Distributional consequences shape public support for the EU carbon border adjustment mechanism: evidence from four European countries

The Paris Agreement's 'ratchet mechanism' has been designed to increase countries' climate ambition over time to keep global temperature rise to well below 2 °C [1–3]. For policymakers, this creates an immediate challenge of needing to balance commitments for increased national climate ambition with maintaining industrial competitiveness at home [4]. Worries over carbon leakage and firm relocation as corporate responses to tighter climate regulation [5–7] have led governments to design policies that can bridge the gap between raising climate ambition and protecting national economies.

To minimize frictions in global trade and to assuage concerns about industrial competitiveness, the international harmonization of carbon pricing has been called for [8, 9]. However, large-scale, intergovernmental coordination is difficult to achieve, and 'climate clubs'—a small set of countries committed to ambitious climate policy that coordinate climate action among each other and enforce carbon border adjustments against outside trading partners—have been proposed as a way forward [10, 11].

The European Union's (EU) carbon border adjustment mechanism (CBAM) that entered its transitional phase on 1 October 2023 and will see its full implementation of CBAM obligations from 1 January 2026 puts climate clubs into practice. The idea is simple: to protect increasingly (carbon) regulated European producers from unregulated (and hence cheaper) imports from outside the EU, a carbon tax will be levied on foreign imports as they enter the EU Single Market. Practically, the border tax will apply to carbon-intensive sectors (cement, iron and steel, aluminum, fertilizers, electricity and hydrogen) and be tied to the price a company pays for emitting a ton of carbon in the EU's internal carbon market [12]. According to the Council of the European Union [13] 'the Carbon Border Adjustment Mechanism is a victory for European climate policy. It will give us a tool to speed up the decarbonization of our industry, while protecting it from companies from countries with less ambitious climate goals.'

The enthusiasm of European policymakers for the CBAM was met with criticism by Brazil, China, India, and South Africa, the EU's largest trade partners, over the introduction of 'trade barriers' [14]. Despite these allegations, legal scholarship emphasizes that border adjustments are consistent with existing World Trade Organization (WTO) jurisprudence on policy-justified trade measures and do not, by design, violate non-discrimination provisions in trade agreements [15, 16]. Apart from legal considerations, structural differences across carbon-intensive, trade-exposed sectors [17–19] as well as international economic and geopolitical challenges resulting from the introduction of a carbon border tax [20, 21] will determine the political feasibility of the CBAM policy. As with other climate policies, the domestic political economy shapes policy support more broadly [22–25] and support for the CBAM specifically [17, 19, 26–28]. This is not least reflected in 'relatively undivided reservations' [26, 27] from domestic producers in the EU, especially those with high electricity demand and complex supply chains [17], as well as exporters that are concerned about their loss in competitiveness from the phase-out of free permits in EU carbon markets [18]. With opposition mainly coming from companies and business associations for now, policy rollbacks may seem unlikely at the moment, but this may change rapidly should the CBAM get politicized, as has recently happened with other green policies [29, 30]. As politicization hinges on broad public resentment and dissatisfaction, understanding public opinion becomes essential, and we offer initial empirical evidence for popular support of the CBAM here.

Building on existing research that shows that popular support for national carbon taxes is generally low [31–35]; that this support can be strengthened under certain conditions [36, 37]; and that costly climate action lowers citizens' appetite for expansive climate policy [25, 38–40], we argue that—similar to preference formation for national policy—public support for supranational CBAM policy derives from how citizens assess the policy's price and employment effects: under a CBAM, prices of carbon-intensive imports will go up (negative price effect), which reduces incentives for firms to relocate over differential carbon regulation in the EU and abroad; this maintains the competitiveness of European industries and secures jobs in the EU (positive employment effect). The extent to which such policy costs matter for people will shape their attitudes towards the CBAM measure.

Separate from these material concerns, the CBAM also challenges citizens' opinions about policy priorities. European advanced democracies have traditionally been advocating for both ambitious climate policy and free trade. Carbon border adjustments, by their very design, however, restrict free trade to protect carbon-intensive producers at home and incentivize trade partners abroad to adopt more stringent climate regulation. This creates an opportunity for testing expectations about whether and how framing the CBAM policy as a trade or climate measure affects the relative importance that the general public assigns to ambitious climate policy and trade liberalization. Our experimental design takes advantage of the fact that the EU's carbon tax is a new, lesser-known policy that has not received extensive mainstream media coverage yet. Unlike other green policies, such as carbon taxes or low emission zones, which have been heavily politicized [29, 30], our respondents' likely lack of extensive familiarity with the CBAM is beneficial for measuring public opinion that reflects citizens' expectations about the material effects of the policy rather than elite cues and party ideology.

2.1. Survey procedure and respondents

2.2. Experimental design, information vignettes and treatment frames

2.3. Outcome measures

2.4. Analysis

3.1. Public support for EU carbon border tax

Comparing the distribution of our $1-5$ outcome measure—ranging from 'strongly oppose' (= 1) to 'strongly support' (= 5)—among control group respondents offers a sense of the general pattern of public support for the EU CBAM measure in the absence of any treatment information (figure 1). Across all four countries, average support among sampled respondents is comparable (Germany: 3.39; Hungary: 3.32; Switzerland: 3.27; UK: 3.39).

Zoom In Zoom Out Reset image size

These descriptive findings are insightful for three reasons: First, for every respondent who strongly opposes (= 1) or opposes (= 2) the CBAM policy, there are at least twice as many respondents who indicate support (= 4) or strong support (= 5), leading to a rather favorable public perception of the CBAM policy across all four countries (Germany: 20.3% opposition to 48.3% support; Hungary: 16.9% opposition to 42.5% support; Switzerland: 21.6% opposition to 42.5% support; UK: 13.0% opposition to 42.0% support) relative to evidence from 12 high-income countries (Germany and the UK included) that (national) taxes on fossil fuels and carbon taxes with equal cash transfers are supported by around 36%–38% of respondents [43]. Hence, compared to standard estimates on the popular acceptance of national carbon taxes [33, 37, 44–46], the supranational CBAM initially receives somewhat higher support. These higher levels of support in our baseline group align closely with survey evidence from stakeholder attitudes towards CBAM in Germany in July 2021, which largely favor policy implementation [27]. Second, patterns of support are consistent across EU member states and non-member states despite differences in the potential material consequences. Third, our data also show that the group of respondents who have not made up their mind about the carbon border policy yet (adding up those in the middle category of the distribution (= 3) and those answering 'Don't know') amounts to roughly a third of respondents in Germany (31.4%) and Switzerland (35.9%), two out of five respondents in Hungary (40.5%), and almost every second respondent in the UK (45.0%). Public opinion about the CBAM is hence malleable and can possibly swing both ways depending on the policy's material effects and how it is framed to citizens.

3.2. Treatment effects: material information about CBAM policy

Figure 2 shows how CBAM support changes when we emphasize the costs of the CBAM policy (increased import prices), its benefits (protected jobs in import-competing sectors), or both. We report three key results.

Zoom In Zoom Out Reset image size

First, respondents who learn about higher costs for imported goods because of the introduction of the CBAM are much less likely to support the policy (panel (A), 'Treatment: Prices'). Policy support for a carbon border tax decreases in the face of personal costs. With reductions in support levels of 6% in Switzerland, 7% in Germany and the UK, and 10% in Hungary, substantive effects are sizable (appendix A). These findings are consistent with existing research on national climate policy [35, 40, 47] and demonstrate that waning support for costly climate action extends beyond national policies to supranational ones. The exact source of policy costs does not seem to matter to citizens as costs are perceived equally negatively, independent of whether they arise from an EU policy, like the CBAM, or are imposed by national governments through a domestic carbon tax.

Second, information vignettes about benefits from job protection produce point estimates that are small and close to zero, without any effects on CBAM support relative to the control group's average support level (panel (A), 'Treatment: Jobs'). This is the case even though our information treatment emphasizes the positive labor market effects of a carbon border tax. Panel (B) offers some evidence that these null results might be due to respondents' expectations that the CBAM policy is not going to affect their own employment prospects (appendix B). Once prompted to assess regional and country-wide effects of the CBAM, the jobs treatment produces strong positive effects in all countries except the UK.

Third, the results for our compound treatment which informs respondents about both price and employment effects in the same informational vignette (appendix C) shows that learning about positive labor market effects somewhat offsets the negative effects of policy costs from price increases alone (panel (A) 'Treatment: Both'). Across all four countries, the point estimates for the compound treatments are not as negative as those for the price treatment, and the effect in Switzerland becomes indistinguishable from the control group.

The above results hold for more fully specified regression models that include socio-economic and political control variables as well as region fixed effects, addressing concerns that differences in, for example, respondents' gender, age, or political ideology drive our results (appendix A). We also demonstrate that our estimated effects do not vary systematically across an extensive set of subgroups in our respondent population, such as when splitting them by income or education levels, employment status, climate concern, party ideology, whether they live in a rural or urban area, and for regions with a heavy manufacturing base (appendix D).

In summary, the empirical findings offer compelling evidence that citizens adjust their policy support for a supranational carbon tax in response to information about the positive and negative material effects of the CBAM policy in expected directions. Consistent with theories of loss aversion [48], the perceived downside effects are stronger than the upside effects.

3.3. Framing effects: climate protection and trade liberalization trade-off

Aside from questions of basic popular support for the EU CBAM, the broader political discussion in the context of carbon border taxes revolves around how societies prioritize climate action and trade liberalization. Following existing environmental survey research that has established 'environment versus economy' trade-off questions as a way to solicit policy priorities from the general public [50, 51], we also ask our respondents about their relative preference for ambitious climate policy compared to trade liberalization. This trade-off outcome measure does not negate the possibility of green growth [52], but reflects the fact that the CBAM, upon introduction, was greeted with grave concern, both nationally and internationally, that it hampers trade under the pretext of climate action [16, 26, 28]. While policymakers, especially from EU trade partners, may see it that way, whether the European general public buys into this narrative or not is ultimately an empirical question. Descriptively, we find little support for a strong trade-off between climate action and trade liberalization, as 50%–60% of our control group respondents across all four countries want their governments to advance both policy goals simultaneously (appendix E).

Results in figure 3 demonstrate that we do not find evidence for strong framing effects. Presenting the CBAM as either a policy that promotes climate action or impedes global trade does not affect how respondents assess the trade-off between climate protection and trade liberalization. All point estimates are estimated rather tightly around zero, and any seemingly statistically significant effects (UK: β = 0.081, $[0.001, 0.162]$, driven by the price treatment; Germany: $\beta = -0.064$, $[-0.120,-0.001]$, driven by the jobs treatment) disappear once we disaggregate the framing conditions into separate informational vignettes about the material consequences of the CBAM (appendix E). Contributing to earlier research documenting limited framing effects in climate policy [31, 53, 54], our null results likely derive from citizens' perceptions that climate protection and trade promotion are not mutually exclusive policies despite political portrayal to the contrary.

Zoom In Zoom Out Reset image size

Creating and sustaining winning coalitions for progressive climate policy is important to reach the goals of the Paris Agreement. Recent years have shown that despite high levels of general support for addressing climate change [55] scaling up climate policy is difficult because of concerns over policy costs [56, 57]. Protest and voter backlash over climate policies are no longer uncommon [37, 58, 59] and were on prominent display, for example, in the recent rejection of the CO₂ law in Switzerland [40].

We contribute to this literature by exploring the impact of distributional effects on public support for the EU's new CBAM policy through survey experiments in four European countries. Doing so at an early stage before elites and political parties had a real chance to politicize the CBAM for their own political gain is a strength of our experimental design. This is especially true because we know from existing research how politically entrenched green policies have become, not only in the US, but also across liberal democracies in Europe [29, 30, 49, 60].

Because of the experimental nature of our study, the findings benefit from high levels of internal validity, yet how these results generalize beyond the four countries we study is less clear. Very consciously, we selected Germany and Hungary as two EU member states together with Switzerland and the UK as non-EU countries to leverage variation in membership status, exposure to the material effects of the CBAM policy, political institutions, and socio-economic development. Across this varied set of countries, our results are remarkably identical, which increases our confidence in the robustness of the empirical patterns documented above. While our findings may extend to other countries at least across Europe, this may very well not be true for developing nations and emerging economies in the Global South, which have recently been argued to also hold important stakes in the CBAM policy [61].

Our findings first build on and add to existing research on the importance of public acceptance of ambitious climate policy [36, 38, 40, 56]. Compared to evidence on public support for national carbon taxes, which tend to be popular with a bit more than a third of respondents [43, 44, 46], support for the CBAM in our control group is about ten percentage points higher. However, support levels reduce by 6%–10%, depending on country, when respondents learn about the CBAM's upward pressure on import prices for targeted goods. This mirrors existing evidence from studies that evaluate ambitious, and hence costly, green policies [35, 39, 40]. While the negative effects are clear, respondents do not increase their support for a carbon border tax when treatment texts emphasize that the CBAM can protect jobs at home. Interestingly, however, this changes when we prompt survey participants to assess the effects not for themselves as individuals but for their region and the country as a whole.

Taken together, this suggests that citizens' opinions are malleable to the messages they receive about policy costs, especially in cases of new, more complex policies, such as the CBAM, for which costs will either only be incurred in the future or are hard to gauge from the outset. Consistent with this uncertainty about the effects of the CBAM on respondents' incomes and jobs, we find that only few citizens have formed strong opinions about carbon border tax policy. This can have important political implications.

First, in times of spiking living costs, championing climate policies that put additional pressure on household budgets is politically difficult. Right-wing, populist parties have begun seizing this opportunity by increasingly positioning themselves as the new climate opposition for electoral gain in multiple countries [29, 30, 49, 60]. Although the CBAM did not feature prominently in debates over climate policy during campaigns in the latest national elections in Germany (September 2021) and Hungary (April 2022), CBAM support among our survey respondents maps onto expected fault lines along mainstream and right-wing populists (figure 4 and appendix F). As a key instrument in the EU's set of ambitious net zero policies, the carbon border tax is a potential target for right-wing, anti-climate mobilization.

Zoom In Zoom Out Reset image size

Second, despite larger relative effects of our price treatments, our findings also point to an opportunity for social democratic and left-leaning parties to campaign on positive CBAM effects for struggling energy-intensive industries and blue-collar jobs that are associated with them. Mentioning that the CBAM can address the problem of carbon leakage and industrial relocation may also sway voters to support the policy. However, the extent to which sociotropic sentiments that are rooted in regional and national considerations can outweigh concerns about individual policy costs remains an open question for future research to explore.

Apart from the immediate distributional politics of the carbon border tax, our study also explicitly addresses how people's policy priorities matter for the increasingly relevant trade-off between global climate protection and free trade. Policy instruments to accelerate climate mitigation have non-negligible ripple effects in global markets and for the liberal world order [21]. Examples include the Inflation Reduction Act (IRA) in the United States, the EU's Green New Deal, or Chinese state aid policies along renewable energy supply chains. In this broader context, we inquire whether citizens' policy priorities for promoting climate protection or free trade and globalization change as a result of framing the EU or respective national governments as leaders in these areas. Our results show that framing the CBAM in a particular way is unlikely to shape policy support. We therefore cannot provide compelling guidance on which of the two global challenges the general public across our four countries would like to see prioritized by policymakers, very possibly because they favor progress on both fronts.

The transition phase of the CBAM will tell how the EU's trading partners are going to respond. For the EU, these outside pressures, albeit vocal now, will ultimately prove less important than securing policy support from the general public in EU member states and across European neighbors. Based on our findings from four European countries, this will require careful attention to the CBAM's price effects—be they real or politically construed by right-wing populists for electoral gain—, because the prospect of increased prices from a carbon border tax, as we show, can considerably sever policy support among citizens. Our research, therefore, highlights the value of complementing studies of sectoral and industrial support [17, 27] with those of popular support.

The research for this manuscript was financially supported by the project 'Beyond Policy Adoption: Implications of Energy Policy on Parties, Publics, and Individuals', funded by the Swiss National Science Foundation (PYAPP1–173642/1). We are grateful to Parrish Bergquist, Alex Gazmararian and Dennis Kolcava for comments on earlier versions of the paper together with audiences at a workshop on the Politics of Climate Change at the University of Lucerne and at the 9th Annual Conference on Environmental Politics and Governance (EPG) at the University of Glasgow. Amanda Franklin-Ryan, Melyssa Pina Sigg, and Greta Kurpicz provided excellent research assistance.

The data that support the findings of this study are openly available at the following DOI: https://doi.org/10.7910/DVN/8VHQ43.

The study was exempt from ethics approval by the Ethics Commission of Northwest and Central Switzerland (EKNZ) because it was deemed not to come under the provisions of human research law (Humanforschungsgesetz). Respondents were required to give informed consent before participating in the survey and they were aware that they were taking part in a research study. We only presented factual information to respondents without any form of deception.

The authors declare no competing interests.