Research articleRecomposing consumption: defining necessities for sustainable and equitable well-being
Abstract
This paper focuses on consumption in the affluent world and the resulting level, composition and distribution of consumption-based emissions. It argues that public policy should foster the recomposition of consumption, while not disadvantaging poorer groups in the population. To combine these two imperatives entails making a distinction between goods and services that are necessary for a basic level of well-being, and those that are surplus to this requirement. The argument proceeds in six stages. First, the paper outlines a theory of universal need, as an alternative conception of well-being to consumer preference satisfaction. Second, it proposes a dual strategy methodology for identifying need satisfiers or necessities in a given social context. Then, it applies this methodology to identify a minimum bundle of necessary consumption items in the UK and speculates how it might be used to identify a maximum bundle for sustainable consumption. The next part looks at corporate barriers and structural obstacles in the path of sustainable consumption. The following part reveals a further problem: mitigation policies can result in perverse distributional outcomes when operating in contexts of great inequality. The final section suggests four ecosocial public policies that would simultaneously advance sustainable and equitable consumption in rich nations.
This article is part of the themed issue ‘Material demand reduction’.
1. Introduction: the issues
Climate change is recognized as an egregious threat to human habitats and welfare. It requires global responses, such as those recognized, though inadequately, at the 2015 Climate Change Conference in Paris. The rich world has a disproportionate responsibility and a duty to mitigate emissions very fast. This paper concentrates on consumption and consumption-based emissions for reasons explained below. It is concerned only with the developed economies of the OECD.
Simplifying greatly, there are three basic strategies for eliminating net carbon emissions by the second half of this century, the goal of the 2015 Paris Agreement:
C1. Ramping up the ‘eco-efficiency of production’, by reducing energy and emission intensities of production. This is the hegemonic strategy today.
C2. ‘Recomposing’ consumption: reducing consumption emissions by switching from high- to low-carbon services and goods, without necessarily cutting overall consumption expenditure.
C3. Post-growth: reducing then stabilizing absolute levels of consumer demand, moving towards a steady-state economy.
The assumption is that C1 alone cannot be adequate, for two main reasons. First, the embodied emissions of consumers in the developed world greatly exceed their production-based emissions, whereas the reverse is true of the developing world (https://www.oecd.org/sti/ind/EmbodiedCO2_Flyer.pdf). For reasons of justice and equity, it is the people and places benefiting from the goods and services emitting greenhouse gases (GHGs) that should count, not the people and places where these are produced. Consumption emissions are more closely related to well-being outcomes than production emissions [2].1 Throughout this paper, the emissions referred to are consumption-based emissions, not production-based or territorial emissions.
Second, it is extremely unlikely that improvements in overall ‘emissions efficiency’ across economies can reduce GHG emissions fast enough to ensure a safe climatic regime ([3], cf. [4,5]). To illustrate this, consider the decomposition of future consumption-based emissions in Sweden (figure 1 [6], cf. [7]). Under business-as-usual they more than double by 2050. If the rate of improvement of eco-efficiency matches that of the previous 15 years (1.4% per annum), Sweden's emissions would almost level out—a considerable achievement given that the ‘low hanging fruit’ will already have been picked. But even a further doubling of this rate (shown by the blue line) would not be enough to achieve the necessary emissions target of 2 tonnes per head by 2050 to have a reasonable chance of keeping global temperature rise below 2°C.
Figure 1. Scenarios for GHG emission from private consumption in Sweden 2012–2050. Source: Larsson [6, p. 3].
Thus, Larsson contends, ‘postmaterial’ (C2) or ‘sufficiency’ (C3) policies will also be needed. The next stage, shown by the green line, models a variety of C2 policies targeting, for instance, beef consumption, frequent air travel and expensive second homes. But these would still not be enough. A sustainable level of emissions is finally achieved in his model by introducing a C3 policy—a reduction in total consumption expenditure achieved by reducing reduced average hours of work. A major goal of this paper is to show how recomposing consumption (C2) might be operationalized.
However, another goal must have equal weight in this exercise: equity and the maintenance of some adequate level of well-being for all citizens. Recomposing consumption in an unfair way by hitting those with low incomes would be unethical, illegitimate and counterproductive. Unfortunately, as discussed later, the goals of sustainability and equity are not easily reconciled in unequal, highly commodified societies like our own.
Thus, we need a conceptual framework that can embrace both well-being and sustainable emissions. I argue that a theory of universal human needs can provide this framework. Since human needs are universal over time and space, and since there are strong moral claims for meeting the needs of all people, they provide a way of understanding well-being in rich countries that does not trespass on the need-satisfactions of other peoples now or of future generations, as the Brundtland Report advocates.
Need theory enables us to distinguish necessities from luxuries.2 To recompose equitably entails making a distinction between goods and services that are necessary for a basic level of well-being, and those that are surplus to this requirement. By prioritizing the former, need theory provides a bridge to relate social, global and intergenerational justice [8]. In the language of Giulio & Fuchs [9], it enables us to define a ‘consumption corridor’ between minimum standards, allowing every individual to live a good life, and maximum standards, ensuring a limit on every individual's use of natural and social resources in order to guarantee a good life for others in the present and in the future.
The argument proceeds in six stages. First (§2), the paper outlines a theory of universal need, as an alternative conception of well-being to consumer preference satisfaction. Second (§3), it proposes a dual strategy methodology for identifying need satisfiers or necessities in a given social context. Then in §4, it applies this methodology to identify a minimum bundle of necessary consumption items in the UK and speculates how it might be used to identify a maximum bundle for sustainable consumption. The next part (§5) looks at corporate barriers and structural obstacles in the path of sustainable consumption. The following part (§6) reveals a further problem: mitigation policies can result in perverse distributional outcomes when operating in contexts of great inequality. The final section (§7) suggests four ecosocial public policies that would simultaneously advance sustainable and equitable consumption in rich nations. A conclusion is given in §8.
2. Human need and sustainable well-being
Need theory challenges the dominant conception of well-being within economics, preference satisfaction theory. This rests on two normative foundations: that individuals are the best judges of their own preferences or wants, and that what is consumed should be determined by the private consumption preferences of individuals. It thus precludes questioning consumer preferences, except within narrow limits. However, it has been subject to numerous challenges.3 An alternative is required.
In [10], Len Doyal and I identify a conceptual space of universal human need, recognize cultural variety in meeting needs, but aim to avoid subordinating the identification of needs to such cultural contexts. Our essential premise is that all individuals, everywhere in the world, at all times present and future, have certain basic needs. These must be met in order for people to avoid harm, to participate in society and to reflect critically upon the conditions in which they find themselves. Only if we understand needs in this way—in universal terms, applied across time and place—can we plan for and measure progress towards our social and environmental goals, both globally and into the future.
The universality of need rests upon the belief that if needs are not satisfied then serious harm of some objective kind will result. This is not the same as subjective feelings like anxiety or unhappiness. It refers to functions not feelings. This harm implies obstacles to successful social participation. All our private and public goals are achieved on the basis of successful social interaction with others. It follows that participation in some form of social life without serious systematic limitations is our most basic human interest.
Basic needs are then the universal preconditions for effective participation in any form of social life. To do this, a person must be able to formulate aims, understand how to achieve them, and act to strive to achieve them in practice. Whatever a person's goals, whatever the cultural practices and values within which she lives, she will require certain prerequisites or basic needs, in order to strive towards those goals. In this way we identify, alongside social participation, health and autonomy as the most basic human needs. Survival is the most basic need, but all people require a modicum of good physical and mental health for effective social participation. Basic autonomy can be defined as the ‘the ability to make informed choices about what should be done and how to go about doing it’. In addition, we distinguish a higher level of critical autonomy: the capacity to compare cultural rules, to reflect upon the rules of one's own culture, to work with others to change them. At this higher level, drawing on imagination, past examples or comparisons with other ways of life, people can begin to question the taken-for-granted cultural frames of their own ways of life.
Human need theory has been advanced from a variety of perspectives [11]. One of these is eudaimonic psychology (as opposed to hedonic psychology), which provides argument and evidence for three universal psychological needs for autonomy, competence and relatedness [12,13]. Related to this, other studies show that individuals whose life goals are more focused on wealth, image and fame than on relationship, personal growth and community evidence less self-esteem, self-actualization and life satisfaction [14,15]. More materialistic individuals are most likely to be dissatisfied with life, to lack vitality and to suffer from anxiety, depression and addiction problems. Kasser [15] concludes that where growth of consumption is a key goal of a nation universal psychological needs are undermined. Indeed, this syndrome becomes self-reinforcing, as such people turn to money and possessions as a way of coping with distress.
Another contender is the capability approach advocated in different forms by Amartya Sen [16] and Martha Nussbaum [17,18]. Sen's version suffers from a fundamental problem: it provides no means for identifying basic functionings or capabilities common to a group of people let alone to all people. Sen famously rejects the search for, and lists of, universal valued functionings. By contrast, Martha Nussbaum has argued for universal ‘human functional capabilities’, derived initially from neo-Aristotelian reasoning subsequently replaced with a more Rawlsian idea of an emerging ‘overlapping consensus’ [17,19]. It is notable that more recently she relies ultimately on the language of ‘need’ [8,18].
Table 1 demonstrates the close agreement on core human universals between these three theoretical approaches [20]. It is important to note that none of these need theories embody any hierarchy of needs, as proposed by Maslow [21].
| theory | ||||
|---|---|---|---|---|
| Doyal & Gough [10] | basic needs | participation | health | autonomy |
| Nussbaum [17] | central human functional capabilities | affiliation | bodily integrity | practical reason |
| Ryan & Deci [12] | psychological needs | relatedness/belonging | autonomy; competence |
Universal human needs have (at least) five theoretical features that aid us in identifying ‘sustainable well-being’.
First, human needs are objective. The truth of the claim that a person needs clean water or a modicum of security in childhood depends on the objective physiological and psychological requirements of human beings and the nature of the satisfier, including its capacity to contribute to the health and autonomy of the person. By contrast, the truth of the claim that a person prefers Bowie to the Beatles depends on the nature of the person's beliefs about and attitudes towards the objects. Put another way, statements about wants are intentional, whereas statements of need are ‘extentional’: their truth depends on ‘the way the world is’ and not ‘the workings of my mind’ [22].
Second, human needs are plural; they cannot be added up and summarized in a single unit of account. In addition to the basic needs for health and autonomy, we identify a list of universal ‘intermediate needs’: water and nutrition, shelter and energy, a non-threatening environment and work practices, significant primary relationships, security in childhood, physical and economic security, education and healthcare. In a wide-ranging study, Alkire [23, ch. 2] surveys over 30 lists of ‘dimensions of human fulfilment’ and again demonstrates a broad overlap of components. It is notable that all such lists include not only material goods but psychological goods, activities and relationships.
Third, needs are non-substitutable: one domain of need-satisfaction or objective well-being cannot be traded off against another. More education is of no immediate help to someone who is ill through lack of vitamin C. Thus certain packages of need satisfiers are necessary for the avoidance of harm. This is quite different from consumer preferences in economic theory where substitutability is the default assumption: given a bundle of two goods it is always possible—by reducing the amount of one fractionally and increasing the amount of the other fractionally—to define a second bundle between which a consumer is ‘indifferent’ [24].
Fourth, needs are satiable. It can be shown that the amount of intermediate needs required to achieve a given level of health and autonomy diminishes as their quantity increases, eventually plateauing [25,26]. Thus the contribution of calories, dwelling space, even levels of childhood security, to basic needs can be satiated. In the case of the basic needs of health and autonomy, thresholds can be conceived where serious harm is avoided such that minimally acceptable levels of social participation can take place. The distributive principle entailed by human need theory is sufficiency: to bring all individuals up to such a threshold. It says nothing at this stage about inequalities above this level, but there are other reasons for justifying much more moderate inequality [27]. Needs come with a built-in distributive principle.
Fifth, needs are cross-generational. This is of great importance since global warming will progressively impose dilemmas of intergenerational equity. We can assert with much confidence that the basic needs of future generations of humans will be the same as those of present humans. To avoid serious harm and to participate and act within future human societies people will require the same logical preconditions: not just survival, but health and autonomy. Future people will have needs for affiliation, cognitive and emotional expression, understanding and critical thought. The epistemology of reasoning about needs remains extentional, not intentional, and thus avoids the indeterminacy of reasoning about future preferences. Until the genetic make-up of Homo sapiens changes significantly, our successors will need a specific amount of the full range of intermediate needs.
Together, this amounts to a remarkable—and pretty obvious—degree of knowledge about the constituents of future peoples' well-being. Compared with the indeterminacy of future generations' preferences or happiness (or of Sen's capabilities), a theory of need provides some firm foundations on which to build sustainability targets for public policy. The abilities to provide these components of objective welfare should be ‘passed down’ to future generations. In O'Neill's [24, p. 33] words: ‘Each generation needs to pass down the conditions for livelihood and good health, for social affiliation, for the development of capacities for practical reasoning, for engaging with the wider natural world and so on’.
Finally, human needs have a sound alternative ethical grounding that preferences do not: they come along with claims of justice and equity in tow.4 Claims of need make moral demands on agents that preferences do not [24]. Universal needs imply ethical obligations on individuals and claims of justice—universal rights and obligations—on social institutions. An important corollary of the moral import of human need is that meeting needs should be given priority over meeting wants, if the two conflict or if resources are scarce. Human needs, present and future, trump present (and future) consumer preferences.
3. Necessities and need satisfiers
When we turn to need satisfiers—the goods, services, activities and relationships that contribute to need-satisfaction in any particular context—things are quite different. The needs for food and shelter apply to all peoples, but there are a large variety of cuisines and forms of dwelling that can meet any given specification of nutrition and protection from the elements. It is essential to draw a sharp distinction between universal needs and specific satisfiers. Without it, need theory could justly be accused of being paternalist, intrusive and insensitive to context and culture.
How then can we identify necessities? In particular, since this article is focused on affluent societies—unequal, highly commodified and with a strong ideology of consumer sovereignty—how can a consensus ever be achieved around such contentious questions? My answer is to set out a broad methodology for identifying need satisfiers in particular contexts that is collective and consensual. The central argument is that identifying satisfiers must draw on two forms of knowledge: the codified knowledge of experts and the experientially grounded knowledge of ordinary people. The process to combine the two we call the ‘dual strategy’ [10, ch. 14].
The codified knowledge of the natural and social sciences enables us to determine the composition of many need satisfiers. It is embodied in the knowledge of practical experts, whether in health and medicine, engineering or biology, technology or policy science, and this knowledge is commonly used to help identify ‘what people need’: the components of a healthy diet, the education needs of children, the damage to our environment from pollutants and so on. Experts have a vital role to play in identifying need satisfiers.
Experientially grounded or practical knowledge is the entire range of understandings and accumulated problem solving of people in their everyday lives and contexts. This too must contribute to deciding what objects, activities and relationships are necessary or essential for well-being in any given context. Policies that ignore this input can be irrelevant, inefficient, stupid or oppressive; for example, building new housing estates far from employment, shops or social activities without public transport and concentrating poor families within them.
Clearly there is a dilemma here. Lauding the indispensable knowledge and power of science, technology, professions and experts runs the risks of uniformity, inflexibility, paternalism, disempowerment and domination, threatening the cultural integrity of groups and the autonomy of individuals. On the other hand, worthy initiatives to use experiential knowledge, to engage and empower citizens, carry the danger that sectional and short-term interests will threaten the identification of longer-term generalizable interests, especially when coupled with power differences in a context of inequality and media concentration.
Thus, we conclude that any rational and effective attempt to resolve disputes over need satisfiers ‘must bring to bear both the codified knowledge of experts and the experiential knowledge of those whose basic needs and daily life world are under consideration. It requires a dual strategy of social policy formation which values compromise, provided that it does not extend to the general character of basic human needs and rights’ [10, p. 141].5 This is admittedly an ideal form of negotiation. In the real world interests, institutions and power imbalances will thwart it. In implementing the dual strategy one can only insist that the debate is as informed, participatory and free of vested interests as is possible.
4. Defining sustainable necessities in practice
The urgent need today is to identify consumption bundles that (i) meet needs for decent living and (ii) are sustainable over time; in other words to define the upper and lower bounds of the consumption corridor. I look at each in turn.
Bradshaw et al. [31] first pioneered a methodology to define a minimum acceptable standard of living, including ‘more than just food, clothes and shelter. It is about having what you need in order to have the opportunities and choices necessary to participate in society’. The methodology used is ‘consensual discussions' among citizens informed at successive stages by expert input. The dual strategy method is now being practically applied to estimate ‘decent living’ minimum standards across a range of EU member states [32]. There are now accepted and proven methods of identifying necessities and distinguishing them from ‘luxuries’: an essential starting point if policies to pursue both human well-being and sustainability are to be developed.6
For example, the 2014 MIS (Minimum Income Standard) study in the UK involved 12 focus groups in which members of the public from a range of social backgrounds were tasked with producing lists of items that households would need in order to reach ‘an acceptable minimum standard of living’. These different groups involved pensioners, working-age adults without children and parents with children. They interacted with experts, including a nutritionist who helped to construct adequate diets and a heating engineer who specified home energy requirements [35].
This exercise has resulted in an agreed minimum consumption bundle that in some respects differs radically from the norm. The UK forums decided that private cars are luxuries and not necessary for a decent standard of living—citizens could use public transport plus taxis instead. The parents group, recognizing that it was increasingly likely that children in social housing would be expected to share bedrooms, concluded that this was not a standard that they agreed with and felt that in general the minimum should include a bedroom for each child of school age. Necessary food expenditures were agreed to be higher than present averages due to the consumption of more fresh fruit and vegetables.
The implication of this research is that focus groups can help to identify a list of necessities at any point in time. The list will change over time in reaction to socio-technical shifts. In 2012, the group discussing the needs of households with children included a cheap second-hand car as a necessity for the first time, due to the decline of public transport. By 2014 all groups, including the pensioners group, regarded a computer, Internet access and a cheap mobile phone subscription as a necessity. But, these apart, the definition of necessities has changed relatively slowly: the UK 2014 budgets were remarkably similar to the 2008 budgets, despite the financial crisis and recession in the meantime, reflecting a consistency in the ways that members of the public interpreted the rationales of necessity [35].
If the entire population were living on the ‘decent life budget’ then emissions would be lower: 37% lower than actual consumption-based UK emissions in 2004 according to research by Druckman & Jackson [36] (figure 2). Almost every category of consumption would deliver lower emissions except for food. Reductions are particularly noticeable in transport, household energy, restaurants and hotels, and miscellaneous household goods and services. In this hypothetical scenario, total UK consumption emissions would fall from 26 tonnes CO2 equivalent per average household to 16 tonnes.
Figure 2. GHG per household per annum. Source: Druckman & Jackson [36, fig. 2].
This is a significant reduction, but the problem is that this still equals 7.3 tonnes per person—well above the 2050 goal of 2 tonnes per person. In the UK and other rich countries, the emissions of the minimum consumption bundle exceed those of the maximum sustainable consumption bundle—and by more than three times! Studies in other countries replicate this finding: in Finland, people receiving minimum income benefits exceed ecologically sustainable lifestyles by a wide margin [37]. This is to be expected and backs up the Swedish modelling in figure 1: C2 policies complement but do not replace C1 policies. To make consumption sustainable within existing socio-technical structures would deprive citizens of a vast range of goods and services that they have agreed are necessary for effective participation in modern life.
But can the dual strategy be applied to pursue a different goal—to arrive at a consensual measure of more sustainable maximum levels of consumption? Returning to four of the UK focus groups and using the same methodology, one study explicitly asked them to consider their carbon footprint and agree on what might be ‘publicly acceptable’ [38]. The results were discouraging. There was some support for reducing heating energy by wearing more clothes at home, maintaining separate temperature zones and switching to energy-saving appliances, and some interest in cycling more and using public transport if it was more convenient and time-saving. There was very little interest in reducing consumption of red meat or eating only seasonal vegetables. The research found that people had got used to having a high degree of choice and did not want to be told what and what not to consume [38]. This is just one small study in one country, but it is not optimistic that within contemporary society consumption can be consensually recomposed to any significant extent.
More transformative approaches will clearly be needed. One approach is to use Max-Neef's [39] framework of human scale development (HSD), developed to enable small communities in Latin America and elsewhere to devise their own routes to a more people-centred idea of development. It involves longer participatory workshops that can question goals, behaviours, satisfiers and infrastructures more radically and over longer time spans. The HSD framework considers need satisfiers to be systematically related and interdependent and then searches for ‘synergic satisfiers’ that simultaneously meet different kinds of needs. To identify these entails a ‘deep learning journey’ going beyond focus groups to longer-term and community-based projects [40]. This can best take place within communities already inspired by sustainability goals, such as eco-villages and transition towns. But two problems remain. First, issues of equity and sustainability are rarely given equal consideration: HSD initiatives won't always pay attention to environmental consequences, and sustainability initiatives won't always pay attention to distributional and poverty issues. Second, there is a consistent problem in scaling up such initiatives in the face of overwhelming power imbalances and system lock-in. It is time to turn to these.
5. Barriers to sustainable consumption in practice
Theories of consumption vary on a wide spectrum from those privileging consumers as agents to those privileging corporate power and structural path dependency.7 This section discusses only the last two obstacles: an asymmetry in power and knowledge between corporations and consumers, and ‘lock-in’ to path-dependent structures. In selecting these for discussion I am making the assumption that neither orthodox consumer theory nor behavioural economic theory adequately address the determinants of consumer preferences. Consequently, the respective policy approaches condoned by these theories—providing better information and ‘nudging’—will not be sufficient [42–44]. Other contributors to this theme issue provide a wider perspective on the role of market actors—corporations, consumers and new market forms—in fostering lower-carbon, more sustainable consumption [45,46].
The corporate framework within which consumers operate and ‘choose’ is heavily biased to create novel consumer goods and services and to encourage absolute increases in consumption [47]. Advertising and marketing—the ‘engineering of consent’ [48]—is of obvious importance in shaping consumer behaviour, not always to the benefit of consumers.8 But these more obvious agency-based examples of corporate power rest on a less visible structural power, both the ability to determine investment and employment, and the discursive power of ideas such as consumer sovereignty and economic growth [51], [52, ch. 4]. For example, Fuchs et al. [53] describe in some detail how such hidden power has delivered abundant cheap meat in modern society and thwarted attempts to regulate and limit it, despite external costs in health and emissions.
Systems of provision lock households into patterns of consumption, as when rural and suburban residents have no viable alternatives to driving the car. ‘Lock-in’ refers to consumption driven by structural and institutional features of the society largely outside the scope of individual choice. For example, many car journeys—to commute, shop, drive children to school, etc.—might be caused by the spatial and social contexts within which people live and to which they must necessarily adapt. The needs framework can be adapted to provide some insights here. Need satisfiers can be viewed as a hierarchy, as, for example, the need for use of a car [54] (table 2).
| basic needs | social participation, health and autonomy | |||
|---|---|---|---|---|
| intermediate needs | income | nutrition | healthcare | relationships |
| satisfier level 1 | employment | shopping | medical visits | social visits |
| satisfier level 2 | travel | |||
| satisfier level 3 | car | |||
Conceiving demand in this way helps analyse the nature of the lock-in. For example, is use of a car a necessary means to travel (level 3)? Does access to work and shopping require travel—to what extent could the Internet replace work, shopping and social interaction (level 2)? To what extent does demand for travel reflect the spatial distribution of physical infrastructure (level 1)? The hierarchy demonstrates that, in defining what is needed and sustainable, attention must be paid to ‘lock-in’ as well as ‘luxury’ [55].
These two system-level barriers to sustainable consumption—corporate shaping of demand and inherited systems of provision—interact. Consumption actions result in unintended consequences that then shape future collective arrangements and future individual consumption actions. To investigate this takes us into social, structural and complexity theory that is beyond the scope of this paper. Giddens [56] refers to this interplay between agency and structure as ‘structuration’. Individual actions result, via a composition effect, in aggregate outcomes that have little relationship with the original intentions of individual actors. Human history is ‘created by intentional activities but in a non-intentional way’ [54].
This approach sheds light on some issues of justice in a climate-constrained world. Car use in a car-dependent setting may be essential for social participation and thus its denial or removal would constitute an unjust harm. Yet if this injustice is overcome at the expense of still higher transport emissions, it will entail further degradation of the natural environment and an unjust imposition on other peoples and on future generations.
This is not to argue that there are no springs of change. Consumption practices are affected by longer-term shifts in tastes and by socio-technical innovations. For example, consumer demand for private transport is falling among young adults across the developed world for a variety of reasons and car companies have not been able to halt that trend [57]. Similarly the entry of new market players such as ZipCar and Uber may result in reductions of car ownership and more efficient use of existing resources [46].
6. Income inequality and equity dilemmas
Consumption demand in market systems is shaped not only by preferences but also by the level and distribution of incomes. Two further problems confronted en route to recomposing consumption are rising inequality and dilemmas of redistribution.
Inequality of incomes has risen since around 1980 across the OECD, though at different rates [58]. Argument and evidence show that this has exacerbated status competition in society, especially over ‘positional goods’. Coupled with corporate power it enhances competitive pressures to consume more and more [59–61]. Standards are inherently relative and involve comparisons with others. When the income distribution becomes stretched, more of the population, even the middle class, become aware they are falling behind the rich. In the neoliberal period since around 1980, accelerating inequality has, notably in the Anglosphere, driven up household debt and, to compensate for lower wage growth, halted the previous decline in average hours of work. Both debt and work hours add further to the link between inequality and unsustainability [62,63].
Inequality can also generate equity dilemmas in implementing carbon mitigation programmes. Though consumption emissions rise with household income, they do so at a slower rate. Lower-income households face noticeably higher emissions per pound spent (figure 3). Thus any rise in carbon prices, when generalized throughout the economy, will impact on lower-income households more. The degree of this regressive impact varies according to the type of consumption. Expenditures on, and emissions from, domestic energy and food take a proportionately higher share of incomes lower down the income scale than spending on, and emissions from, transport, consumer goods and personal services [64]. As a result, carbon mitigation programmes can increase poverty and inequality.9
Figure 3. Per capita GHG emissions per pound of income, by income decile and sector, UK 2006. Source: Gough et al. [64, table 8].
This follows because necessities have in general a higher-than-average carbon footprint (appendix A). This poses a potential tension between securing emission reductions and ensuring an equitable distribution. It suggests that simply redistributing income to low-income households might, ceteris paribus, raise rather than lower total emissions as more money is available to spend on high-emitting basic goods. This lends support to a conclusion of Koch & Fritz [65, p. 698] about welfare states: ‘The same mechanism that defuses the socio-economic inequalities inherent in capitalist development ensures the inclusion of an increasing number of people in environmentally problematic production and consumption processes’.
Thus carbon pricing is always inequitable; yet redistributing incomes will, all other things being equal, increase emissions! There appears to be a fundamental tension between meeting basic needs and cutting emissions.
However, much depends on where the redistributed income and associated emissions come from—if it is due to a cap on excessive emissions by the affluent this could more than offset the higher emissions of low-income households on, for example, heating and food. Moreover, comparative research finds that carbon intensities for housing and domestic energy are much lower in Norway and Sweden than other OECD countries, reflecting the extensive presence of district heating, biomass, hydro-electricity and better insulation in the two Nordic countries [66]. In which case taxing energy would not discriminate against poorer households. Different forms of technology and infrastructure can thus profoundly improve the eco-efficiency of specific sectors and thus the potential equity--sustainability trade-off pictured in the table [67] in the appendix. Nevertheless, we have demonstrated that there are potential trade-offs between sustainability and equity when reforming the sphere of consumption.
| low GHG less than 1 tonne CO2 equiv./£000 | high GHG greater than 1 tonne CO2 equiv./£000 | |
|---|---|---|
| necessities: | alcoholic beverages/tobacco (0.7%) | all domestic energy (electricity, gas, other fuels) (26.9%) |
| income elasticity < 1 | communication (1.2%) | food (12.9%) |
| luxuries: | clothing and footwear (2.6%) | all transport (vehicle fuels, other transport) (22.5%) |
| income elasticity > 1 | other housing (2.3%) | |
| furnishings (5.0%) | ||
| recreation and culture (8.7%) | ||
| restaurants and hotels (5.0%) | ||
| private health (0.5%) | ||
| private education (0.3%) | ||
| miscellaneous (4.2%) |
7. State policies for recomposing consumption
The conclusion thus far is that changing consumer preferences will be constrained by corporate power, system lock-in and the interaction between the two. Growing inequality also makes recomposing consumption more difficult and creates dilemmas in redistributing incomes in high-carbon economies. The upshot of the above is that recomposing consumption will require in addition some hefty top-down state interventions. Many will be required to influence the eco-efficiency of production (the C1 goal).10 But a range of novel C2 interventions will also be needed to recompose consumption in a fair way: a suite of ‘ecosocial policies’ that simultaneously and explicitly pursue both equity/justice and sustainability/sufficiency goals. I conclude with some suggestions here.
(a) Taxing high-carbon luxuries: smart value-added tax
The economist Robert Frank [59] has long argued for a progressive consumption tax, on sustainability as well as equity grounds. A major justification is that it would curb the rate of expansion in mass wants and desires. However, it may not be progressive, since it would benefit high-saving, richer households. But in all OECD countries except the USA there exists an explicit tax on consumption—value-added tax (VAT)—that raises about a fifth of all tax revenues and is a major funder of social programmes. The VAT rate in most EU countries today varies between 20% and 25%, but in all countries there are exemptions and lower rates applied to certain goods and services. These usually include basic foodstuffs but vary greatly across countries, following pressures from particular industries and lobby groups.11
The argument for a ‘smart VAT’ is to introduce deliberate variations in the rate, higher to discourage bad consumption and lower to encourage desirable consumption. The proposal has mainly been advocated on health and well-being grounds, to improve healthy eating and discourage obesity [69]. But it could also be amended to take account of sustainability. Thus high-GHG goods that harmed well-being would attract the highest VAT rates, while low-carbon goods that improve well-being would be taxed at lower or even negative rates (amounting to a subsidy).
This entails an assault on common notions of consumer sovereignty and would meet corporate, retail and consumer opposition discussed above. To secure legitimacy and public support, Fell [69] realizes that forms of citizenship engagement along dual strategy lines would be necessary. To decide what goods are virtuous and what harmful he proposes regular deliberative dialogue in focus groups informed by environmental and social experts. This exercise could draw on the MIS studies and equivalent exercises such as the Northern Ireland ‘healthy food basket’. These decision would then be fed up to a second stage of public decision-making to decide on the different VAT rates. Fell concedes that this stage would likely require economic expertise, such as the Office of Budgetary Responsibility in the UK, but to be effective smart VAT rates should vary widely between perhaps +25% and −20%. To publicize what is happening they should be displayed on all goods in shops and should be monitored and adjusted as necessary.
There is no reason why the emissions footprint of consumption practices should not be integrated into this process alongside health and other basic need goals. Smart VAT provides a broad framework within which specific proposals to tax high-carbon non-essentials could fit, such as a frequent flyer levy or a global tax on business-class flight tickets [70]. It provides a workable method to scale up a dual strategy for sustainable consumption.
(b) Social tariffs
Water, electricity and gas utilities are, up to certain level of consumption, basic but high-carbon necessities. Direct regulation of utility prices would provide a simple method of securing both greater sustainability and equity. This would require energy and water companies to operate a ‘rising block tariff’, with lower tariffs for initial units of electricity, gas or water consumed, and higher tariffs for successive units. At present energy tariffs work in the opposite way, mirroring the cost structure of utilities faced with large fixed costs to maintain the network. To reverse this would recognize the basic need component of the first block of household energy and the progressive choice element in successive units.12 The total average price of domestic energy would continue to rise over time, as part of the carbon pricing strategy, but the distribution of the burden would be skewed more to higher-consumption households. Related tariffs schemes have been implemented in some countries in southern Europe [71]. It would be difficult to administer in the privatized energy system of most countries today and is opposed by energy suppliers.
(c) Widening social consumption
Turning to the expenditure side of fiscal policy, health, education and social care are further necessities which for other reasons are usually supplied free or at low cost to all citizens via collective public funding. There are several reasons for raising the share of state social consumption as part of an ecosocial strategy. First, tax-financed social consumption such as health services, social care and education is inherently redistributive: allocation according to need, risk or citizenship, not market demand, automatically serves redistributive social goals—even if the tax system is neutral rather than progressive. Second, research shows that this saves carbon. For example, the US healthcare system directly accounts for 8% of emissions in the USA, compared with 3% of UK emissions directly stemming from the NHS. This is due both to the greater macro-efficiency and lower expenditure shares of health in the UK, but also to lower emissions per pound or dollar spent, due to better allocation of resources and procurement practices and to explicit carbon-saving programmes [72].
(d) Rationing: personal carbon allowances
The most direct and radical way to simultaneously recompose consumption in a sustainable and equitable direction would be ‘downstream’ rationing of carbon between households. A country's total GHG emissions would be capped (decreasing year by year) and this amount divided into equal annual allowances for each adult resident, with variable rates for children [73,74]. In effect, a dual accounting standard and currency is developed: energy, goods and services would have both a money price and a carbon price. Some propose a trading element whereby those who emit less carbon than the average could sell their surplus allowances and gain, while higher emitters would pay a market price for their excess. Advocates claim that a PCAT (personal carbon allowance and trading) scheme covering domestic energy, road fuel and air travel would curb consumption in a progressive way. However, PCAT does not avoid all issues of fairness; for example, those living in inefficient or underutilised housing, dependent on car travel, or with special needs would face difficulties. Too many exceptions to the standard allowance could undermine the scheme, but too few would result in rough justice, which could undermine public support. Integration with upstream cap and trade systems, e.g. the Emissions Trading System (ETS), would raise problems. There are also several practical as well as political issues to be faced [74].
These are some ways in which democratic states can begin to shift contemporary consumption practices to both reduce emissions and pursue the equitable and just satisfaction of basic human needs. In concentrating on ‘top-down’ policies I am not underestimating the powerful role of local action and bottom-up perspectives [75–77]. But states will remain central actors in the Anthropocene [78].
8. Conclusion
To have a hope of mitigating climate change fast enough means targeting consumption in rich countries alongside production. Yet consumer demand rarely figures in climate policy because it questions consumers' tastes and challenges some formidable corporate interests and structural obstacles. This paper makes the case for a new policy domain—to ‘recompose’ consumption by switching from high- to low-carbon goods and services, without necessarily cutting overall consumption expenditure. At the same time this recomposition must be fair, on grounds of justice and legitimacy, especially since cutting emissions can have regressive distributional effects.
To target both goals—sustainability and equity—in consumption we need to distinguish necessities or basic goods from luxuries or surplus goods. Theories of human need can identify universal and objective components of well-being applicable across time and space. Identifying need satisfiers or necessities requires a further methodology fostering a dialogue of citizens and experts. We show where these methods have been used to identify bundles of satisfiers for an acceptable minimum standard of living and then turn to examine problems in applying this method to identify a maximum sustainable consumption standard. Under present socio-technical conditions, the minimum exceeds the maximum, so demand policies must be implemented alongside eco-efficiency policies in production.
Recomposing consumption along these lines can provide guidance of what consumption items to target and what not. It will require the efforts of numerous agents, citizens, civil society, business and others. But a major role for government will endure and expand as a conduit for linking the dual strategy to political action. To ensure that emission reduction and social justice go hand in hand, the paper ends by outlining a range of ecosocial policies.
The recomposition of consumption advocated here has three advantages. First, it forms a distinct strategy for climate mitigation alongside green growth and improvements in emissions efficiency. Second, it relates consumption to fundamental aspects of well-being. Third, it embodies a distributional ethic and a case for redistribution. It could play a central role in a strategy for material demand reduction.
Competing interests
The author declares that there are no competing interests.
Funding
No funding has been received for this article.
Acknowledgements
I am grateful to three anonymous referees of this paper, and at an earlier stage to many colleagues who advised on the work in progress. The Grantham Research Institute on Climate Change at the LSE provided helpful research assistance for this paper.
Appendix A
It is possible to use a different ‘consumer behaviour’ method to distinguish necessities from luxuries according to the income elasticity—the ratio of the percentage change in expenditure emissions to the percentage change in household income—of different consumption items. Necessities have an income or expenditure elasticity of less than 1, non-necessities or ‘luxuries’ greater than 1 [79]. This will not necessarily agree with the social consensual method outlined in the paper, for example, regarding alcohol and tobacco as necessary goods, but there is a strong overlap. Table 3 plots income elasticities against the emissions intensity of the different categories in the form of a 2 × 2 matrix. It is clear that the core necessities in the top right-hand cell are high emitters of carbon or other GHGs and contrast with luxuries in the cell diagonally opposite. The scale of high-carbon necessities would be still greater if it also included essential transport—the data used does not permit any disaggregation of transport.
Footnotes
1 It is true that people and places where goods are produced benefit from employment, wage income and profit, but they also suffer pollution and health costs. The need approach that informs this paper defines well-being in terms of non-monetary outcomes.
2 ‘Luxuries’ suggests mansions and private yachts, but I cannot think of a better term to refer to everything that is not a ‘necessity’.
3 In brief, on the grounds of subjectivity, epistemic irrationality, endogenous and adaptive preferences, the limitlessness of wants, the absence of moral evaluation and the non-specificity of future preferences [8].
4 ‘For standard economic analysis everything is a preference: the epicure's wish for a little more seasoning, the starving child's wish for a little water, the collector's wish for one more painting, and the homeless person's wish for privacy and warmth, all are preferences’ [28, p. 55].
6 The account that follows concerns rich OECD countries, but the ‘socially perceived necessities’ approach has been used elsewhere, such as South Africa [33]. The South African study found that deprived townships arrived at similar lists to the white population. For example, under protective housing they list as necessities: mains electricity, a house protective against weather, warm, fridge, separate bedrooms for adults and children, a flush toilet. Poorer countries will arrive at a less demanding list of necessities, but there is evidence that across the world the list of necessities is converging. For example in housing again: a minimum floor space per person, solid construction, continuous electricity supply; 90% of the world's adult population now has access to a basic mobile phone [34].
7 This contrast echoes Duesenberry's famous aphorism: ‘Economics is all about how people make choices; sociology is all about how they don't have any choices to make’ [41, p. 233].
8 Two examples: one half of poor UK households persuaded to switch energy suppliers following doorstep sales found themselves with a worse energy deal than before [49]. The vast majority of adults in Great Britain (82%) have regretted a purchase in the past year, amounting to 2–10% of total consumer spending [50].
9 For example, the UK imposes a suite of obligations on energy companies to aid poorer and less well-housed groups: improve insulation, renew boilers, etc. Yet because these improvements are paid for out of higher energy bills they bear more heavily on lower-income households and paradoxically increase rather than reduce ‘fuel poverty’ [49].
10 Four policy pillars for GHG mitigation can be distinguished: carbon pricing, strategic carbon-saving investment, public regulation, and behaviour change and public engagement (cf. [68]). All of these can contribute to the achievement of C1 eco-efficiency policies. But some can also influence C2, the composition of consumption. For example, the minimum consumption scenario cited above assumes that UK housing is insulated to a higher quality than at present, which will require an ambitious programme of retrofitting entailing public guidance, regulation, loans and subsidy. I do not consider other C1 interventions in this paper.
