Why a uniform carbon tax is unjust, no matter how the revenue is used, and should be accompanied by a limitarian carbon tax

ABSTRACT

A uniform carbon tax with equal per capita dividends is usually advocated as a cost-effective way of reducing greenhouse gas (GHG) emissions without increasing, and in many cases even reducing, economic inequality, in particular because of the positive balance between the carbon taxes paid by the worse off and the carbon dividends they receive back. In this article, I argue that a uniform carbon tax reform is unjust regardless of how the revenue is used, because it does not discourage the rich from indulging in wasteful emissions. This is at odds with a normative account of the distribution of the GHG budget that I define as emissions limitarianism, which is a climate derivative of economic limitarianism. Emissions limitarianism holds that, under low GHG budget conditions, emissions permits should not be used to achieve functionings that over-consume the GHG budget without producing significant welfare gains relative to the achievement of alternative functionings. Accordingly, I argue that a uniform carbon tax should be complemented by a limitarian carbon tax, which restricts the capabilities of the rich to achieve wasteful emission-generating functionings. The article adopts a methodological multi-level perspective on capabilities as a measure of welfare, distinguishing between basic, secondary and tertiary capabilities.

KEYWORDS:

• Capabilities
• carbon dividends
• carbon tax
• limitarianism
• luxury carbon tax

1. Introduction

The world is on track to warm by more than 3°C above pre-industrial levels by the end of this century, based on current policies (IPCC 2023, 23). There is a two-thirds chance that the global average temperature will exceed 1.5°C above pre-industrial levels, at least temporarily, before 2030 (WMO 2023). Beyond 1.5°C, the risks of triggering various climate tipping points become significant, with massive losses expected to both human and natural systems, and the risks increase as global warming continues (Dietz et al. 2021; IPCC 2023, 19). It is still possible to stay within the greenhouse gas (GHG) budget associated with the 1.5°C target, many experts say, but it will require an unprecedented transformation endeavour (IEA 2023; IPCC 2023, 21–26). Negative emissions will certainly play a role in offsetting harder-to-abate emissions, but the bulk of the effort to achieve climate neutrality can only come from an unprecedented reduction in positive emissions (IEA 2022; IPCC 2023, 21–24).

Faced with this unique social challenge, many scholars defend the carbon tax as an indispensable climate policy tool (Baranzini et al. 2017; Edenhofer et al. 2021; Heath 2021; Metcalf 2019; Mintz-Woo et al. 2021; Rabe 2018). On the one hand, the carbon tax equalises the marginal abatement costs of different economic agents by inducing them to abate their emissions up to the point where the cost of abating an additional unit of CO₂ equals the carbon price (static efficiency). On the other hand, the carbon tax provides significant incentive for economic agents to invest in low-carbon technologies that are expected to reduce emissions further in the future (dynamic efficiency).¹ In addition, many carbon tax advocates believe that if carbon revenues were distributed equally per capita, the carbon tax would reduce wealth inequality rather than exacerbate it (Baranzini et al. 2017, 7–8; Budolfson et al. 2021; Economists’ Statement on Carbon Dividends 2019; Klenert et al. 2018). This is because low-income households would receive more carbon dividends (either tax rebates or cash transfers) than the carbon taxes they pay, high-income households would pay more carbon taxes than the dividends they obtain, and middle-income households would pay carbon taxes roughly equal to the dividends they receive back (Boyce 2019; Metcalf 2019). In short, a uniform carbon tax reform with equal per capita carbon dividends, henceforth UCT + ECDs, can lead to a double ethical outcome: a cost-effective reduction in GHG emissions and an improvement in economic equality, thanks in particular to the positive balance between the carbon tax burden and the carbon dividends for the worse off (Mintz-Woo 2021).²

The article is primarily addressed to those who advocate UCT + ECDs as an essential component of a climate policy package for the climate transition. There is an open debate about whether market-based policies are consistent with the mitigation ambitions of the international community. For example, some argue that optimising the economic system through financial incentives is not enough and that systemic change is needed instead (Rosenbloom et al. 2020; Tvinnereim and Mehling 2018). I do not enter the debate about whether carbon pricing is a sufficient or even a bad idea for the climate transition, at least in some circumstances. I simply argue that if what the proponents of UCT + ECDs say is true, then the distributional effects of a carbon tax reform need to be subjected to a more elaborate ethical scrutiny than is usually the case.³

I will argue that under conditions of wealth inequality and polarisation such as we have today, both globally and in most countries, UCT + ECDs falls short of the moral demands of a just climate transition. Most members of the global wealthiest 1% (which I identify as the rich), and almost all members of the global wealthiest 0.1% (which I identify as the super-rich), have sufficient economic resources to bail themselves out of a collective mitigation effort induced by market-based mitigation policies that are equal for all (see Nielsen et al. 2021; Otto et al. 2019). This makes UCT + ECDs not only problematic for the effectiveness of the climate transition, but also unjust, as it allows the super-rich to appropriate significant shares of the greenhouse gas (GHG) budget, i.e. GHG emissions still compatible with a reasonable mitigation target, without generating significant welfare gains compared to alternative, less polluting behaviour.

The article has two main ancillary aims. The first is to develop the normative account that explains why UCT + ECDs is unjust, even when the double ethical outcome is given (a cost-effective reduction in GHG emissions, plus an increase in economic equality compared to the pre-tax scenario). To this end, I will articulate an emissions-based version of limitarianism (Robeyns 2019; 2022), that applies to the distribution of the GHG budget independently of broader considerations of socio-economic justice – I will define it as emissions limitarianism to distinguish it from economic limitarianism. The second is to argue that UCT + ECDs can pass the justice test if it is accompanied by a special form of luxury carbon tax (Benoit 2020; Oswald et al. 2023), which I define as a limitarian carbon tax.

Before proceeding, I would like to clarify two methodological points. The first point is that my discourse is to be understood entirely in terms of ideal theory. On a non-ideal level, in fact, any unilateral action by a single country (or group of countries) against luxury emissions is indeed prone to luxury carbon leakage. The latter is an even more serious problem than classical carbon leakage because the rich can move their assets and financial interests across national borders with great ease. While possible solutions to traditional carbon leakage can, at least in theory, be extended to luxury emissions (think, for example, of carbon border adjustment mechanisms, climate clubs, etc., see Falkner, Nasiritousi, and Reischl 2021), I am not in a position to judge, at least here, if and which solution is more effective. It may therefore be the case that the ethical reasons for introducing a limitarian carbon tax alongside the uniform carbon tax are outweighed by pragmatic reasons against it. What I want to show in this article, however, is that the ethical reasons for integrating a uniform carbon tax reform with a limitarian carbon tax do exist and need to be taken into account in the policy-making process.

The second methodological point is that, in presenting emissions limitarianism, I will treat capabilities as a ‘multi-level concept’ (Day, Walker, and Simcock 2016; Smith and Seward 2009, 228), and thus distinguish between basic, secondary and tertiary capabilities. Basic capabilities are ‘basic reasons for actions’ (Alkire 2002, 185), i.e. vectors of human being that cannot be reduced to a lower denominator and are therefore incommensurable and non-hierarchical (Grisez, Boyle, and Finnis 1987) – a famous example being the ten basic capabilities identified by Martha Nussbaum (2011, 17–45) as ‘central’. Secondary capabilities are ‘precursors’ of basic capabilities (Day, Walker, and Simcock 2016, 259), i.e. more specific capabilities that usually need to be combined in order for one or more basic capabilities to materialise. Moreover, as many secondary capabilities are more or less vague, they can materialise through two or more tertiary capabilities.

For example, consider the capability to get from point A to point B in a reasonable time – we can call this the A-to-B capability. A given moving time is reasonable if it is compatible with the purpose of the move – so if the purpose of getting from A to B is to spend the weekend in B, and the move takes two days to get to B and two days to get back to A, the duration of the move is unreasonable in relation to its purpose. The A-to-B capability is a secondary capability. On the one hand, it is instrumental to one or more lower-level capabilities, i.e. basic capabilities, depending on the case and the person. The assumption is that the lower the level of a capability, the more directly it contributes to individual welfare. Accordingly, the A-to-B capability may be a precursor of the capability to acquire knowledge, the capability to form and maintain love relationships, the capability to enjoy oneself (which philosophers such as John Finnis 2011, 87; and Martha Nussbaum 2011, 34, would more elegantly label ‘play’), and so on. On the other hand, the A-to-B capability can materialise in different ways, indicating tertiary capabilities: being able to take the train from A to B, being able to take a private jet from A to B, being able to drive from A to B, and so on. In short, any individual who has access to at least one of the above tertiary capabilities can be said to enjoy the secondary A-to-B capability, which in turn is a precursor to one or more basic capabilities.⁴

2. Why wasteful emissions are unjust: emissions limitarianism

Whenever a secondary or tertiary capability X to achieve an emission-generating functioning (e.g. the capability to fly privately) is substitutable for a same-level alternative capability Y to achieve a much less emission-generating functioning (e.g. the capability to fly with airlines or to travel by train), without unreasonably interfering with the enjoyment of a lower-level capability (e.g. love, play, knowledge), there can be no moral claim to capability X under severe GHG budget constraints. This is because if a given individual enjoys capability Y, where enjoyment implies not only technical but also economic viability, then achieving the functioning of capability X is wasteful in the sense that the individual would consume a much larger share of the GHG budget than achieving the functioning of capability Y, without this leading to significant welfare gains – i.e. without this being instrumental in the exercise of a broader set of lower-level capabilities. I label this normative account of the GHG budget distribution as emissions limitarianism.

The term limitarianism has recently been introduced in the literature by Ingrid Robeyns (2019; 2022), in connection with a theory of socio-economic justice known as economic limitarianism, and in a few years it has sparked a new philosophical debate. Economic limitarianism argues that it is both possible and necessary to set a limit to wealth that no one should be allowed to exceed. The economic limitarian claim is based on both an axiological premise and one or more triggering elements.

The axiological premise, in Robeyns’ formulation, is that welfare is a ‘satiable concept’; thus, it is possible to identify a ‘riches line’ above which marginal positive changes in wealth yield nihil marginal contributions to individual ‘flourishing’. Therefore, depriving someone of ‘surplus wealth’, i.e. wealth above the riches line, entails very little opportunity cost in terms of welfare. Under hypothetical conditions of widespread abundance and substantive political equality, the riches line would be insufficient to justify the limitarian claim. Accordingly, Robeyns (2019) identifies three main triggering elements, later discussed in the literature (Andersson 2021, paper no. 4, 7–14; Huseby 2022; Volacu and Dumitru 2019), that can make the limitarian claim hold true. First, the social and political spheres are arranged (or rather disarranged) in such a way that prominence in the economic sphere can lead to excessive political influence and power (see also Zwarthoed 2018). Second, a large part of the world's population is unable to meet basic needs due to a lack of economic resources. Third, climate adaptation and mitigation require major public investments that many governments are unable to mobilise (for Robeyns this trigger is one of the possible specifications of the second trigger). If at least one of these triggering elements is present, limitarians claim that it is morally legitimate to take surplus wealth from the rich, given the existence of the riches line.

In a more recent publication, Robeyns (2022, 254–255) clarifies that although the riches line is part of the justification for the limitarian threshold, the two concepts need not be equivalent; she therefore distinguishes between ‘excess wealth’, which is above the limitarian threshold, and surplus wealth, which is above the riches line, making it clear that the two types of wealth do not always coincide. Indeed, in some cases, it is possible that the limitarian threshold may be set lower than the riches line for reasons that are not axiological, but rather related to one of the triggering elements mentioned above; for example, in some cases it may be necessary to take away more than surplus wealth from the very well-off, in order to ensure the good functioning of democratic institutions. Moreover, the riches line can also be interpreted in a more moderate way, without undermining the limitarian claim; for example, it can be assumed that the riches line does not indicate the point beyond which marginal increases in wealth cease to produce marginal positive changes in welfare, but rather the point beyond which marginal increases in wealth produce positive changes in welfare that should be considered relatively less important, morally speaking – thus leaving open the possibility of trade-offs across the limitarian threshold, and especially between the top and the bottom (see Timmer 2021, 761–762).

I will propose an emissions limitarian account that differs from economic limitarianism in several ways. First, the axiological premise I defend is not generally about money, but about GHG emissions. Achieving some emission-generating functionings is wasteful in that it depletes sustainable emissions (i.e. emissions compatible with the GHG budget associated with a reasonable mitigation target) without making a significant contribution to individual welfare. This is because individuals could achieve much less polluting functionings instead, without compromising their exercise of lower-level capabilities. Second, like economic limitarianism, the emissions limitarian claim I defend cannot stand on axiological premises alone. In a world with an unlimited GHG budget and/or negative emissions possibilities, it would not necessarily be wrong to indulge in frivolous actions that generate a great deal of GHG emissions (at least in relation to the climate problem, I am not addressing the broader issue of pollution here), just as in general all superfluous and extravagant goods and services that people have access to are not prima facie morally wrong for economic limitarians. Hence the need for a triggering element, which in this case is purely climatic rather than socio-economic or related to democratic processes, and which consists of a tight budget of sustainable GHG emissions (IEA 2023; IPCC 2023). While it is true that negative emissions technologies could theoretically allow GHG emissions to be produced without depleting the GHG budget, these are solutions that have so far been deployed on a very small scale, and it is not possible to predict if and when the technical and economic barriers to much wider deployment will be overcome. Given these uncertainties, it would be unjustified to pretend that we have more sustainable emissions than we actually do (Shue 2017).

Third, the distribuendum of the emissions limitarian claim is not wealth, but GHG emissions, or more specifically emissions permits. This is because, while it is generally true that increases in wealth tend to lead to increases in GHG emissions, this is not necessarily always the case – e.g. a person who is slightly above the limitarian wealth threshold might emit less than a person who is slightly below the threshold. Thus, emissions limitarianism does not claim that surplus wealth as such is morally problematic under certain conditions, but rather that we should prevent the rich from converting their wealth into wasteful emissions-generating functionings. Emissions limitarianism can be seen as a necessary but not sufficient condition for a just distribution of GHG emissions permits. Emissions limitarianism is thus compatible with any account of the just distribution of non-wasteful emissions permits, insofar as it does not defend a comprehensive account of the just distribution of emissions permits, but simply says that such a distribution cannot allow for the wasteful use of emissions permits.⁵

The central question for someone wanting to embrace emissions limitarianism, then, is when the achievement of an emission-generating functioning should be considered wasteful according to the definition given above.⁶ To clarify and operationalise the notion of wasteful emissions, I find it useful to distinguish between basic, secondary, and tertiary capabilities, as discussed in the introduction (see also Day, Walker, and Simcock 2016; Smith and Seward 2009). One way of framing this distinction is to say that a basic capability is a reason for action that requires no further justification, because it is directly constitutive of individual welfare (see also Finnis 2011, 85; Grisez, Boyle, and Finnis 1987, 103). I do not intend to provide an exhaustive list of basic capabilities here, as this is not my purpose and there is no space for it. Examples of basic capabilities include being able to feed oneself, being able to maintain good health, being able to enjoy social respect (Day, Walker, and Simcock 2016), knowledge, aesthetic experience, sociability, play, practical reason (Finnis 2011), emotions (Nussbaum 2011), and so on. Again, I do not mean that this list is necessarily exhaustive, nor do I mean that all these things are necessary for a flourishing life, but simply that these capabilities are the '“«primary colors»” of values’ (Alkire 2002, 186), i.e. the basic colors (e.g. red, blue, yellow) that cannot be deconstructed into more basic colors and from which all other colors are obtained by mixing.

Typically, each basic capability can materialise through two or more alternative secondary capabilities. A secondary capability indicates a way in which a given basic capability can materialise, without qualitative specifications. Some basic capabilities can only materialise through a single secondary capability. For example, the basic capability of being fed could only materialise through the secondary capability of having access to food (broadly defined). Some other basic capabilities, however, can materialise through two or more secondary capabilities – in fact, in many cases a mix of different secondary capabilities is required. For example, the basic capability to play, which consists of having a reasonable opportunity to access leisure activities for their own sake, could be said to materialise as long as the individual enjoys a number of secondary capabilities: i.e. having access to the national artistic heritage, having access to public information, being able to travel, having access to the Internet, and so on. Moreover, almost any secondary capability can materialise through a wide range of tertiary capabilities, which indicate different means of implementation, distinguished on a qualitative basis – where quality is usually associated with different price ranges. For example, the secondary capability of travelling can materialise as long as the individual has access to one of these alternative tertiary capabilities: being able to travel by car, being able to fly economy class, being able to fly first class, being able to travel by train, and so on. The same is true for those secondary capabilities that are the only way a given basic capability can materialise. For example, we said that having access to food is the only reasonable way of being fed, but the same secondary capability of having access to food can materialise in both the capability of eating in a gourmet restaurant and in the capability of eating in a fast food restaurant.

People typically choose to realise a different mix of either secondary or tertiary capabilities, i.e. to achieve the relative functionings, based on a range of personal and/or collective considerations. An individual who has to decide how to commute to work – where commuting is a secondary capability that is usually instrumental to the basic capability of practical reason (see Finnis 2011, 88) – will consider all the options available to them that correspond to tertiary capabilities: driving, cycling, taking public transport, and so on. They will choose which tertiary capability to realise by weighing several factors, such as price, comfort, time taken, environmental impact, and so on. But climate change poses a major problem. We cannot all simply choose to achieve the functionings we like best, because this conflicts with the climate transition – i.e. if all the people who now eat red meat several days per week do not adjust their diet, or if all the people who fly several times per month do not change their behaviour, we could never achieve net-zero CO₂ emissions by mid-century (see IEA 2022).

According to emissions limitarianism, a just climate transition requires that the few remaining sustainable emissions are not wasted. The way to ensure this is to limit all those higher-level capabilities (either tertiary or even secondary), whose functionings are comparatively very climate-damaging, that can be foregone without compromising anyone’s enjoyment of lower-level (either secondary or basic) capabilities. Obviously, even the removal of a tertiary capability (and more generally of any higher-level capability) may cause some subjective discomfort, but the axiological assumption of emissions limitarianism is that this discomfort will not translate into significant welfare losses if there is an alternative same-level capability available that can guarantee the enjoyment of the lower-level capability to which the replaced capability was instrumental.

Consider, for example, the tertiary capability of private aviation. It is much more polluting than commercial flights – between 4 and 14 times more –, and a train – 50 times more (see Transport and Environment 2021, 3). The main secondary capability to which the private-flight capability is instrumental is getting from point A to point B in a reasonable time and on reasonable terms – we previously called it the A-to-B capability. The latter tends to be instrumental to one (or a mixture) of the following basic capabilities: practical reason, play, emotions, affiliation, and so on. The question we should be asking is whether the individual has a valid alternative to the private-flight capability within their tertiary capability set, the realisation of which is less environmentally damaging and which also allows them to enjoy the A-to-B capability. If they do have such an alternative, then achieving the private-flight functioning causes wasteful emissions and is therefore unjust on emissions limitarian grounds.

Suppose, for example, that point A is Amsterdam and point B is Paris. There are many commercial flights departing from A to B at different times of the day, not to mention high-speed trains. Both commercial flights and trains are obviously cheaper than a private flight, so the price issue does not apply. The private flight may be more pleasant than other travel alternatives (I suppose), and it gives the traveller a distinctive social status. However, neither the comfort of the private flight nor the status associated with its use are functional to the enjoyment of the A-to-B capability. Furthermore, even if the private-flight capability were primarily a precursor to the capability of social status affirmation rather than to the A-to-B capability, it can be assumed that a very wealthy person could affirm their social status with alternative functionings that are less polluting than a private flight. Therefore, under the circumstances just described, private flying is perfectly substitutable for air and/or rail travel.

A possible objection to emissions limitarianism is that there might be some cases where the emissions of a seemingly extravagant functioning, such as private flying, are not wasteful, so that achieving this specific functioning should not be considered as morally wrong. Imagine, for example, that point A is Amsterdam and point C is a place on the edge of the Amazon rainforest that is not served by airline routes and that would take several days to reach. In this case, for the purposes of the A-to-C capability, the private-flight capability is not fully substitutable. In order to understand whether it is morally justified to deprive someone of the private-flight capability when it is not fully substitutable for the purposes of the A-to-C capability, we need to ask whether and to what extent travel from A to C is necessary for one or more basic capabilities to materialise. For example, suppose the reason John needs to travel quickly from A to C is that he has been bitten by a very rare poisonous snake and the only way to save his life is to be inoculated with an equally rare antidote, which of course is only available at location C. In this case, the enjoyment of the private-flight capability is the only way for John to enjoy the A-to-C capability, on which the enjoyment of any basic capability (i.e. survival) depends. Accordingly, restricting the private-flight capability is likely to be unfair to John, regardless of the GHG budget constraints. It is very likely, however, that the reasons why people use private jets, in the vast majority of cases, have to do with matters of business, love, tourism, and so on; in other words, all the human needs that can be reasonably satisfied even without the capability to fly privately. In short, those who embrace emissions limitarianism must necessarily bite the bullet of this objection, but they can also rest assured that the injustices that can arise from the implementation of emissions limitarianism are quite rare.

The same discourse applies to many other tertiary capabilities available to the super-rich, among which both space tourism and yacht holidays are worth mentioning. Space tourism can be either suborbital, lasting just a few minutes, just enough time to look at the Earth from above, or orbital, lasting even a few days, like a real holiday. The first suborbital sightseeing trips took place in the summer of 2021 aboard two spacecrafts, one by Virgin Galactic and the other by Blue Origin, carrying six and four passengers respectively – prices for suborbital trips are in the hundreds of thousands of USD. The first orbital space trip, on the other hand, was organised by SpaceX in September 2021, with four people on board, and lasted about three days – prices for orbital trips run into the millions of USD. Both types of space tourism require the launch of a rocket, which involves the combustion of fuel, which in turn produces GHG emissions – and it should be emphasised that GHG emissions are even more damaging to the climate when they are released directly into the stratosphere, as is the case with rocket launches. Several scientists are studying the climate impact of rocket launches, and of course each case is different, so we cannot draw any firm conclusions. However, based on existing data, it is safe to assume that a rocket launch, lasting just a few seconds, causes more than 100 tCO_2e (Kokkinakis and Drikakis 2022; Marais 2021), which is more than the average member of the global top 1% emits in a year (Chancel 2022).

I see four main reasons why people spend a lot of money to be launched into the sky (see also Heilweil 2021): the unique experience of seeing the Earth from above, being in the history books, confirming a certain social status, feeling a strong adrenaline rush. All of these are secondary capabilities that lead back to the basic capabilities of play and practical reason. Some of them can be easily replaced by less polluting alternatives (e.g. although I am no expert, I would guess that some extreme sports can give similar or equivalent sensations to the adrenaline rush of space travel); some others, unfortunately, cannot (if you want to see the Earth from above, live and with your own eyes, there is no alternative to going up into space). A very wealthy person, however, has access to a range of alternative secondary capabilities that can satisfy the basic capabilities of play and practical reason – if this were not the case, it would be a case of cultivating of expensive tastes, in terms of GHG emissions, that cannot justify the use of the GHG budget share that they require (see Kaplow 2006).

Similarly, on average, a superyacht (i.e. longer than 24 metres in length) emits the equivalent of 202 cars per year from an average US household (Lynch et al. 2019, 381–382). The reason people pay for a superyacht holiday is to spend their leisure time in a way that meets certain recreational standards. Needless to say, there are many less polluting functionings that a would-be superyacht owner could substitute for without sacrificing the level of enjoyment and satisfaction they seek – the first that comes to mind is a holiday on a sailboat built with sustainable materials.

3. Why a uniform carbon tax has trouble dealing with the wasteful emissions of the rich

UCT + ECDs provide the financial incentive for most people to give up wasteful emission-generating functionings, but not for high-income households. We have already seen that UCT + ECDs can give both low- and middle-income households more carbon dividends than they pay in carbon taxes, and thus also the opportunity to keep their carbon consumption unchanged. However, people whose welfare is not yet fully satiated by their wealth (Robeyns 2022, 254) are particularly interested in maximising the positive income change induced by UCT + ECDs by reducing their carbon consumption in order to increase the difference between carbon dividends and carbon taxes (see also Boyce 2018, 58–59). For example, if UCT + ECDs make car use much more expensive than public transport use, most people will have a significant incentive to replace car use with public transport for commuting to work, even if the carbon dividends were sufficient to cover the increased cost of car use. There are two main reasons for this. First, if individuals continue to enjoy the lower-level capability to get to work in a reasonable time and on reasonable terms even after foregoing the car use functioning (thanks to the tertiary capability to use public transport), they will be induced to forego the car use functioning in order to increase their income (despite the fact that the car use capability is shielded by carbon dividends). Second, carbon dividends may be paid long after the individual has made the consumption choice. For example, if carbon dividends were paid once a year, the car-use capability would only be guaranteed in a broad time perspective for both low- and middle-income individuals – i.e. in relation to the final balance of the year, but not in relation to individual acts of consumption.

In short, UCT + ECDs provide effective financial incentives for most people to reduce wasteful emissions, even the poor who receive more in carbon dividends than they pay in carbon taxes. Wasteful emissions are the lowest-hanging fruit because their opportunity cost in terms of welfare (using capabilities to realise functionings as a metric) is relatively low, so they are the ones that most people can be expected to reduce first and most in response to UCT + ECDs.

Conversely, high-income households, and especially the super-rich, have sufficient resources to remain indifferent to the financial incentive provided by UCT + ECDs (Castano Garcia et al. 2021; Chancel 2022, 935). Some data on carbon inequality recently published by Lucas Chancel (2022) may help to clarify this (see also Barros and Wilk 2021; Kenner 2019). A very small minority of the world’s population, the wealthiest global 1%, is responsible for almost 17% of global CO_2e emissions (with average per capita emissions of 101 tCO_2e); compared to less than 12% of CO_2e emissions by the poorest half of the global population (with average per capita emissions of 1.4 tCO_2e) – the global top 1% includes all people whose wealth is just over USD 800,000 up to the richest people in the world (Chancel et al. 2022, 28); according to Credit Suisse (2023, 21), instead, the number of global millionaires exceeds the global top 1%. An even smaller minority, the global top 0.1%, are responsible for more than 7% of global CO_2e emissions (with an annual per capita average of 425 tCO_2e) – all those with at least USD 3.5 million fall into this rather exclusive group (Chancel et al. 2022, 28). This means that a uniform carbon tax set at USD 75 (IMF Communications Department 2022) would impose an average carbon tax burden of just over USD 7,500 per year on the global top 1%, a group of people in which the poorest have just under USD 1 million and the richest a few hundred billion USD. The average carbon tax burden on the global top 0.1%, the poorest of whom have more than USD 3 million, would be just under USD 32,000 per year. These are rough calculations, of course, but they serve to give an idea of how the burden of a uniform carbon tax could be absorbed by the members of the global top 1%, and even more so by members of the global top 0.1%, without leading to significant losses in capabilities.

For all these reasons, UCT + ECD_S allow the rich, and especially the super-rich, to retain the capability to achieve wasteful emission-generating functionings, while creating barriers and/or effective disincentives for everyone else to indulge in wasteful emissions.

4. Why emissions limitarianism?

I have proposed emissions limitarianism as a justification for why UCT + ECDs is unjust, even if the double ethical outcome (less GHG emissions, more economic equality) is given. It is now important to clarify why I argue that we (also) need this specific normative account to assess the distributional effects of a carbon pricing reform, instead of relying on established normative theories, such as economic limitarianism or the classical distinction between subsistence and luxury emissions, first proposed by Henry Shue (1993) and later discussed in the literature.

I suggest that there are at least two reasons why it is important to pursue emissions limitarianism independently of (and perhaps before) economic limitarianism: emissions limitarianism can address some of the main objections raised against economic limitarianism, and achieving the goals of economic limitarianism does not guarantee achieving the goals of emissions limitarianism.

First, unlike economic limitarianism, emissions limitarianism does not enter into the normative debate about whether surplus wealth is legitimate or not; it simply asserts that the rich should not convert their wealth, whether surplus or not, into wasteful emission-generating functionings. This has both normative and pragmatic advantages. On the one hand, emissions limitarianism can fully or even partially circumvent some of the main objections that can be raised against economic limitarianism. The main concern is that, even if the axiological claim of economic limitarianism were correct, it would be socially counterproductive for society and/or morally illegitimate to overtax the rich. Economic limitarianism would be counterproductive in that it would deprive the most innovative and fertile people of the incentives to go further and further in their entrepreneurial activities, which generate marginal increases in wealth that also benefit the non-rich, albeit indirectly – for example, through job opportunities, new technologies, and, more generally, new investments (see Nielsen and Axelsen 2022, 744–748; Robeyns 2019, 260–262). Moreover, economic limitarianism might be morally questionable in those cases where the rich claim to have ‘earned’ their wealth, insofar as they have played by the rules of the capitalist game that largely govern contemporary societies, without exploiting and without evading taxes (see Halldenius 2022, 786–789). On the other hand, if we simply consider the interest the super-rich have in remaining super-rich, and the economic and political influence they wield, it is safe to assume that emissions limitarianism is more politically feasible than economic limitarianism – in other words, it is easier to get a super-rich person to accept giving up the possibility of space travel than a significant (in monetary terms) portion of their wealth. The relatively lower ambition of emissions limitarianism is, in my view, a first argument for pursuing emissions limitarianism independently of the broader societal transformation that underlies emissions limitarianism, given the urgency of the climate crisis.

Admittedly, one could get the impression that there is a form of perfectionism hidden in emissions limitarianism, in that it argues that it is all right for rich people to spend their money in some ways but not in others. And this, in turn, could be seen as a disadvantage of emissions limitarianism compared to economic limitarianism, which does not judge individuals’ consumption choices. It should be stressed, however, that the reasons why emissions limitarianism condemns certain consumption choices are simply the distributional consequences of those choices in relation to sustainable emissions, and have nothing to do with any particular view of the good life.⁷ For example, if it were magically possible to remove much of the CO₂ that has accumulated in the atmosphere at relatively low cost, then the issue of wastefulness would no longer arise with respect to such functionings as space travel or yacht holidays, and these would therefore no longer be unjust on the basis of emissions limitarianism.

Second, there is no absolute guarantee that the normative demands of emissions limitarianism will be fully satisfied by economic limitarianism. Indeed, even if all individuals were kept below the limitarian threshold, it is conceivable that at least some people would continue to indulge in wasteful emission-generating functionings; for example, driving a large SUV instead of a less polluting car that offers the same standards of comfort and performance and/or reliability. It could be argued, however, that economic limitarianism combined with UCT + ECDs could do the same job as emissions limitarianism if, for example, the limitarian threshold (demarcating excess wealth) were set below the wealth threshold that demarcates access to the global top 0.1% – or, in other words, if economic limitarianism flattened the wealth that makes people behaviourally non-plastic with respect to the uniform carbon tax. However, whether and when economic limitarian policies are sufficient to make all social groups behaviourally plastic towards UCT + ECDs is an empirical question, depending inter alia on where the riches line is set, which cannot be definitively resolved, and which may produce different answers in different social contexts. This is a second argument why a limitarian carbon tax should complement UCT + ECDs, while the limitarian wealth threshold should do the broader distributive work for which it is intended.

A second alleged reason why emissions limitarianism might be redundant is that the normative account of subsistence emissions vs. luxury emissions (hereafter SEvsLE) would be sufficient to justify a luxury carbon tax as a complement to UCT + ECDs. SEvsLE argues that subsistence emissions should be prioritised over luxury emissions. This is because subsistence emissions aim to satisfy basic needs, whereas luxury emissions aim to satisfy less urgent needs (Shue 1993). Applying this theory to a tight GHG budget scenario, such as the one we are experiencing today, one might conclude that all luxury emissions are wrong, and therefore perhaps to be banned, because they take carbon space away from subsistence emissions (see Tank 2022). However, SEvsLE does not fully overlap with emissions limitarianism, as the two normative accounts put forward different moral arguments against two different types of emission-generating functionings, namely luxury and wasteful ones. To be more precise, according to SEvsLE, in the presence of GHG constraints, it is justified to limit the capabilities to realise emission-generating functionings that do not serve the satisfaction of subsistence needs, because the latter are morally urgent, and we cannot run the risk that there will not be enough of the GHG budget left to cover basic necessities. Instead, according to emissions limitarianism, it is morally wrong to waste a scarce GHG budget, regardless of whether the saved emissions permits are used to satisfy more or less subsistence needs.

This has significant practical implications. First, consider a hypothetical, but not unrealistic scenario in which people can meet most of their subsistence needs without eroding the GHG budget; this could be the result of significant investment in the decarbonising core sectors such as transport, food and energy production (see Shue 2019). According to SEvsLE, in this low-carbon subsistence scenario, a UCT + ECDs that fails to curb luxury emissions would be less morally problematic than in the current scenario, because subsistence needs could be met by consuming much less, if any, of the GHG budget; and thus the emissions permits that luxury emissions would ‘steal’, so to speak, from subsistence emissions would be far fewer (see also McLaughlin 2019). Conversely, in the same low-carbon subsistence scenario, wasteful emissions would remain morally problematic for emissions limitarianism, as long as they ‘waste’ parts of the GHG budget that could be used to produce welfare gains for others, though not necessarily subsistence-related welfare.⁸

Second, emissions limitarianism allows a distinction to be made, within the emissions of the rich, between those that are more or less morally wrong on the basis of their wastefulness, whereas for SEvsLE all non-subsistence emissions are prima facie morally problematic; and this is obviously a drawback when decisions have to be made about which emission-generating functionings to cut first. More precisely, the identification of luxury emissions is derivative of the identification of subsistence emissions, and the latter in turn depends on both the specification of what counts as subsistence needs and which subsistence needs require GHG emissions to be met. This means that all emissions that are not necessary for subsistence are considered luxury emissions. Emissions limitarianism, on the other hand, starts with the emissions associated with tertiary capabilities and asks which of these emissions can be reduced without undermining anyone's secondary capabilities, and then does the same with secondary capabilities in relation to basic capabilities. This means that not all emissions associated with the exercise of capabilities other than basic capabilities are morally problematic for emissions limitarianism, but only those associated with capabilities that are both highly polluting and fully substitutable by less polluting alternatives, as discussed in section 2.

5. Implementing emissions limitarianism

So far, I have argued that UCT + ECDs do not meet the demands of emissions limitarianism and therefore fall short of achieving a just climate transition. However, this does not mean that UCT + ECDs should be discarded, or at least not necessarily; a uniform carbon tax reform could be complemented by a public policy that induces the super-rich to give up wasteful emissions (see also Gore 2021, 8–9; Otto et al. 2019). One public policy that could move in this direction was recently proposed by Philippe Benoit (2020), who suggested overtaxing the purchase of luxury and highly polluting goods and services, instead of (or even in addition to) the emissions they produce (see also Shue 2021, 139–141; Wallace and Welton 2023).

How high the rate of a luxury carbon tax should be is a political economy question that is obviously open to debate. For example, Benoit (2020, 4) suggests that a luxury carbon tax of USD 100,000 should apply to a space trip that can be purchased for USD 250,000. This is beyond the scope of this article. Instead, it is important to emphasise that the rate of the luxury carbon tax is in addition to, and much higher than, the social cost of carbon associated with luxury emission-generating functionings – that is, the discounted monetary value of the climate damage caused by marginal GHGs. This is because, unlike the uniform carbon tax, the purpose of the luxury carbon tax is not to confront people with the monetary value of their climate externalities (Heath 2021, 149–171), but rather to confront the rich with a carbon tax so high that they cannot remain indifferent.

Suppose, for example, that the luxury carbon tax is introduced in parallel with the uniform carbon tax, and that the latter is more or less equal to the social cost of carbon. What are the implications for a yacht owner? On the one hand, a yacht owner will have to internalise the social cost of both the production emissions (i.e. the emissions generated during the construction of the yacht) and the use related emissions (i.e. mainly the emissions from fuel combustion); and this will be done through the uniform carbon tax. On the other hand, the yacht owner will have to pay an additional luxury carbon tax for the fact that the GHG emissions associated with the yacht's life cycle are ‘discretionary extravagant’ (Benoit 2020, 2). There are several ways in which a luxury carbon tax could be implemented, such as at the time of purchase, or as a form of annual property tax.

A different approach to luxury taxation has been proposed in a very recent study by Oswald and colleagues (2023). According to them, in order to accelerate the climate transition and make it more equitable, carbon tax rates should be differentiated according to the purpose of the goods and services to which they are applied, with higher rates for luxury goods and lower rates for necessities – luxury goods and services being, for the purposes of the study, those whose income elasticity of demand is positive and more than proportional, while necessities have a positive but less than proportional income elasticity of demand. Oswald and colleagues conclude that carbon tax rates differentiated in this way would have significant positive effects both in terms of mitigation and in terms of wealth redistribution – with carbon revenues being used for retrofitting households and for payments to poorer households.

I will not get into the question of which version of a luxury carbon tax is better. I will simply point out that for a carbon tax targeting the consumption of the super-rich to be justified on the basis of emissions limitarianism, the tax must target not luxury emission-generating functionings as such, but wasteful emission-generating functionings. Whether and when a luxury emission-generating functioning is also wasteful is a question to which the account of emissions limitarianism as I have presented it in section 2 has the ambition to provide an answer. For these reasons, I maintain that UCT + ECDs should be complemented by a limitarian carbon tax (LMCT), which is not always equivalent to a luxury carbon tax. More specifically, the two taxes share the same logic, but not necessarily the same target. In the remainder of this article, I would like to address two issues related to the UCT + ECDs + LMCT proposal. First, if wasteful emissions are morally wrong, why not ban them? Second, assuming that LMCT is, all things considered, the preferred policy tool for implementing emissions limitarianism, how should we spend the revenues that LMCT raises?

I will start with the first question. Wasteful emissions-generating functionings could be effectively curbed by banning the sale and/or use of motor yachts of a certain size, of private jets, of SUVs, and by banning private companies from taking humans near or beyond suborbital space, and so on. There are two ways to introduce a ban. One way is to pass a regulation with more or less immediate effect. The other is to adopt (or even to announce) a regulation in the more or less near future (e.g. in three or five years). The immediate effect option is obviously optimal for climate change mitigation, but it would have disastrous consequences for actors in the luxury industry, especially those who have invested in the production of the goods and services that would be affected by the ban, and consequently also for workers – which would imply, among other things, that the political feasibility of the immediate effect option is rather limited. The delayed effect option, on the other hand, would give stakeholders in the luxury goods industry time to restructure the sector in the most sustainable way possible, but would have little immediate impact on climate change mitigation (before the ban comes into force).

A limitarian carbon tax (LMCT) is a market-based instrument that offers a good compromise between the two ban options. Specifically, if the LMCT is set high enough, we can imagine that many wealthy consumers will decide to abstain from wasteful emissions in the very short term; a smaller fraction may instead remain indifferent to the tax, either because they are very rich, or because they are too fond of luxury, or because they make it a matter of principle. Consumers who reduce wasteful emissions immediately after the introduction of the LMCT will make an important contribution to climate change mitigation. Consumers who instead remain indifferent to the tax (probably members of the global top 0.1% or even 0.01%) will pay a very high cost. This will have two important consequences. On the one hand, it will allow industrial players to continue to make some profit from (polluting) luxury goods and services during the industrial transition phase. On the other hand, the LMCT will generate significant revenues.

It should be emphasised that neither the limitarian ban nor the LMCT succeeds in perfectly implementing emissions limitarianism. Indeed, one problem with this normative account is that it would have to be implemented by means of a policy tool that analyses the reasons behind every single functioning, even the most polluting and apparently most extravagant, in order to understand whether it is actually wasteful or not. Of course, this is extremely difficult, if not impossible, so we can only approximate. The fact that a given capability can, in the vast majority of cases, be replaced by a capability to achieve a less polluting functioning without jeopardising the enjoyment of the lower-level capability at stake, is sufficient to classify the realization of this capability as wasteful and thus either prohibit or to overtax it. This means, therefore, that any realistic policy of implementing emissions limitarianism runs the risk of causing injustice – in the previous section I made the hypothetical case of a person who, in order to take a life-saving antidote, must necessarily take a private flight. I will not go into the question of how often a capability that is generally wasteful turns out not to be so in one or more isolated cases, because that would require empirical evidence that is impossible to provide here. Suffice it to say that the LMCT deals with the problem of imperfect implementation better than a ban. With the latter, there would be no way for the life-threatened traveller to get to where the precious antidote is, since private planes would not sell and/or could not take off, whereas the LMCT would impose a surcharge, albeit a high one, on the private flight. Of course, the traveller in question may not be able to afford to pay the LMCT, but this could be partly remedied by asking others for help, given the dramatic nature of the situation. In short, the LMCT certainly fails to fully implement emissions limitarianism, although this serves as its justification, but I believe that alternative policy instruments are no better, and in many cases worse.

The second issue that needs to be addressed is how to spend the limitarian carbon revenues. First, it should be noted that while one of the purposes of a limitarian wealth tax is to transfer surplus wealth from the better off to people with unmet basic needs (so that this wealth is no longer surplus, i.e. it actually contributes to increasing someone's welfare), the purpose of the LMCT is not to raise revenue for a specific purpose, but to make it very costly for the rich, and especially the super-rich, to appropriate portions of the GHG budget for wasteful purposes. It follows that it is not possible to derive from the normative justification of the LMCT a moral obligation to allocate its revenues to a particular class of recipients. However, without claiming to exhaust the discourse, I will briefly describe four possible uses of limitarian carbon revenues and offer some arguments in favour of one of them.

The first possible use of limitarian carbon revenues is compensatory. Over the past 30 years, the global top 1% has emitted more GHGs than the global bottom 50% (the former group’s emissions have averaged about 15% of the total and the latter group’s emissions about 10%, see Chancel 2022). This may justify using limitarian carbon revenues to compensate and/or to provide adaptation funds to the main victims of climate change. The second possible use of limitarian carbon revenues is instead transition-oriented, and it aims at keeping the 1.5°C target alive. This requires revenues to be directed towards both decarbonisation strategies and the development of negative emissions technologies. Of course, the transition-oriented use of limitarian carbon revenues would also ease the mitigation burden on the 99% of the world's population. The third possible use of limitarian carbon revenues is intra-industry, and its main concern is that the reduction of wasteful emissions does not adversely affect vulnerable stakeholders in the luxury sector, from workers to suppliers to blameless investors. The fourth use is distributive in the broadest sense, and involves giving limitarian carbon revenues to the less well-off, regardless of the climate issue.

Despite the fact that all four possible uses of limitarian carbon revenues are compatible with the moral underpinnings of the LMCT, I believe that a mix of moral and pragmatic reasons should lead us to lean towards a combined use, in which intra-industry recipients are given lexical priority. On the one hand, it would be highly unfair for at least some blameless stakeholders of the luxury industry to suffer the negative consequences of an LMCT-induced production slowdown. Certainly, both the owners and the guilty investors of the luxury industry have a moral duty to use at least part of the profits they have made over the years in a highly polluting sector to protect the victims of the luxury transition (see Atteridge and Strambo 2020); but if this is not enough, or if these actors fail in their responsibilities, luxury carbon revenues will have to fill the gap. On the other hand, it is very difficult to imagine that the LMCT proposal will gain sufficiently broad political support if the luxury industry is not given the necessary financial resources for a green reorientation, in order to also safeguard jobs and investments. The recycling of limitarian carbon revenues within the luxury industry to fill the gap that cannot be filled by actors in the same luxury industry is therefore a crucial component of an LMCT proposal that is both fair and politically feasible.

Once the intra-industry use has been fulfilled, all three other uses have, in my view, equal normative force. Again, the choice depends very much on pragmatic considerations, and in particular on which needs are the most pressing. For example, if climate finance in the future continues to favour investment in climate mitigation at the expense of adaptation (Oxfam 2023), then a substantial part of limitarian carbon revenues could be used to meet the adaptation needs of the most vulnerable.

6. Conclusions

I argued that achieving some emission-generating functionings is wasteful, insofar as it unduly consumes sustainable emissions, without producing significant marginal improvements in welfare, compared to achieving alternative functionings. Under severe GHG budget constraints, achieving wasteful emission-generating functionings is morally wrong.

Many believe that a uniform carbon tax with equal carbon dividends (UCT + ECDs) exhausts the moral demands that can be placed on a climate transition policy, insofar as UCT + ECDs reduce aggregate GHG emissions in both an ambitious and cost-effective way (thanks to the UCT), without producing inequitable distributional effects, indeed by making the distribution of wealth more equitable compared to the pre-tax scenario (thanks to the ECDs). I argued that UCT + ECDs is unjust in that it forces the vast majority of people to reduce their wasteful emission-generating functionings, while excluding the rich, and especially the super-rich, who have enough disposable wealth to absorb the economic burden of the UCT without losing the capabilities to realise wasteful emission-generation functionings. For UCT + ECDs to pass the test of justice, it must be accompanied by a limitarian carbon tax that prevents the rich from depleting the GHG budget for wasteful purposes, just as the UCT does for everyone else.

As I said in the introduction, my argument is primarily one of ideal theory. In some cases, it may be right, but at the same time counterproductive, for a government to implement emissions limitarianism through, say, a limitarian carbon tax. This is because many rich people could simply move their luxury goods abroad, at least on paper. In response to this problem, some final remarks need to be made about the practical relevance of what has been discussed so far. The first consideration is that the main purpose of the article is to defend a normative claim, i.e. that UCT + ECDs without a LMCT is unfair. This claim, in my view, must play a role in the choice of policymakers on how to implement carbon tax reform. Whether and under what circumstances there are non-normative factors that should lead policymakers not to fully and/or immediately implement the normative claim I defend is a question that is primarily for them to resolve and that I do not address in the article. I would also add in this regard that the risk of what we might call wealth leakage concerns not only the implementation of emissions limitarianism, but limitarianism more generally, including economic limitarianism – and even more generally, any account of distributive justice that aims to redistribute a significant portion of wealth from the top down. The same applies to the implementation of any sufficiently ambitious climate policy package, whether or not it includes policy instruments targeted at the richest, which are likely to have the side-effect of inducing companies to relocate production abroad and increase foreign imports at the expense of domestic production. Both types of leakage create clear disincentives for any government to be the first to adopt socially ambitious policies. In short, the problem of moving from normative theory to practice is not unique to emissions limitarianism.

The second consideration is that it is now a well-established argument in the literature that many of the problems mentioned above can be addressed through coordinated and multilateral action by a sufficient number of countries. For example, both the efforts to create a global minimum tax to prevent unfair tax competition and those to create a climate club to prevent carbon leakage should be seen in this light. The former aims to get countries to agree on a minimum level of (large) corporate taxation below which no signatory should go (Bunn and Bray 2023). The second aims to build a group of countries that agree on more or less equivalent climate policies (not necessarily carbon pricing) and then pressure other countries to join the club – more invasively, for example through trade sanctions, or less invasively, by tapping into the common sense and responsibility of those outside the club (Falkner, Nasiritousi, and Reischl 2021). The limitarian carbon tax could, or rather in my view should, become the subject of an international agreement, possibly binding, between a sufficiently large number of countries (e.g. the G20) for a rapid and fair climate transition. However, a full discussion of this is beyond the scope of this article.

Acknowledgements

I gratefully acknowledge financial support provided to the Financial Ethics Research Group at University of Gothenburg, in particular from Vinnova (Sweden’s innovation agency) for the Sustainable Finance Lab, and from Mistra (the Swedish Foundation for Strategic Environmental Research) for the Finance to Revive Biodiversity program (Mistra FinBio). Earlier versions of this article were presented at a workshop on normative energy ethics organised by the Institute for Technology Assessment and Systems Analysis at the Karlsruhe Institute of Technology, at the Braga Meetings on Ethics and Political Philosophy at the University of Minho, at the Colloquium on Moral and Political Philosophy at the University of Graz, and at the Practical Philosophy Research Seminars at the University of Gothenburg. I am very grateful to the organisers and participants of these events for their valuable comments and advice. I would also like to thank the anonymous reviewers of the Journal of Global Ethics for their invaluable help in developing and clarifying many of the key points of the article.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Fausto Corvino

Fausto Corvino, At the time of writing, Fausto Corvino was Postdoctoral Researcher in Practical Philosophy at the University of Gothenburg (Sweden), where he was a member of the Financial Ethics Research Group. At the time of publication, he is a Marie Skłodowska-Curie Fellow at the Hoover Chair in Economic and Social Ethics, Institut supérieur de Philosophie, UCLouvain (Belgium). His research interests lie in intergenerational and global justice, climate ethics and economic ethics. Most recently, his research has focused on the ethics of market-based approaches to climate policy.

Notes

1 A carbon price can be introduced either in the form of a carbon tax or a cap-and-trade system, or by a combination of the two – in the former, the carbon price is set by policymakers, while in the latter, policymakers set the cap on emissions and let the market for emissions permits determine the carbon price. In this article I will focus on the carbon tax as the simplest and cheapest policy instrument to reach sectors such as transport and households, but everything I say can of course be extended to the cap-and-trade system. Moreover, a carbon price can be applied not only to CO₂ emissions, but also to other GHGs on the basis of their global warming potential expressed in terms of carbon equivalence (CO_2e) – e.g. how many units of CO₂ are needed to have the same global warming impact as one unit of CH₄.

2 For a critical discussion of the unfairness of the carbon tax, in abstraction from the use of revenues, see Tank 2020; Povitkina et al. 2021.

3 In this article, I only consider the distributional aspects of carbon pricing, leaving aside the more strictly ethical concern that putting a price on emissions permits means “commodifying” the atmosphere. For a discussion of this objection to carbon pricing see Neuteleers 2022.

4 One could also introduce the concept of fourth- or fifth-level capabilities, but for the sake of simplicity in this article I will limit the analysis to the third level.

5 I am grateful to an anonymous reviewer for prompting me to develop this point.

6 The axiological claim holds as long as well-being is measured in terms of capabilities. Accordingly, it cannot be ruled out that even capabilities to achieve wasteful emission-generating functionings can contribute to individual welfare, if the latter is measured in terms of subjective well-being. For example, having the most expensive car or house in the neighbourhood, city or world may contribute significantly to individual satisfaction, at least for some, while being irrelevant in terms of objective indicators of quality of life.

7 I am grateful to an anonymous reviewer for prompting me with this clarification.

8 Although it must be emphasised that full decarbonisation removes the normative force of both accounts as the GHG budget becomes irrelevant, we are still a long way from this and the purpose of both accounts is to contribute to a just transition to a carbon neutral future.