whereas lump sum allocation will be more
penalizing poor. We therefore examined whether the respondents connect the own costs with some cost
distribution rule in order to ensure that the attributes are uncorrelated. By interacting income with attributes, we
find that deviations from mean income in the sample do not seem to significantly change the means of the
parameter of cost distributions (the results are provided in SI Annex).
9
would be reflected in the prices of consumer goods. Therefore, all households would have less
money for purchasing of other goods. The costs for a household varied between EUR 20 and
EUR 150 a month and the nominal values were recalculated into national currencies using
purchasing power standard. Both attributes and their levels were chosen based on the results of
the literature review and the qualitative pre-survey.
In each choice situation, respondents were presented with three options, namely two proposed
new policy options and a status quo option, that is, the continuation of the already implemented
policies that will reach the 20% reduction target by 2020, but that will not reach the stricter
target in later years.
The experimental design of our study consisted of 72 choice-tasks, blocked into twelve
questionnaire versions with six choice tasks per respondent. The order of choice tasks in each
version was randomized for each respondent, to mitigate potential anchoring or framing
effects. The design was optimized for D-efficiency (Ferrini & Scarpa, 2007) of the
multinomial logit model using Bayesian priors (Scarpa & Rose, 2008).
4. Modelling preference for climate change mitigation policies
4.1. Econometric model
In this section we infer respondents’ preferences and WTP for changes in the EU climate policy
from the choices they made in the discrete choice experiments (Carson & Czajkowski, 2014).
The state-of-the-art in modelling consumers’ preferences using discrete choice data is the
mixed logit (MXL) model, in which respondent i’s utility associated with choosing alternative
j out of the J available alternatives in choice task t can be expressed as:
ijt ijt i ijt i ijt
V p a e+=+Xb
, (1)
where
X
represents a vector of alternative-specific attributes,
p
is an additively separable
cost, and vectors
b
and
are coefficients. Note that the coefficients are indexed by respondents
in the MXL model respondents’ coefficients can differ and are assumed to follow an a priori
specified multivariate parametric distribution.
10
Assuming the stochastic component e to be identically and independently, extreme value type
1 distributed leads to convenient expression of choice probabilities an individual will choose
alternative j if
ijt ikt
VV
for all
k
and
, and the probability that alternative
j
is chosen from a
set of
J
alternatives becomes
( ) ( ) ( )
( )
( ) ( )
( )
1
exp
|exp
ijt ijt i
J
ikt ikt i
k
i i i
i i i
pa
P j J pa


=
=+
+
b
b
X
X
(2)
Given that we wish to estimate WTP for non-monetary attributes
X
, it is convenient to
introduce a modification which is equivalent to using a money-metric utility function
(estimating preference parameters in WTP space, Train & Weeks, 2005):
( )
ii
ijt i i ijt ijt ijt i i ijt i ijt ijt
ii
U a p a p
a
 

= + = +
++

b
XXβ
. (3)
In this specification, the estimates obtained by a researcher are a product of the scale and
marginal utility of income
ii
a
and the scale-free coefficients
i
β
corresponding to each of the
choice attributes
X
, which can be readily interpreted as respondents’ marginal WTP for non-
monetary attributes.
In the MXL models, all attribute coefficients are random and freely correlated; all coefficients
are assumed to be normally distributed, with the exception of the coefficient representing the
marginal utility of income
ii
a
, which is assumed to follow a lognormal distribution to
constrain its sign. For the lognormally distributed coefficient, the estimated mean and standard
deviation of the underlying normal distribution is reported. The cost enters the model with a
negative sign and was scaled by a factor of 100 to facilitate convergence. Because the
coefficients are assumed random variables following the specified probability distributions, the
model is typically estimated using the maximum simulated likelihood method (Revelt & Train,
1998).
11
We estimate further a latent class (LC) model that tackles preference heterogeneity in a
different way than MXL instead of assuming continuous parametric distributions of
individual parameters it uses a finite number of preference types (latent classes) which are
probability weighted to best describe population level preferences. Each respondents’
probability of membership in any given class may be defined as a function of the respondent’s
characteristics. This approach might be of particular interest, if population level preferences are
better characterized by a mixture of a few discrete preference types rather than a continuous
distribution.
4.2. Marginal WTP for policy attributes
The discrete choice experiment data was used to model respondents’ preferences using the
approach outlined just above.
7
Table 3 presents the results of the MXL models for each of the
countries. The models are estimated in WTP space and hence the coefficients of the non-
monetary attributes can readily be interpreted as marginal WTP.
8
All WTP values are expressed
in nominal 2014 EUR.
< Table 3 about here >
The results reveal stark differences between the countries. With respect to the preferences for
more ambitious targets of the EU climate policies, while the Czech households would be on
average willing to pay around EUR 13 each month for achieving 40% GHG emission
reductions by 2030 with no further reductions or EUR 17 for 80% reductions by 2050, and the
UK households would be willing to pay EUR 46 and EUR 44, respectively, the Polish
households’ mean WTP for adopting these targets vs. the current policy was not statistically
different from zero. The mean WTP values for adopting 40% and 80% targets are not
statistically different from one another in any of the examined countries, indicating the utility
7
We examined whether results are affected by excluding respondents who protested against the contingent
scenario, or who always chose stringent reduction targets but did not consider their costs. We found only a
minor effect on WTP, thus we do not exclude neither group of respondents from the main analysis. These results
are reported in SI Annex.
8
The models were estimated using custom code developed in Matlab which is made available from
github.com/czaj/DCE under Creative Commons by 4.0 license. The maximum likelihood function was simulated
using 10,000 Sobol draws (Czajkowski and Budziński, 2015), using different starting points and optimization
techniques, to make sure it reached the global optimum.
12
change from adopting the 80% reduction vs. the 40% reduction is negligible. In line with these
results, the citizens of the Czech Republic and the UK are generally supportive of the proposed
climate policy, while the Poles prefer the status quo policy to any new policy option. However,
the preferences in all countries are highly heterogeneous, as indicated by the estimates of the
standard deviations of the WTP distributions relative to the means of these distributions.
Regarding the preferences for burden sharing rules among the EU countries, households from
the Czech Republic and the UK significantly prefer the distribution of policy costs based on
the GHG emissions of the EU countries to linear-to-wealth or per-capita based rules, for which
they are statistically indifferent. The mean WTP for the burden sharing linked to GHG emission
volumes is EUR 15 (s.d. EUR 22.1), and EUR 27 (s.d. EUR 47.4), respectively. In contrast,
Polish households are not so eager to implement emission based burden sharing among
countries they are, on average, indifferent between different burden sharing rules.
Finally, with respect to the distribution of policy costs within a country, respondents from all
surveyed countries are in favour of implementing the polluter-pays principle, i.e. cost
distribution based on emissions, and disapprove of even a lump sum distribution of the costs
(per person). The estimated mean WTP for the most preferred option of this attribute is EUR
33 in the UK, EUR 21 in the Czech Republic, and EUR 13 in Poland. In the UK, progressive
cost distribution (the richer should pay more) was preferred to the distribution proportional to
income. However, the emission-based option remained the most preferred. Once again, the
preferences for these attributes are highly heterogeneous, as indicated by the significant and
relatively high estimates of the standard deviations of WTP distributions.
4.3. Total WTP for policy packages
Figure 3 reports the WTP simulation results using the parameter estimates obtained from the
MXL model with the correlated attribute levels.
9
The WTP values for ‘a policy package’ are
simulated for eight different combinations of burden sharing and cost distribution for both 40%
and 80% GHG mitigation policies.
9
SI Annex reports the WTP estimates using the parameter estimates from multinomial logit and mixed logit
without correlated attributes.
13
At first, we found that the WTP values are very similar for the 40% and 80% reduction targets
for the same policy package in each country. The British households are willing to pay the
most, followed by Czech households. The estimated mean WTP of the Polish households is
negative for each policy package departing from the status quo, although the 95% confidence
interval covers positive values for each package with costs proportional to their emission
volumes.
The order of the eight policy packages is very similar in each of the three countries, with the
policies that link the cost distribution to GHG emission volumes considered the best. The
British and the Czech households similarly prefer the policy package that links burden sharing
as well as cost distribution to emission volumes (EU GHG MS GHG), and the estimated
mean WTP per household is EUR 145 and EUR 62 respectively. The second most preferred
policy package varies across the three countries. The British households prefer the EU GHG
MS rich package, followed by packages that either link cost distribution to emissions or
progressively to household income. As the second most preferred, Czech households prefer a
package that consists of some linking to emission volumes (EU GHG MS rich, EU linear
MS GHG, or EU person MS GHG), followed by policies based upon cost progressively
linked to income. Polish households’ WTP for the ‘EU GDP MS GHG package is similar to
that of Czech and UK households. The least preferred package in all cases is the package that
is based upon proportionality principle linked to the size of a country or a family, respectively
EU person MS person.
< Figure 3 around here >
4.4. WTP across segments
Most respondents were generally in favour of the emission reductions (65% of the Czechs, 58%
of the Poles and 55% of the British were in favour of the 40% reduction target; see Figure S1
in SI Annex). As found in the MXL models, WTP for the stringent reductions varies
considerably. This heterogeneity is also confirmed by the LC models. In each country, we
identified three different classes that we may call ‘against’, ‘modest’, and ‘green.
14
Respondents who are ‘against’ strongly preferred the status quo policy and have negative or
zero marginal WTP for stringent reductions. The modest’ class consisted of respondents who
support stringent policies by modest WTP, but also have negative WTP for the status quo
option. The ‘green’ class was characterized by respondents who were found to be willing to
pay the largest amounts for the stringent reduction targets and have disutility from the status
quo. While there were 44% and 39% ‘green’ supporters in the UK and in the Czech Republic,
respectively, there were only 25% respondents in the green’ class in Poland, where ‘against’
with 45% is the dominant class. Respondents in the ‘green’ class strongly preferred the
proportional principle of distribution linked to emission volumes, but they disliked the lump
sum cost allocation (see Table 4 and SI Annex for details).
< Table 4 about here >
Less educated respondents were more likely to be in the against’ class in each country,
although the effect of education was not consistent across various levels of education. The
Czech males and older respondents were more likely to belong to the ‘against’ class. Males
were also more likely to be members of the ‘green’ class in the Czech Republic and in the UK,
indicating that females were more likely to be in the ‘modest’ class. The Poles with lower
education were more likely to be among ‘greens’, while respondents who completed a
vocational school in the Czech Republic, or bachelor degree in the UK were less likely to be
‘green. Effects of income, household size, or number of children in household were not
significant.
If respondents think that it is very unlikely that other countries in the world will reduce their
emissions adequately, they are willing to pay more for the SQ option in each country (EUR 36
in CZ, EUR 54 in UK, and EUR 39 in PL; see SI Annex). WTP is also larger for both stringent
reductions in the Czech Republic (EUR 7 and EUR 10).
The share of respondents who believe that the stricter emission reduction policies will be
implemented is largest among the Poles (23%), compared to the Czech (13%) and British
(18%). The believers have a much stronger preference for policies with stricter reductions. The
WTP of Polish believers for policy package with stringent emission reductions is weakened by
negative marginal WTP for the 40% and 80% targets. On the contrary, respondents who do not
think that stringent mitigation policy will be implemented strongly prefer the status quo; hence
15
they are willing to pay much less for the policy package with stringent reduction targets (see SI
Annex).
5. Conclusions
Our study examines public acceptability of climate mitigation policy in three EU member
states, focusing on various emission reduction targets and cost distributions. We conducted a
survey to elicit the preferences of Czech, Polish and British citizens aged 18 to 69 years for
three different policy strategies differing in the planned emission reduction targets: 20%
reduction reached by 2020 with no further reductions, which reflects the current policy
situation, 40% reduction by 2030 (on top of the 20% reduction by 2020) agreed by the EU
Council (EC 2014) and submitted as the INDC to the COP 21 Agreement, and a more stringent
80% reduction target by 2050 (in addition to the earlier reductions). We note that these three
targets actually correspond to three different EU GHG emission trajectories, and consequently
lead to three different projected increases in global average temperature by 2100 and likely
impacts. Preferences for reaching these three GHG reduction targets are elicited through a
policy package that proposes several options to share the burden among EU Member States and
to distribute the costs among citizens of the examined countries.
Specifically, we estimated the marginal willingness to pay for each policy attribute using
discrete choice experiments and stated preference data, and then simulated the WTP for several
different policy packages. For the most preferred policy package linking burden sharing as well
as cost distribution proportionally to emission volumes, households would be on average
willing to pay almost EUR 62 a month in the Czech Republic and EUR 145 in the UK. For
whole populations, these values can be translated into annual benefits of EUR 0.27 billion and
EUR 2.72 billion for the 40% reduction policy alternative. Lower WTP values in the Czech
Republic are understandable given the negative experience of the citizens with previous policy
failures (Rais, 2014). It seems that the Poles are indifferent between the status quo (20%
reduction) and the two more strict targets, implying that the Poles are not ready to pay for more
strict GHG reducing policy, which corresponds to the lower share (55%) of Polish citizens in
favour of the 40% reduction target.
16
The benefits that we estimated can be compared to costs of policies. According to EC (2011),
the EU would need to invest on average 1.5% of its GDP annually to reach the 2050 target,
while AR5 (IPCC 2014) reports a median estimate of the costs to reach 450 ppm CO2eq
concentration target by 2100 at 1.7% of GDP in 2030 and 3.4% in 2050. This cost range can
be translated in around EUR 33-80 per month and household in Poland, EUR 40-110 in the
Czech Republic, and EUR 120-280 in the UK. For most of the combinations of stringent policy
packages (as reported in Figure 3), the benefits exceed the lower bound of the costs in Czech
Republic and UK, but the benefits do not exceed the costs if the costs are close to the upper
bound estimate, or if the mitigation policy uses a less favoured cost allocation and/or burden
sharing rule. The benefits do not exceed the costs in any case in Poland.
Still, how GHG emissions might be reduced is also a concern in Poland as well as in the other
two countries. The WTP for climate policy depends on the distribution of policy costs both
among citizens of the country and among countries of the EU. There is a clear preference for
distribution based on emissions in both instances in the UK and the Czech Republic, which
lends support to previous conclusions about a preference for polluter-pays principle (Carlsson
et al., 2013; Dietz & Atkinson, 2010; Schleich et al., 2016). The estimated values for the
emission-based distribution of costs are comparable to those estimated in previous studies:
Carlsson and colleagues (2013) report EUR 16 (2014 PPP, survey in 2009) per month for U.S.
households for international distribution based on current responsibility for emissions
(compared to EUR PPS 25 in the Czech Republic and EUR PPS 22 in the UK reported here for
EU-GHG) and Dietz and Atkinson (2010) report EUR 11 a month (2014 PPP, survey in 2005)
for polluter-pays option of the London transport scheme (WTP for MS-GHG is EUR PPS 35
in CZ and EUR PPS 27 in the UK).
However, Polish citizens were indifferent between the different principles in the case of
distribution among the EU countries, which can be indicative of self-serving bias (Carlsson et
al., 2013) given the known high coal and emission intensity of the Polish industry. The
preferences for cost distribution options were highly heterogeneous. Although the class of
respondents who are ‘against’ stricter emission reductions is the dominant group in Poland
(45%), there are also 25% of ‘green’ supporters and 29% of ‘modest’ respondents. The Polish
‘green’ class shared preference for the cost distribution linked to emission levels with ‘green’
supporters in other countries. The ‘against’ classes in Poland and UK were found to be
indifferent with respect to the cost allocation rule, as they do not like the mitigation policy at
17
all. Moreover, the members of the ‘against’ class in Poland didn’t care much about any policy
attributes and strongly preferred the current policy.
The different WTP values found among the countries analysed in this study are likely to be a
consequence of the different political and socioeconomic situations. Even though both Poland
and the Czech Republic belong to post-communist countries with carbon-intensive economies
(having the third and fourth highest volume of GHG emissions per GDP in the EU), the Polish
situation differs from the Czech context. Poland is much more dependent on hydrocarbon fuels,
mostly coal. Energy security, stability of supply or employment arguments are often used by
industry, the power lobby and labour unions to convince the Polish policy-makers to rely on
domestic coal for the needs of the country’s power sector (Bukowski, Maśnicki, Śniegocki, &
Trzeciakowski, 2015). Indeed, during the negotiations within the EU before COP21 in Paris,
both Poland and the Czech Republic called for additional compensatory mechanisms and
opposed the adoption of binding renewable and energy efficiency targets (Joint Statement,
2014), with Poland playing a leading role in this opposition. Moreover, Poland declared upon
signature of the Paris Agreement that Poland is not obliged to provide financial resources to
assist developing country Parties with mitigation and adaptation efforts.
The Czech government seems to be more in favour of introducing measures to achieve more
stringent emission reduction targets. The State Energy Policy (2015) has already been
expecting a gradually decreasing ratio of fossil fuels to primary energy sources, from 77% in
2015 to 56% in 2040. A new proposal for the Climate Change Act that reflects the 80%
reduction target by 2050 is going to be prepared for the Czech Government by September 2016.
The United Kingdom was the first EU member state that approved the Climate Change Act in
2008 that has set binding GHG emission reduction targets; 34% by 2020 and 80% by 2050,
with a detailed five-year plan to prepare or implement specific measures.
Public acceptability of policies to reach the EU’s GHG emission reduction targets may be
increased by taking into account the distributional consequences, especially by introducing a
distribution of costs based on the emissions of individual households, i.e. implementing the
polluter-pays principle (in all three surveyed countries) and of the EU Member States.
Special attention should be paid to the communication of climate change policy design to the
Polish public, as the Poles are, contrary to the Czechs and the British, on average not willing to
18
financially contribute to support a stricter GHG reduction policy. We found that people prefer
the stricter emission targets more if they believe that the policy is likely to be implemented.
This suggests that efforts to provide clear information about the policy roadmap supported by
a binding policy commitment may be effective at increasing public support for such policies.
19
References
Berrens, R. P., Bohara, A. K., Jenkins-Smith, H. C., Silva, C. L., & Weimer, D. L. (2004).
Information and effort in contingent valuation surveys: application to global climate
change using national internet samples. Journal of Environmental Economics and
Management, 47(2), 331363. http://doi.org/10.1016/S0095-0696(03)00094-9
Bechtel, M. M., & Scheve, K. F. (2013). Mass support for global climate agreements depends
on institutional design. Proceedings of the National Academy of Sciences, 110(34),
1376313768. http://doi.org/10.1073/pnas.1306374110
Brännlund, R., & Persson, L. (2012). To tax, or not to tax: preferences for climate policy
attributes. Climate Policy, 12(6), 704721.
http://doi.org/10.1080/14693062.2012.675732
Brick, K., & Visser, M. (2015). What is fair? An experimental guide to climate negotiations.
European Economic Review, 74, 7995.
http://doi.org/10.1016/j.euroecorev.2014.11.010
Bristow, A. L., Wardman, M., Zanni, A. M., & Chintakayala, P. K. (2010). Public
acceptability of personal carbon trading and carbon tax. Ecological Economics, 69(9),
18241837. http://doi.org/10.1016/j.ecolecon.2010.04.021
Bukowski M., Maśnicki J., Śniegocki A., & Trzeciakowski R. (2015).Quo vadis? Prospects
for the development of the coal mining sector in Poland. WISE Institute.
Cai, B., Cameron, T. A., & Gerdes, G. R. (2010). Distributional Preferences and the
Incidence of Costs and Benefits in Climate Change Policy. Environmental and
Resource Economics, 46(4), 429458. http://doi.org/10.1007/s10640-010-9348-7
Carlsson, F., Kataria, M., Krupnick, A., Lampi, E., Löfgren, Å., Qin, P., & Sterner, T. (2013).
A fair share: Burden-sharing preferences in the United States and China. Resource
and Energy Economics, 35(1), 117. http://doi.org/10.1016/j.reseneeco.2012.11.001
Carson, R. T., Louviere, J. J., & Wei, E. (2010). Alternative Australian climate change plans:
The public’s views. Energy Policy, 38(2), 902911.
http://doi.org/10.1016/j.enpol.2009.10.041
Carson, R. T., & Czajkowski, M. (2014). The discrete choice experiment approach to
environmental contingent valuation. In S. Hess & A. Daly (Eds.), Handbook of Choice
Modelling (pp. 202235). Edward Elgar. Retrieved from
http://econpapers.repec.org/bookchap/elgeechap/14820_5f9.htm
20
Cazorla, M. V., & Toman, M. A. P. (2010). International equity and climate change policy. In
M. A. P. Toman (Ed.), Climate Change Economics and Policy: An RFF Anthology
Cole, S., & Brännlund, R. (2009). Climate policy measures: what do people prefer? Umeå
University, Department of Economics. Retrieved from
http://www.usbe.umu.se/digitalAssets/7/7737_ues767.pdf
Cools, M., Brijs, K., Tormans, H., Moons, E., Janssens, D., & Wets, G. (2011). The socio-
cognitive links between road pricing acceptability and changes in travel-behavior.
Transportation Research Part A: Policy and Practice, 45(8), 779788.
http://doi.org/10.1016/j.tra.2011.06.006
Czajkowski, M., & Budziński, W. (2015, January). An insight into the numerical simulation
bias a comparison of efficiency and performance of different types of quasi Monte
Carlo simulation methods under a wide range of experimental conditions. Paper
presented at Environmental Choice Modelling Conference, Copenhagen.
Dietz, S., & Atkinson, G. (2010). The Equity-Efficiency Trade-off in Environmental Policy:
Evidence from Stated Preferences. Land Economics, 86(3), 423443.
Dietz, T., Dan, A., & Shwom, R. (2007). Support for Climate Change Policy: Social
Psychological and Social Structural Influences. Rural Sociology, 72(2), 185214.
http://doi.org/10.1526/003601107781170026
European Commission (2011). Communication from the Commission to the European
Parliament, the Council, the European Economic and Social Committee and the
Committee of the Regions: A Roadmap for moving to a competitive low carbon
economy in 2050. COM(2011). Brussels.
European Commission. (2015a). Energy Union Package: A Framework Strategy for a
Resilient Energy Union with a Forward-Looking Climate Change
Policy. COM(2015). Brussels.
European Commission. (2015b). Climate Action Progress Report 2015 (Report from the
Commission to the European Parliament and the Council). Brussels. Retrieved
from http://ec.europa.eu/clima/policies/strategies/progress/docs/progress_report_2015
_en.pdf
European Council (2014). Conclusions of European Council on 2030 Climate and Energy
Policy Framework. EUCO 169/14, Brussels, 23. and 24.10.2014
European Environment Agency. (2015). Trends and projections in Europe 2015 Tracking
progress towards Europe’s climate and energy targets (No. 4/2015). Luxembourg:
21
Publications Office of the European Union. Retrieved
fromhttp://www.eea.europa.eu/publications/trends-and-projections-in-europe-2015
Eurostat. (2014). Level of internet access - households. Retrieved from
http://ec.europa.eu/eurostat/web/information-society/data/main-tables
Ferrini, S., & Scarpa, R. (2007). Designs with a priori information for nonmarket valuation
with choice experiments: A Monte Carlo study. Journal of Environmental Economics
and Management, 53(3), 342363. http://doi.org/10.1016/j.jeem.2006.10.007
Fishburn, P. C., & Sarin, R.K. (1994). Fairness and social risk I: unaggregated analyses.
Management Science, 40 (9), 1174-1188.
Harring, N., & Jagers, S. (2013). Should We Trust in Values? Explaining Public Support for
Pro-Environmental Taxes. Sustainability, 5(1), 210227.
http://doi.org/10.3390/su5010210
IPCC (2014). Summary for Policymakers. In: Climate Change 2014: Mitigation of Climate
Change. Contribution of Working Group III to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change [Edenhofer, O., R. Pichs-Madruga, Y.
Sokona, E. Farahani, S. Kadner, K. Seyboth, A. Adler, I. Baum, S. Brunner, P.
Eickemeier, B. Kriemann, J. Savolainen, S. Schlömer, C. von Stechow, T. Zwickel and
J.C. Minx (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New
York, NY, USA.
Joint Statement of Ministers of the Environment of the Visegrad Group and Bulgaria and
Romania (2014). Bratislava, September 30, 2014.
Kim, J., Schmöcker, J.-D., Fujii, S., & Noland, R. B. (2013). Attitudes towards road pricing
and environmental taxation among US and UK students. Transportation Research
Part A: Policy and Practice, 48, 5062. http://doi.org/10.1016/j.tra.2012.10.005
Layton, D. F., & Levine, R. A. (2003). How Much Does the Far Future Matter? A Hierarchical
Bayesian Analysis of the Public’s Willingness to Mitigate Ecological Impacts of
Climate Change. Journal of the American Statistical Association, 98(463), 533544.
http://doi.org/10.1198/016214503000000341
Leiserowitz, A. (2006). Climate Change Risk Perception and Policy Preferences: The Role of
Affect, Imagery, and Values. Climatic Change, 77(1-2), 4572.
http://doi.org/10.1007/s10584-006-9059-9
Li, H., Berrens, R. P., Bohara, A. K., Jenkins-Smith, H. C., Silva, C. L., & Weimer, D. L.
(2004). Would developing country commitments affect US households’ support for a
22
modified Kyoto Protocol? Ecological Economics, 48(3), 329343.
http://doi.org/10.1016/j.ecolecon.2003.10.010
Miketa, A., & Schrattenholzer, L. (2006). Equity implications of two burden-sharing rules for
stabilizing greenhouse-gas concentrations. Energy Policy, 34(7), 877891.
http://doi.org/10.1016/j.enpol.2004.08.050
O’Connor, R. E., Bard, R. J., & Fisher, A. (1999). Risk Perceptions, General Environmental
Beliefs, and Willingness to Address Climate Change. Risk Analysis, 19(3), 461471.
http://doi.org/10.1111/j.1539-6924.1999.tb00421.x
Paris Agreement. (2015, December 12). Paris: United Nations Treaty Collection.
https://treaties.un.org/Pages/ViewDetails.aspx?src=IND&mtdsg_no=XXVII-7-
d&chapter=27&clang=_en#EndDec
Poortinga, W., Spence, A., Demski, C., & Pidgeon, N. F. (2012). Individual-motivational
factors in the acceptability of demand-side and supply-side measures to reduce carbon
emissions. Energy Policy, 48, 812819. http://doi.org/10.1016/j.enpol.2012.06.029
Rais, J. (2014, February 25). Was the support of renewables in the Czech Republic failure?
Retrieved from http://4liberty.eu/was-the-support-of-renewables-in-the-czech-
republic-failure/
Revelt, D., & Train, K. (1998). Mixed Logit with Repeated Choices: Households’ Choices of
Appliance Efficiency Level. Review of Economics and Statistics, 80(4), 647657.
http://doi.org/10.1162/003465398557735
Reynolds, T. W., Bostrom, A., Read, D., & Morgan, M. G. (2010). Now What Do People
Know About Global Climate Change? Survey Studies of Educated Laypeople. Risk
Analysis: An International Journal, 30(10), 15201538.
Ringius, L., Torvanger, A., & Underdal, A. (2002). Burden Sharing and Fairness Principles in
International Climate Policy. International Environmental Agreements, 2(1), 122.
http://doi.org/10.1023/A:1015041613785
Rose, A., Stevens, B., Edmonds, J., & Wise, M. (1998). International Equity and
Differentiation in Global Warming Policy. Environmental and Resource Economics,
12(1), 2551. http://doi.org/10.1023/A:1008262407777
Sælen, H., & Kallbekken, S. (2011). A choice experiment on fuel taxation and earmarking in
Norway. Ecological Economics, 70(11), 21812190.
http://doi.org/10.1016/j.ecolecon.2011.06.024
Scarpa, R., & Rose, J. M. (2008). Design efficiency for non-market valuation with choice
modelling: how to measure it, what to report and why*. Australian Journal of
23
Agricultural and Resource Economics, 52(3), 253282. http://doi.org/10.1111/j.1467-
8489.2007.00436.x
Schade, J., & Schlag, B. (2003). Acceptability of urban transport pricing strategies.
Transportation Research Part F: Traffic Psychology and Behaviour, 6(1), 4561.
http://doi.org/10.1016/S1369-8478(02)00046-3
Schleich, J., Dütschke, E., Schwirplies, C., & Ziegler, A. (2016). Citizens’ perceptions of
justice in international climate policy: an empirical analysis. Climate Policy, 16(1),
5067. http://doi.org/10.1080/14693062.2014.979129
State energy policy of the Czech Republic (2015, approved). Ministry of Industry and Trade,
Prague December 2014.
Steg, L., Dreijerink, L., & Abrahamse, W. (2005). Factors influencing the acceptability of
energy policies: A test of VBN theory. Journal of Environmental Psychology, 25(4),
415425. http://doi.org/10.1016/j.jenvp.2005.08.003
Ščasný, M., Massetti, E., Melichar, J., & Carrara, S. (2015). Quantifying the Ancillary Benefits
of the Representative Concentration Pathways on Air Quality in Europe. Environmental
and Resource Economics, 62(2), 383415. http://doi.org/10.1007/s10640-015-9969-y
Train, K., & Weeks, M. (2005). Discrete Choice Models in Preference Space and Willingness-
to-Pay Space. In R. Scarpa & A. Alberini (Eds.), Applications of Simulation Methods
in Environmental and Resource Economics (Vol. 6, pp. 116). Berlin/Heidelberg:
Springer-Verlag. Retrieved from http://link.springer.com/10.1007/1-4020-3684-1_1
Wiser, R. H. (2003). Using Contingent Valuation to Explore Willingness to Pay for Renewable
Energy: A Comparison of Collective and Voluntary Payment Vehicles. Berkeley:
Lawrence Berkeley National Laboratory.
Wiser, R. H. (2007). Using contingent valuation to explore willingness to pay for renewable
energy: A comparison of collective and voluntary payment vehicles. Ecological
Economics, 62(34), 419432. http://doi.org/10.1016/j.ecolecon.2006.07.003
Young, H. P. (1994). Equity: In Theory and Practice. Princeton: Princeton University Press.
Zvěřinová, I., Ščasný, M., & Kyselá, E. (2014). What Influences Public Acceptance of the
Current Policies to Reduce GHG Emissions? (WP2 Deliverable 2.5.). Prague: Charles
University Environment Center. Retrieved from http://cecilia2050.eu/publications/239
24
Tables
Table 1. National samples and identification of speeders
country
mode
N (completed)
% of speeders
N valid
Czech Republic
CAWI
1,270
9.4%
1,150
CAPI
431
NA
431
Poland
CAWI
974
14.1%
837
CAPI
429
NA
429
United Kingdom
CAWI
1,420
11.9%
1,251
Note: Here and in the following tables, N denotes number of observations.
Table 2. Design of the choice experiment for acceptability of policy instruments
Note: In parentheses, we introduce abbreviations for attributes and levels that we use
throughout this paper.
Attribute
Levels
Emissions reduction target for the
European Union (increase in global
average temperature by 2100)
-20% by 2020 (+2.64.8°C by 2100)
-40% by 2030 (+1.22.8°C by 2100)
-80 by 2050 (+0.7–2.2°C by 2100)
Status Quo: 20% reduction by 2020; current policy
Distribution of costs among the
European Union countries (EU)
richer states pay more (GDP)
states with higher population pay more
(person)
higher emitting states pay more (GHG)
Status Quo: richer states pay more
Distribution of costs among the
citizens of [member state](MS)
everyone pays the same amount (person)
everyone pays the same income percentage
(income)
the rich pay a higher income percentage
(rich)
those who emit more pay more (GHG)
Status Quo: everyone pays the same amount
Increased monthly costs for your
household
20, €33, €65, €95, €130, €150
Status Quo: 0
25
Table 3. The MXL model estimates of consumers’ WTP based on the discrete choice
experiment data
Czech Republic
United Kingdom
Poland
means
standard
deviations
means
standard
deviations
means
standard
deviations
coefficient
(s.e.)
coefficient
(s.e.)
coefficient
(s.e.)
coefficient
(s.e.)
coefficient
(s.e.)
coefficient
(s.e.)
target = 20% (ref.)
0
0
0
0
0
0
target = 40%
13.19***
(2.78)
37.05***
(2.75)
45.79***
(6.26)
97.55***
(10.00)
-1.76
(2.53)
31.15***
(2.94)
target = 80%
17.04***
(3.02)
46.77***
(2.89)
43.57***
(8.36)
149.21***
(10.11)
-0.14
(2.57)
32.59***
(1.73)
burden sharing EU = GDP (ref.)
0
0
0
0
0
0
burden sharing EU = person
3.61
(2.56)
25.26***
(3.04)
-8.39
(5.43)
25.85***
(8.07)
-3.08
(2.16)
10.48***
(1.68)
burden sharing EU = GHG
14.98***
(2.56)
22.14***
(2.62)
27.13***
(6.21)
47.36***
(6.70)
-2.49
(2.33)
11.88***
(1.90)
cost distribution MS = income
(ref.)
0
0
0
0
0
0
cost distribution MS = person
-12.78***
(2.96)
16.33***
(2.88)
-15.72**
(7.49)
41.23***
(7.37)
-6.28**
(3.05)
14.75***
(1.59)
cost distribution MS = rich
3.81
(3.09)
33.07***
(3.28)
23.21***
(8.02)
61.75***
(12.15)
4.47
(3.02)
22.23***
(2.32)
cost distribution MS = GHG
20.94***
(3.41)
49.10***
(4.01)
33.17***
(9.12)
81.69***
(8.65)
12.48***
(3.50)
24.31***
(3.17)
status quo
(alternative specific constant)
-12.82***
(3.82)
81.80***
(5.44)
-39.07***
(8.82)
161.16***
(16.57)
15.95***
(3.03)
83.65***
(4.83)
-cost(100 EUR)*scale
1.39***
(0.07)
1.33***
(0.17)
0.58***
(0.08)
1.78***
(0.24)
2.02***
(0.16)
2.27***
(0.36)
Model diagnostics
Log-likelihood (constant only)
-10,064.28
-7,933.06
-6,914.07
Log-likelihood
-7,711.40
-6,014.17
-5,022.28
McFadden’s pseudo R2
0.2338
0.2419
0.2736
Ben-Akiva Lerman’s pseudo R2
0.4746
0.4846
0.5763
AIC/N
1.6373
1.6170
1.3367
N (observations)
9,486
7,506
7,596
k (parameters)
54
54
54
26
Table 4: Latent Class Model, summary of the estimates
Czech Republic
United Kingdom
Poland
against
modest
green
against
modest
green
against
modest
green
SQ*
308
-24
-128
217
-27
-323
0
-8
-135
target = 40%*
-72
9
58
-32
44
139
0
4
26
target = 80%*
0
11
67
-34
45
215
0
2
24
burden sharing EU
GDP
GHG
person
GHG
GHG
indiffer.
GHG
indiffer.
indiffer.
indiffer.
cost distribution
MS
rich
GHG
GHG
(person)
GHG
(person)
indiffer.
GHG
(income)
GHG
rich
(person)
indiffer.
indiffer.
GHG
rich
(person)
class probability
33%
28%
39%
38%
18%
44%
45%
29%
25%
Note: *Mean WTP in Euro; cost distribution rule shown in brackets indicates the least preferred
option.
27
Figures
Figure 1. Example of choice card
Policy characteristics
Policy A
Policy B
Current policy
Emissions reduction target
for the European Union
Increase in global average
temperature by 2100 if the
rest of the world complies
equivalently
80% reduction
by 2050
0.7 °C to 2.2 °C
40% reduction
by 2030
1.2 °C to 2.8 °C
20% reduction
by 2020
2.6 °C to 4.8 °C
Likely impacts
Mild
Moderate
Severe
Distribution of costs among
the European Union
countries
states with higher
population pay
more
higher emitting
states pay more
richer states pay
more
Distribution of costs among
the citizens of [the UK]
everyone pays the
same income
percentage
those who emit
more pay more
everyone pays the
same income
percentage
Increased monthly costs for
your household
£130
£130
£0
Which policy do you
consider the best taking
into account you and your
household?
Policy A
Policy B
Current policy
28
Figure 2. Choice experiments design: Emissions reduction targets for the European Union
20% reduction by 2020
40% reduction by 2030
80% reduction by 2050
Amount of greenhouse
gas emissions
emissions remain more- or less
as now, may slightly increase
(black dotted line)
-20% by 2020
-40% by 2030
then, remain stable
(orange)
-20% by 2020
-40% by 2030
-80% by 2050
(dark red line)
Policy status
policy that has been agreed at
the EU and is currently being
implemented
EU commitment, measures not
implemented yet
EU commitment, measures not
implemented yet
Increase in global
average temperature by
2100 relative to
1986-2005 level
- if the rest of the world
adopts equivalent
emission reduction
targets
2.6 ºC to 4.8 ºC
1.2 ºC to 2.8 ºC
0.7 ºC to 2.2 ºC
Likely impacts
Severe
large drop in agricultural
production
loss of most coastal areas
substantial threat to human
health caused by disease,
malnutrition, heat waves,
floods and droughts
widespread extinction of
animal and plant species, loss
of their habitats
Moderate
moderate drop in
agricultural production
loss of many coastal areas
some threat to human health
caused by disease,
malnutrition, heat waves,
floods and droughts
extinction of some animal
and plant species and loss of
their habitats (especially
coral reefs, arctic animals)
Mild
the most severe impacts of
climate change are prevented
some effects of global
warming will be felt,
however not as severe as in
the other reduction scenarios
29
Figure 3. Estimates of implicit WTP for policy packages, in EUR per month and household.
Note: Burden sharing across the EU Member States (EU) and distribution of the cost within a
member state (MS) is linked to greenhouse gas emissions released by a country or household
respectively (GHG), to GDP or income, progressively to household income (rich), or is
based on lump sum allocation (person). Estimates of means depicted as points and 95%
confidence intervals as lines.
-50
0
50
100
150
200
WTP in PPS EUR a month
UK 80% target
UK 40% target
CZ 80% target
CZ 40% target
PL 80% target
PL 40% target
EU GHG
MS GHG EU GHG
MS rich EU GDP
MS GHG EU person
MS GHG EU GDP
MS rich EU person
MS rich EU GDP
MS income EU person
MS person
... Against this backdrop, this article investigates public preferences for various EU climate change mitigation (ECCM) packages. While prior research has extensively examined Europeans' support for climate-mitigation policies at the national level (e.g., Beiser-McGrath & Busemeyer, 2023;Fairbrother et al., 2019;Hagen et al., 2016;Jagers et al., 2019;Maestre-Andrés et al., 2021;Savin et al., 2020;Wicki et al., 2019) and global levels (e.g., Bayer & Genovese, 2023;Bechtel et al., 2019Bechtel et al., , 2022Bechtel & Scheve, 2013;Beiser-McGrath & Bernauer, 2019;Gampfer et al., 2014), surprisingly little attention has been given to support for EU-level climate-mitigation policies (Coleman et al., 2023;Janik et al., 2021;Panarello & Gatto, 2023;Ščasný et al., 2017). ...
... Furthermore, previous opinion research on climate policy and European solidarity suggests that citizens are not equally sensitive to policy issues (Baute & de Ruijter, 2022;Bechtel et al., 2014;Beiser-McGrath & Busemeyer, 2023;Bergquist et al., 2020;Brannlund & Persson, 2012;Gerhards et al., 2019;Heermann et al., 2023;Tatham & Peters, 2023). Therefore, this article explores whether multidimensional preferences for the design of EU climate change mitigation policies are influenced by respondents' income position (Bergquist et al., 2020;Ščasný et al., 2017), ideology (Janik et al., 2021;Panarello & Gatto, 2023) or climate attitudes (Schaffer & Umit, 2023). These can be considered relevant categories around which a potential divide in public preferences may be structured. ...
Article
Full-text available
In the fight against climate change, the European Union has developed a new growth strategy to transform Europe into the first climate-neutral continent by 2050. To support EU member states in their transition towards greener economies, climate change mitigation policies are being implemented at the EU-level. However, such policies can be designed in different ways, and gaining citizens’ support is crucial for the political feasibility of the European green transition. Drawing on data from an original conjoint experiment conducted in Germany (N = 5,796), this article investigates how policy design shapes public support for EU climate change mitigation. To this end, the study theoretically and empirically distinguishes four policy dimensions that address the distributive politics of the European green transition: sectoral scope, social spending, financing structure and cross-country distribution. The results confirm that all four policy dimensions significantly impact public support. Specifically, the study reveals that support is greatest for EU policy packages that target financial support at the renewable energy sector, include social investment policies, are financed by increasing taxes on the rich, and distribute resources across EU member states based on population size. Furthermore, citizens’ sensitivity to the policy design varies slightly by income position, left-right ideology and climate attitudes.
... DCEs are a type of stated preference method that is commonly used to elicit customer preferences for hypothetical policies or goods. Consequently, DCEs have found widespread application in evaluating the effectiveness of planned policies across various domains, including emission mitigation [44,101,117], residential electricity demand response [109], incentives for EV purchases [74], as well as discerning consumer preferences regarding EV charging in general [24,120,122] and its integration with renewable energy sources [91,125]. Despite its extensive utilization, to the best of our knowledge our study represents the first attempt to employ this methodology specifically in assessing consumer responses to incentives tailored for shifting EV charging demand to meet supply constraints. ...
Article
Full-text available
Electric vehicle (EV) charging patterns are expected to strain electricity grids, particularly during peak residential demand in the evening. Demand-side management emerges as a potential strategy to manage this surge and align demand with supply constraints. This study investigates consumer preferences related to electric vehicle charging, focusing on the determinants influencing charging decisions during periods of electricity supply constraints or renewable energy abundance. We employ a discrete choice experiment on a representative sample of the Israeli population to assess preferences for various charging scenarios and extrapolate marginal willingness to pay values to understand the effectiveness of different incentive schemes. Key findings include a preference for evening charging, with average discounts of 20% to shift demand to nighttime, 23% to daytime, and 37% to weekends. Distinct preferences were observed among Sabbath observers, suggesting that lifestyle factors influence charging patterns. Based on our analysis, a 0.07 USD/kWh discount is needed to incentivize daytime charging, while marginal willingness to pay for green electricity reaches 0.06 USD/kWh, suggesting that communicating the renewable share in the energy mix may help reduce the discount needed. We also find that uniform tariffs may create unintended consequences, such as new peak loads in high-EV, low-renewable regions. Therefore, we recommend regionally differentiated time-of-use tariffs, further refined by season, day type, and time of day. These findings support the design of targeted incentives that reflect consumer preferences while improving grid resilience and renewable energy integration, especially in isolated electricity systems like Israel.
... We explored different model specifications, such as the Multinomial logistic model, the Random parameter logit model (RPLM), the RPLM with error component and the RPLM with correlated parameters. From this exploratory analysis, the RPLM with correlated parameter was selected, which has been widely used in the analysis of discrete choice models in environmental economics (Layton and Brown, 2000;Carlsson et al., 2003;Scarpa et al., 2008;Scasný et al., 2017;Waldman et al., 2017;Alberini et al., 2018;Bae and Rishi, 2018;Frontuto et al., 2020;Mariel and Artabe, 2020). The RPLM allows utility coefficients to vary over individuals and accounts for the panel structure of the data where each respondent made several Education levels: 1 = no studies; 2 = primary school; 3 = secondary school; 4 = bachelor or professional training; 5 = university or higher. ...
Article
Full-text available
We design an eco-scheme under the framework of the Common Agricultural Policy (CAP) 2023–2027 integrating the two most relevant agricultural practices for pollinator conservation: setting aside conservation land areas and reducing pesticides. Our main objectives are: (i) to measure farmers’ willingness to accept the implementation of pollinators-friendly agricultural practices, and (ii) to assess the effect of farmers’ environmental concerns and the source of recommendation on said willingness. A choice experiment is used to measure farmers’ preference for alternative designs of the eco-scheme in two cropping systems representative of the Aragonese and Mediterranean agriculture: rainfed extensive crops and irrigated/permanent crops. As we find, Aragonese farmers are willing to uptake agricultural practices for pollinator conservation to certain extent if they are paid to do so. This is especially true for setting aside land for conservation where more demanding practices could be accepted within current Spanish unitary payments (per ha payment). The reduction or elimination of pesticides would require payments far beyond current Spanish unitary payments. Irrigated/permanent crop farmers require larger payments than rainfed crop farmers. Farmers with pro-environmental attitudes selected more environmentally-demanding alternative levels both for sparing agricultural land and reducing pesticides. Finally, the uptake of the eco-scheme could be easier if cooperatives play an active role in its promotion. The hypothetical eco-scheme presented here could be readily implemented within CAP eco-schemes while integrating the objectives of the European Pollinators Initiative, the Biodiversity Strategy 2030 and the Farm to Fork Strategy.
... Finally, most reviewed surveys were conducted around 2010. Some recent ones also focus on citizens' attitudes towards climate policies (see Carattini et al. 2018; Maestre-Andrés, Drews, and van den Bergh 2019 for reviews of the literature, as well as the more recent studies by Carattini, Kallbekken, and Orlov 2019;Carlsson et al. 2021;Andor, Lange, and Sommer 2022;Dechezleprêtre et al. 2022;Ščasný et al. 2017;Sommer, Mattauch, and Pahle 2022) or towards other environmental issues (Börger et al. 2024). While they partially address our problem, their main focus is not on attitudes towards global climate fairness. ...
Article
Full-text available
Are countries contributing their fair share to the Paris Agreement? The answer to this central question can rely on knowledge about existing– and potentially shared– views about fairness. Yet, current studies on existing fairness views are rare and often have a limited practical scope. In this article, we design and administer a survey to elicit fairness judgements among French and US citizens regarding these issues. We find that in both countries, most respondents think that principles of climate justice should be settled internationally even if they go against some countries’ interests, and express a preference for common (rather than differentiated) responsibilities - all the more likely when they are concerned about climate change. We observe support for two theoretical effort sharing rules: a convergence towards equal per capita emissions, and an operationalized version of grandfathering. Our survey also shows inconsistencies in observed fairness judgements, and that citizens have difficulties coordinating on simple judgements regarding existing nationally determined contributions. We eventually call for the progressive development of standardised surveys on these questions.
... We estimated a model with interactions between the climate measure and inequality attributes, resulting in worsened fit (Table A6 in Appendix). Thus, we focused on main effects, consistent with previous studies (e.g., Linnerud et al., 2022;Ščasný et al., 2017). However, policy simulations from the interaction model find policies 6 and 2 have the highest acceptance (see Figure A7 in Appendix), with similar WTP estimates as in Figure 2. ...
Article
Full-text available
Effective carbon taxation is essential for achieving the green transition. However, there is typically stiff opposition to carbon taxation due to adverse equity and other impacts. Hence, it is essential to get a better understanding of which factors, including the use of tax revenue, can increase acceptability. To date, stated preference methods, rarely used to analyse this issue, have focused only on households. We conduct two national choice experiment surveys of Norwegian households and companies. The experiments include carbon taxes on transport fuel and associated emission reductions and different revenue recycling options as attributes. We find that both groups are more accepting of higher tax levels if the revenues finance climate change mitigation efforts. There is some heterogeneity among the groups with regard to using revenues to reduce different dimensions of inequality. Simulating policy scenarios, we find acceptance for the highest carbon tax among both groups when the revenues are used both to finance climate change mitigation efforts and to reduce rural-urban inequalities. This policy option points to an acceptable carbon tax close to an estimated level necessary for reaching the most ambitious climate target set by the Norwegian government. In Norway, further research should explore whether an effective carbon tax level could be achieved with modest efficiency loss, while also alleviating inequality. ARTICLE HISTORY JEL CLASSIFICATION H23; Q48; Q58; R48 Key policy insights. Policymakers need a better understanding of the factors, including the use of associated tax revenues, that increase carbon tax acceptability in order to ensure that the tax is high enough to effectively reduce emissions.. Simultaneously considering levels of carbon tax acceptance among households and companies could increase the chance of implementing higher, more effective carbon tax levels by identifying mutual interests and points of divergence.. Our results show that companies and households concur to a certain extent on the most socially acceptable policy option; earmarking carbon tax revenues to finance climate change mitigation efforts and to reduce rural-urban inequalities.
... Fischer, 2020;Haug & Wold, 2017;Huber & Murray, n.d.;Igbatayo et al., 2022;Kundo et al., 2021Kundo et al., , 2023Nishtar & Khan, 2023;Scognamillo & Sitko, 2021;Weldegebriel & Prowse, 2013) Social capital, social protection climate communication, and public acceptance Effective communication is crucial in covering social protection programs against climate shocks. When information about resource scarcity circulates in a socially cohesive society, individuals understand the available support during climate-related challenges (Bakaki & Bernauer, 2018;Bendixen et al., 2022;Bergquist et al., 2020;Coirolo et al., 2013;Giordono et al., 2023;Hannibal & Vedlitz, 2018;Hanson-Easey et al., 2018;Harris & Howe, 2023;Hart et al., 2015;Houser et al., 2022;Kruse & Atkinson, 2022;León et al., 2022;Linde, 2018;Merrill et al., 2018;Netzel et al., 2021;Pianta & Brutschin, 2022;Pillay & van den Bergh, 2016;Ščasný et al., 2017;Yazar & York, 2022) a crucial role in positively impacting climate change support. ...
Article
Full-text available
Traditional climate action strategies often overlook the critical role of social cohesion. This study explores how social protection programs and community unity can bolster climate resilience and public support for decarbonization efforts. The study adopted a systematic methodology that employed quantitative (scientometric/bibliometric analysis) and qualitative (content analysis) approaches to analyze existing social protection programs, community cohesion indicators, and public attitudes toward climate change action. Our findings revealed a crucial link: communities with robust social cohesion demonstrate a heightened commitment to climate action. This translates to increased engagement in sustainable practices, information sharing, and mutual support among community members. This suggests policymakers can leverage existing social networks, civic engagement and safety nets to build public support for decarbonization efforts. The study highlighted that building community engagement and inclusivity is critical to fostering a collective commitment to climate action and successfully transitioning to low-carbon societies.
Article
Full-text available
Green preferences are often seen as crucial for mitigating climate change. Yet, it remains unclear whether they alone can drive the shift toward a low-carbon economy and what the distributional consequences might be. This paper studies the macroeconomic, environmental, and distributional effects of green preferences among consumers and producers using the agent-based integrated assessment MATRIX model. We compare scenarios with varying pro-environmental attitudes to conventional supply-side climate policies like carbon taxes and cap-and-trade mechanisms, with and without abatement investment subsidies and alternative redistribution strategies. Without an active policy, achieving a low-carbon transition requires unrealistically high values of green preferences among consumers and producers. Conversely, carbon taxes and cap-and-trade mechanisms can reach that objective, but at the cost of increased instability and inequality. Moderate abatement subsidies can balance those effects, reducing emissions while mitigating both economic and distributional challenges, especially when environmental revenues fund social transfers instead of tax cuts.
Article
Full-text available
This paper analyzes the drivers of carbon taxes acceptability with survey data and a randomized labeling treatment. Based on a sample of more than 300 individuals, it assesses the effect on acceptability of specific policy designs and individuals’ perceptions of carbon taxes advantages and disadvantages. We find that the lack of perception of primary and ancillary benefits is one of the main barriers to the acceptability of carbon taxes. In addition, policy design matters for acceptability and in particular earmarking fiscal revenues for environmental purposes can lead to larger support. We also find an effect of labeling, comparing the wording “climate contribution” with “carbon tax”. We argue that proper policy design coupled with effective communication on the effects of carbon taxes may lead to a substantial improvement in acceptability.
Article
Full-text available
This paper presents economic benefit estimates of air quality improvements in Europe that occur as a side effect of GHG emission reductions. We consider two climate policy scenarios from two representative concentration pathways (RCPs), in which radiative forcing levels are reached in 2100. The policy tool is a global uniform tax on all GHG emissions in the integrated assessment model WITCH. The resulting consumption patterns of fossil fuels are used to estimate the physical impacts and the economic benefits of pollution reductions on human health and on key assets by implementing the most advanced version of the ExternE methodology with its impact pathway analysis. The mitigation scenario compatible with (formula presented.) (RCP 2.6) reduces total pollution costs in Europe by 84 %. Discounted cumulative ancillary benefits are equal to about €1.7 trillion between 2015 and 2100, or €17 per abated tonne of (formula presented.) in Europe. The less strict climate policy scenario (RCP 4.5) generates benefits equal to €15.5 per abated tonne of CO2\hbox {CO}_{2}CO2. Without discounting, the ancillary benefits are equal to €46 (RCP 2.6) and €51 (RCP 4.5) per tonne of (formula presented.) abated. For both scenarios, the local benefits per tonne of (formula presented.) decline over time and vary significantly across countries.
Article
Full-text available
In this paper we are concerned with what explains public acceptance and support of environmental taxes. We examine findings in environmental psychology emphasizing that people’s (environmental) value-orientation is the dominant driver determining individuals’ support for pro-environmental policy instruments. We introduce a complementary model, mainly drawing upon findings in political science, suggesting that people’s support for policy instruments is dependent on their level of political trust and their trust in other citizens. More specifically, we analyze whether political trust and inter-personal trust affect individuals’ support for an increased carbon dioxide tax in Sweden, while checking their value orientation, self-interest, and various socio-economic values. We make use of survey data obtained from a mail questionnaire sent out to a random sample of 3,000 individuals in 2009. We find that apart from people’s values, beliefs, and norms, both political trust and interpersonal trust have significant effects on people's attitudes toward an increased tax on carbon dioxide.
Technical Report
Full-text available
Climate politics cannot be successful unless a majority of various groups of people support climate policies. Public resistance and related reluctance among politicians to pursue unpopular policies are factors that can inhibit the successful implementation of climate policies. This report therefore analysis factors that affect public acceptability of current climate policies. The methods used in this report include a systematic literature review and secondary data analysis. While this provides a very good overview of possible explanatory factors, the report also examines the determinants systematically by combining different theoretical and methodological approaches. Those are economic and social-psychological frameworks and mixed-method research design. The report finds amongst other things that both the characteristics of people (social-psychological, socio-economic and demographic) and the characteristics of proposed policies are important factors influencing their public acceptability. Moreover, people generally tend to prefer policy instruments resulting in lower prices of environmentally friendly products and services (e.g. subsidies for renewable energy sources) opposite to instruments increasing the prices of those environmentally harmful (e.g. fossil fuel taxation). The report concludes that a policy instrument labelled as ‘tax’ is significantly less acceptable than an unlabelled policy instrument, even though they have the same characteristics.
Article
Full-text available
This study investigates the determinants of public acceptability of road pricing and environmental taxation policies. The strength and direction of causal paths between psychological determinants and the acceptability of these policies are measured with survey data from students in New Jersey, USA and London. The estimated models show that a number of well-established psychological determinants provide an explanation for the acceptability of both policies and in both locations despite various differences in the policy scenarios. Scenario fairness appears to be the most important direct determinant of acceptability in both countries. We further verify the effect of “specific trust in government” on scenario fairness and other direct determinants that indicate the important role of government performance for achieving acceptability for these measures. Our findings further suggest that awareness of wider environmental issues, such as climate change, can lead to the support of specific sustainable transport policies, such as road pricing, which do not address climate change issues directly.
Article
A Primer on Nonmarket Valuation is unique in its clear descriptions of the most commonly used nonmarket valuation techniques and their implementation. Individuals working for government agencies, attorneys involved with natural resource damage assessments, graduate students, and others will appreciate the non-technical and practical tone of this book. The first section of the book provides the context and theoretical foundation of nonmarket valuation, along with practical data issues. The middle two sections of the Primer describe the major stated and revealed nonmarket valuation techniques. For each technique, the steps involved in implementation are laid out and described. Both practitioners of nonmarket valuation and those who are new to the field will come away from these methods chapters with a thorough understanding of how to design, implement, and analyze a nonmarket valuation study.
Article
We present a theory of the basis of support for a social movement. Three types of support (citizenship actions, policy support and acceptance, and personal-sphere behaviors that accord with movement principles) are empirically distinct from each other and from committed activism. Drawing on theoretical work on values and norm-activation processes, we propose a value-belief-norm (VBN) theory of movement support. Individuals who accept a movement's basic values, believe that valued objects are threatened, and believe that their actions can help restore those values experience an obligation (personal norm) for pro-movement action that creates a predisposition to provide support; the particular type of support that results is dependent on the individual's capabilities and constraints. Data from a national survey of 420 respondents suggest that the VBN theory, when compared with other prevalent theories, offers the best available account of support for the environmental movement.
Article
Understanding the acceptance of and support for transportation policies focused on the environment, such fuel economy standards, is important because of the positive impact policies can have on the environment and overall sustainability goals. This study investigates the acceptance of and support for fuel economy standards through an online survey of Maine residents. Specifically, we assess the Corporate Average Fuel Economy (CAFE) standards, which aim to increase fuel economy of vehicles, while decreasing greenhouses gas emissions and foreign fuel dependence in the United States. We assess how perceptions of the policy and economic views of the market affect acceptance and support. We differentiate acceptance and support on two dimensions, a temporal and attitudinal–behavioral dimension. In doing so, we improve upon traditional measures of these variables and provide evidence that acceptance and support are distinct constructs. We find that perceived fairness, perceived effectiveness, and a subscription to a free-market ideology play a role in acceptance and support. The implications of the findings are discussed in relation to survey methods, policy communications, and an interdisciplinary understanding of environmental policy.
Article
Relying on a recent survey of more than 3400 participants from China, Germany, and the US, this article empirically analyses citizens' perceptions of key guiding principles for sharing mitigation costs across countries, justification of climate policy and trust in climate policy. Our findings suggest that the ranking of the main principles for burden-sharing is identical in China, Germany, and the US: accountability followed by capability, egalitarianism, and sovereignty. Thus, on a general level, citizens across these countries seem to have a common (normative) understanding of fairness. We therefore find no evidence that citizens' (stated) fairness preferences are detrimental to future burden-sharing agreements. In all three countries a majority of citizens considers international climate policy to be justified, but citizens' perceptions differ across specific items and countries. Finally, a substantial portion of citizens in all countries exhibit a lack of trust in international climate agreements. Policy relevance Disagreement over the distribution of mitigation costs across countries is blocking current negotiations about a new international climate change agreement to be adopted in 2015. At the heart of this disagreement are different perceptions of distributive justice among those involved in climate policy making. Our findings show that there is no difference in the ranking of fairness principles across citizens in China, Germany, and the US, suggesting that the common ground for crafting a future agreement is larger than expected. In particular, the accountability principle should weigh heavily when deciding on the burden-sharing. In addition, our findings suggest that in order to gain support among citizens, international climate policy may need to take measures to improve trust.