Mechanisms explaining the impact of economic inequality on environmental deterioration

Ecological Economics 116 (2015) 191–200

Contents lists available at ScienceDirect

Ecological Economics journal homepage: www.elsevier.com/locate/ecolecon

Surveys

Mechanisms explaining the impact of economic inequality on environmental deterioration Alexandre Berthe, Luc Elie ⁎ a

GREThA — UMR CNRS 5113, University of Bordeaux, Avenue Léon Duguit, 33608 Pessac, France

a r t i c l e

i n f o

Article history: Received 15 September 2014 Received in revised form 10 April 2015 Accepted 27 April 2015 Available online 15 May 2015 Keywords: Economic inequality Environmental pressures Environmental policies

a b s t r a c t Rising economic inequality, often considered intrinsically harmful, is increasingly being viewed as having a number of secondary impacts as well, including impacts on health and economic growth. The ongoing nature of today's environmental crisis also raises questions about inequality's role in environmental deterioration. Despite the large number of papers that have been written on this topic, no theoretical or empirical consensus presently exists. Firstly, our article identifies that authors' conclusions in this area depend on their hypotheses regarding 1) the relationship between individual income and individual environmental pressure, 2) the impact of inequality on the social norms that influence individual environmental pressure, 3) the interests that social groups have in degrading or protecting the environment, 4) how these interests play out in terms of political demands, and 5) how these political demands translate into political decisions. Secondly, the study shows that, despite enabling a general test of the causal relationship between inequality and the environment, the empirical methods utilised do not account for the full range of theoretical mechanisms in play. Hence the suggestion that a research programme be launched to conduct empirical studies of the five aforementioned hypotheses by applying a recursive approach. © 2015 Elsevier B.V. All rights reserved.

1. Introduction The contemporary period is characterised by intensification of global social and environmental crises. The social crisis has manifested in rising economic inequality since the 1980s in most countries worldwide, with levels sometimes approximating those seen in the United States in the early 20th century (Piketty, 2014; OECD, 2011, 2014). The environmental crisis has manifested, especially since the 1950s, in the rapid rise in environmental pressures1 (Steffen et al., 2011), resulting notably into widespread changes in natural ecosystems (Millenium Ecosystem Assessment, 2005) and climatic disturbances (IPCC, 2013). The simultaneous worsening of each of these crises raises a question as to whether they are mutually reinforcing. It is against this background that an environmental justice literature analyses how, and to what extent, the environmental crisis reinforces social inequality by superimposing environmental inequality onto economic inequality (Laurent, 2011). In particular, this literature shows that poor people are highly dependent on their environment and suffer more from the effects of pollution (Martinez-Alier, 2002) and climate change (Olsson et al., 2014). In the opposite way, the social crisis could exacerbate the environmental crisis

⁎ Corresponding author. E-mail addresses: [email protected] (A. Berthe), [email protected] (L. Elie). 1 The rest of the paper uses terms such as environmental deterioration or pressure to refer both to the overuse of resources and to pollution.

http://dx.doi.org/10.1016/j.ecolecon.2015.04.026 0921-8009/© 2015 Elsevier B.V. All rights reserved.

and prevent its resolution. Indeed, some studies analyse how social inequalities usually inhibit cooperation in the case of local commonpool resource management (Baland et al., 2007). However, this literature does not explore the impact of the social crisis on the environment at a larger scale, i.e. the national or regional level. A number of studies investigate this last phenomenon by analysing the effects of economic inequality on environmental policies and pressures. Despite many different contributions, this literature has yet to reach a theoretical or empirical consensus. From a theoretical perspective, Boyce (1994), Magnani (2000) and Wilkinson and Pickett (2010) propose a variety of original explanations that imply that inequality negatively affects the environment. Scruggs (1998) and Heerink et al. (2001) develop explanations defending the opposite view. In the first group, Wilkinson and Pickett (2010) consider that inequality has led to individuals adopting consumerist and individualistic behaviours toward the environment. Magnani (2000) theorises that inequality means poorer median segments, hence policies promoting growth rather than environmental protection. Lastly, according to Boyce (1994), inequality reinforces the power of the affluent, who has no interest in protecting the environment. In the second group, Heerink et al. (2001) consider that inequality implies a concentration of wealth amongst the affluent, whose economic behaviours generate fewer environmental pressures. Scruggs (1998) asserts that by concentrating power in situations of great inequality, affluent groups actually encourage environmental protection policies. Such a range of explanations implies that there is a competition between them. The present paper develops

192

A. Berthe, L. Elie / Ecological Economics 116 (2015) 191–200 Individual environmental pressure

(1) Individual income

Individual environmental pressure

(2) Individual income

Individual environmental pressure

(3) Individual income

Fig. 1. Individual environmental pressures and incomes. Source: authors, adapted from Scruggs (1998).

an analytical structure that helps detecting the mechanisms that participate to the differentiation of the explanations. Furthermore, it demonstrates the necessity of comparing such mechanisms using observable data. Existing empirical studies can help to substantiate the hypothesis that inequality adds to environmental deterioration, but they have not been conducted at a sufficiently detailed level to advance understanding of which mechanisms are actually at work. The rest of the paper is organised as follows. Section 2 offers a theoretical structure of this field and compares existing developments. Section 3 analyses empirical studies' contributions and limitations. Section 4 offers a conclusion. 2. From Inequality to Environmental Pressure: Structuring a Theoretical Field The present section summarises and compares theories explaining how economic inequality impacts environmental quality and places them within a coherent framework. Boyce (1994) is one of the first to focus on inequality as a potential cause of environmental deterioration. After the answer proposed by Scruggs (1998), some authors attempted to integrate these opposing analyses (Boyce, 2003, 2007; Heerink et al., 2001). Nevertheless, academic literature would subsequently hone in on this topic without taking into account the full range of conceived mechanisms. The theoretical literature can be structured in two channels relating inequality and environmental pressure: one based upon the economic behaviours of households (Section 2.1); and the second based upon the determination of environmental policies (Section 2.2). 2.1. How Income Distribution Affects Environmental Pressure Through the Economic Behaviour of Households Economic behaviours of households, based mainly on consumption of goods and services, have a direct influence on environmental pressure. By influencing the level and content of aggregate consumption, the level of inequality impacts environmental pressure.

2.1.1. Methodological Individualist Approaches: Environmental Pressures Aggregated From Individual Economic Choices In an approach based on methodological individualism, Scruggs (1998) shows that the effect of income distribution on environmental pressures depends upon the relationship between environmental degradation and income at the individual level. The author envisions three types of relationships that determine how variations in income inequality affect aggregate pressure (Fig. 1). These three situations differ in the way they describe how individual income level affects environmental pressures through individual economic behaviour. All these relationships assume that environmental pressure increases in line with income (except for the area covered by the dotted lines in Situation 1). However, they each provide a different hypothesis regarding the direction of the marginal variation in environmental pressure. In Situation 1, the curve is concave, signifying a marginal decrease in environmental pressure. In this case, a poorer person's utilisation of an additional unit of income generates greater environmental pressure than the utilisation of this unit by a more affluent person. A society in which affluent persons have extra income at their disposal would therefore, ceteris paribus, generate less environmental pressure than a benchmark society might do. In other words, the direction of the marginal variation in environmental pressure fundamentally determines how inequality affects aggregate environmental pressures as they relate to economic choices.2 In Situation 1, increased inequality results in reduced environmental pressures. Situation 2 is the opposite, with marginal environmental pressure here increasing for each individual, which implies a society where less inequality generates fewer aggregate environmental pressures. Lastly, Situation 3 is an intermediary case where income redistribution has no impact on environmental deterioration as long as global revenue remains the same. Having adopted this approach, Scruggs (1998) and Heerink et al. (2001) consider that Situation 1 is the most representative of reality. To justify this hypothesis, they assume that environmental quality is a superior good, i.e. one whose demand increases at an increasing rate as income rises. In this way, Scruggs (1998) utilises Inglehart's postmaterialism theory (Inglehart, 1990), which holds that a minimum level of affluence and satisfaction of material needs are required for environmental preferences to materialise. Moreover, to explain the first curve, Heerink et al. (2001) claim that affluent households are able to substitute polluting goods with environmentally friendly goods.3 Given this hypothesis, Heerink et al. (2001) consider supplementing the first curve with a decreasing segment beyond a certain income threshold so as to draw an environmental Kuznets curve at the level of an individual household (see the section with a dotted line in Fig. 1). The idea is that more affluent households are associated with lower levels of individual environmental pressure. However, even if post-materialist values develop as income increases, they do not necessarily lead to changes in consumption behaviour. For instance, according to Fleurbaey et al. (2014), environmental values explain approximately 20% of the environmental behaviour leading to greenhouse gas (GHG) emissions. This value–action gap can be explained by the competition between the environmental concerns with other concerns, such as specific kinds of consumption in which affluent people engage as their desire and ability to reach a higher standard of living increases (more energy intensive transportation and housing, etc.). Other explanations for the value–action gap include consumer misinformation about the ecological impacts of their pro-environmental behaviour (named behaviour–impact gap, see Csutora, 2012). Furthermore, other arguments can relativise the reality of Situation 1. In particular, Roca 2 Heerink and Folmer (1994) formally demonstrate the links between the level of inequality and the average level of an output in case of non-linear relationships between income and this output at the individual level. 3 The analysis developed here can apply to the environmental pressure due to consumption by normal households or individual producers during their production process. Regarding the second household type, Heerink et al. (2001) evoke the example of agricultural households in developing countries (Heerink et al., 2001: 360–361).

A. Berthe, L. Elie / Ecological Economics 116 (2015) 191–200

(2003) identifies that richer households can displace the costs of the consumption-related environmental degradation to other places, voluntarily or otherwise, in order to continue their environmentally damaging practices without suffering the consequences. Furthermore, the total effect of consumption on the environment depends on the content but also on the magnitude of that consumption. Indeed, Situation 1 might also be explained by the fact that the marginal propensity to consume is commonly recognised as something that tends to decrease with income.4 After all, even if poorer and more affluent actors both consume products that are identical in terms of environmental pressure, the fact that the latter saves a higher proportion of their income may imply that they generate less environmental pressure per monetary unit.5 A number of empirical analyses can be mobilised to determine the shape of the curve. Heerink et al. (2001) base their individual environmental Kuznets curve theory on an empirical study demonstrating, for a sample of California households, that vehicles owned by more affluent households produce less GHG emissions than vehicles owned by their poorer counterparts (Kahn, 1998). However, the use that Heerink et al. (2001) make of this study has been questioned by Cox et al. (2012) who have shown, using a sample of Scottish households, that including the number of miles travelled has a greater effect on GHG emissions than vehicle efficiency. This is due to the fact that, on average, more affluent households travel greater distances than poorer ones. On a broader empirical level, an OECD review (OECD, 2008) shows that most studies conclude that consumption behaviour-related environmental pressure increases with household income. This can be observed in relation to waste, recycling, transportation choices and domestic use of energy and water. The only exception is organic food consumption, which could be explained by the aforementioned Roca (2003) idea, i.e. the fact that the affluent can displace the negative effects of their consumption. Yet, analyses by Cox et al. (2012) and the OECD (2008) show empirically that affluent households generate greater environmental pressure individually, but do not analyse if it also happens in marginal terms. However, by calculating the income elasticities of environmental pressures,6 the empirical literature identifies nonlinearity between income and environmental pressures. Indeed, the choice between Situations 1 and 2 in Fig. 1 depends on the value of these elasticities.7 For instance, Liu et al. (2013) and Büchs and Schnepf (2013) identify a concave relationship between household income and consumption behaviour-related GHG emissions.8 Other studies, more numerous, focus on the impact of household levels of consumption on environmental pressures and find similar results, although elasticities are generally higher than in studies analysing the relationship between income and environmental pressures.9 These results are consistent with those from Liu et al. (2013) and Büchs and Schnepf (2013) taken in the context of a decreasing marginal propensity to consume (Dynan et al., 2004). Finally, several analyses distinguish between the effect of the consumption level and the effect of the consumption content on 4 Keynes' fundamental psychological law (Keynes, 1936). Recent studies seem to confirm a decreasing marginal propensity to consume as income rises in a given society (Dynan et al., 2004). 5 See, at the end of the present section, a discussion on how savings potentially affect environmental pressures. 6 Most of the studies identified here utilise input/output methods to take into account direct and indirect environmental pressures due to consumption. 7 Where elasticity is included between 0 and 1, it corresponds to 1. Where it is above 1, it corresponds to 2. Ceteris paribus, the lower the elasticity, the better the inequality is for the environment. 8 Although elasticities range a great deal depending on studies: 0.25 in north China (Liu et al., 2013) and 0.6 in the UK (Büchs and Schnepf, 2013). 9 In the identified studies, expenditure elasticities range from 0.84 to 0.94 for GHG emissions (Kerkhof et al., 2009; Girod and De Haan, 2010; Duarte et al., 2012; Roca, Serrano, 2007), from 0.64 to 1 for energy consumption (Lenzen et al., 2006), from 0.86 to 0.96 for acidifiant gases (Kerkhof et al., 2009; Roca and Serrano, 2007), and from 0.71 to 0.81 for eutrophicant pollutants (Kerkhof et al., 2009; Roca and Serrano, 2007). We can note a 1.42 expenditure elasticity level for smog-forming pollutants in the case of The Netherlands, which shows the necessity to conduct separate analyses for different pollutants (Kerkhof et al., 2009).

193

environmental pressures. This is done by calculating both income and expenditure elasticities of environmental pressures due to consumption. They find that income elasticity is always below the expenditure elasticity, and that income elasticities are lower than 1.10 These results confirm both the decreasing relationship between income and consumption levels (level effect) and between the consumption level and the related environmental pressures (content effect). Nevertheless, the question about the shape of the curve is not conclusive for at least two reasons. First, most of the studies identified focuses on an analysis of GHG emissions and on a limited set of countries, although relationship could be different depending on the pressures and contexts studied. Second, some studies question the hypothesis of a concave relationship. For example, a meta-analysis of 36 studies reveals significant diversity in the income elasticity of households' electricity consumption11 (Espey and Espey, 2004). More crucially, Baiocchi et al. (2010) reveal a more complex relationship between income and GHG emissions using quadratic and cubic terms of the logarithm of income in their econometric equations in the case of British households. Their results show that the income elasticity is a Ushaped function of income, i.e. the elasticity is below 1 for mean income and above 1 at the two extremeties of the income distribution, which suggests the existence of a carbon-intensive lifestyle of the affluent. Overall, this set of facts raises questions about the relationship between individual income and environmental pressure. Thus, a systematic and more detailed review of the analyses studying the non-linearities between individual income and individual environmental pressures would be necessary and could demonstrate the necessity of additional analyses. Nevertheless, even in those cases where Situation 1 is a good representation of reality, inequality might have adverse dynamic and holistic effects on environmental pressure. The literature identified in this section does not take into account dynamic consequences of economic behaviours on environmental pressures, especially through demand-pull technology change and savings behaviours. First, Vona and Patriarca (2011) show that high levels of inequality can prevent the development and diffusion of new environmental technology. Indeed, high inequality may imply that few consumers have access to eco-friendly goods, leading to fewer positive technological externalities and, therefore, a stagnation of prices at high levels. Consequently, the speed and the extent of environmental technology adoption are less important in such a context. Second, the economic behaviour of savings can have an impact on environmental pressures if the savings are utilised for productive investments. This hypothesis is not considered in the studies identified above (Ala-Mantila et al., 2014) and as a result the increasing marginal propensity to save of the affluent could have environmental consequences that remain ignored. Thus, increasing inequality could modify environmental pressures by increasing the level of savings or the way they are utilised. The next sub-section examines holistic approaches used to explain the impact of inequality on the environment through social norms.

2.1.2. Contribution of Holistic Approaches: The Role of Social Norms The aformentioned analyses tend not to account for differences in the social norms that are likely to have been caused by different levels of inequality. This is because such differences are likely to modify the penetration of values such as consumerism, individualism and shorttermism. Variations in the level of dissemination of these norms might lead to variations in the level of individual environmental pressure for a given income level, thereby changing the relationship assumed in 10 The identified studies present a range of expenditure elasticities from 0.6 to 0.802 and income elasticities from 0.34 to 0.607 for GHG emissions (Weber and Matthews, 2008; Ala-Mantila et al., 2014). Studying Ecological Footprint in Australia, Lenzen and Murray (2001) identify an expenditure elasticity of 0.64 and an income elasticity of 0.38. 11 In the case of domestic electricity consumption, Espey and Espey (2004) show that short-term income elasticity varies, depending on which study, from 0.04 to 3.48, with long-term elasticity varying between 0.02 and 5.74.

194

A. Berthe, L. Elie / Ecological Economics 116 (2015) 191–200

Individual environmental pressure

In this channel, the authors' oppositions tend to reflect the way in which each has responded to three questions: 1) Does environmental protection coincide with the interests of a particular social group? 2) Which social groups express a political demand for environmental quality? 3) How is political conflict in this area resolved and what level of environmental protection is adopted in response?

Individual income Fig. 2. Variation of social norms affecting individual environmental pressures. Source: authors.

the preceding sub-section between individual income and individual environmental pressure (Fig. 2). First, Veblen (1899) identifies the role that consumption plays as a social value rather than as a factor of individual utility. This is what is commonly referred to as conspicuous consumption. In a very unequal society, differences in social status are greater and therefore more visible, exacerbating status competition. Such competition notably involves an accentuation of conspicuous consumption.12 By extension, inequality can increase middle and less affluent social groups' desire to copy the dominant social group's modes of consumption, thereby causing greater environmental pressure by increasing the individual consumption level and by changing their consumption content (because the products are consumed on the basis of the social status they confer rather than their environmental quality). Having said that, greater equality can be construed as a factor limiting the social pressure to consume (Wilkinson and Pickett, 2010). Thus, even in consideration of the hypothesis of a concave relationship between environmental pressure and income, the introduction of consumerism might lead to a generalised increase in the level of an individual's income-related environmental pressure curve. Inequality's beneficial effects on the environment should therefore be relativised. Second, Wilkinson and Pickett (2010) along with Boyce (2003, 2007) hypothesise that an unequal society undermines social cohesion and trust in others. As such, it reduces actors' propensity to incorporate social responsibility objectives into their choices. Civic-minded behaviours such as waste sorting and, more generally, protecting the collective environment are more likely in this case to suffer the consequences. Third, Boyce (1994) refers to the possibility that inequality affects consumption due to both wealthier and poorer actors' short-termism. In the case of a highly unequal society, for instance, poorer actors' focus on their short-term material concerns means that they do not have the luxury of considering the long-term impacts of their consumption. Conversely, their wealthier counterparts might find themselves in a less secure position (fear of downwards mobility or political upheaval) and therefore also be more short-term oriented. In sum, the possible impact of inequality on environmental pressures via individual economic behaviours is varied and likely to have both a mechanical effect, depending on marginal pressures reflecting individuals' economic behaviours, and a social effect, through changes in the social norms. Nevertheless, the level of environmental deterioration is not fully explained by individual economic behaviour but is also dependent on environmental policy choices.

2.2. Inequality and Determination of Environmental Policies The influence of inequality on the adoption of environmental policies is also described in a variety of ways in the theoretical literature. 12 There is a counter-argument to this idea, namely the possibility of a conspicuous consumption of high quality environmental goods. For the moment, however, this probably remains the exception.

2.2.1. Does Environmental Protection Coincide With the Interests of a Particular Social Group? The first question is whether a particular social group derives a net benefit or disadvantage from environmental deterioration. According to Roemer (1993), like Boyce (1994, 2003, 2007), affluent social groups are probably those whose consumption activity creates the most environmental degradation and whose income depends most on reducing production costs. In addition, many aspects of the environment do not constitute public goods in the strict sense of the term, meaning that they can be privatised (Boyce, 2003, 2007). In this case, more affluent social groups have the ability to substitute public environmental goods for private environmental goods if the formers were to deteriorate. In the end, affluent social groups are not seen here as suffering from a deterioration of the public environmental goods, reducing their interest in protecting it. Furthermore, the affluent can avoid the impacts of environmental pressures through trade and territorial segregation (Roca, 2003). Conversely, the more deprived social groups are more likely to budget around natural resources. Therefore, they depend more on the conservation of the resources and suffer more from the effects of pollution — as revealed in environmental justice studies (Bullard, 1990; Martinez-Alier, 2002). Aside from Boyce, authors do not tend to hypothesise about individuals' income-related interests but instead about their environmental quality preferences reflected in a demand for environmental protection policies. 2.2.2. Which Social Categories Express a Political Demand for Environment Quality? According to Boyce, interests usually crystallise in political demands. In this analysis, common social groups tend to request environmental protection policies, unlike their privileged counterparts. The systematic nature of this connection is not obvious, however. The mediation between the interests and political demands expressed by a given social group depends on the political ideas of this group, themselves shaped by its interests, short-term material constraints and social norms. As a result, a number of studies do not start by hypothesising that the preferences of social groups are determined by their sole interests but instead brought short-term material constraints into the equation. Indeed, Scruggs (1998) explains wealthy persons' increased preference for the environment using the post-materialist analysis that Inglehart (1990) first developed. The idea is that once short-term material constraints are satisfied, actors tend to adopt post-materialist values, which include environmentalism. Following Ng and Wang (1993), Magnani (2000), for her part, hypothesises that actors' well-being tends to be based on their relative as opposed to absolute income, meaning that they replace demands for pro-growth policies with environmental policies as their income rises in comparison with a society's average income. Similarly, Franzen and Vogl (2013) highlight the fact that inequality causes public opinion to focus on economic and social instead of environmental problems. Lastly, one of the main obstacles to the adoption of economic instruments for regulating environmental problems is, according to Ekins (1999), these instruments' distributive impact. They add to the cost of the environmental goods involved, which account for a greater proportion of poor households' budgets. Their introduction can therefore have a socially regressive fiscal effect, particularly if no compensatory mechanisms are brought in (Serret and Johnstone, 2006). Such considerations highlight the role poor households' budget constraints play in the demand for environmental policies. Wilkinson and Pickett (2010) highlight the role that inequality plays in shaping social norms, and how these norms shape political demands for environmental protection. In this view, significant inequality

A. Berthe, L. Elie / Ecological Economics 116 (2015) 191–200

exacerbates social status competition, thereby reinforcing individualism and consumerism. Environmental policies increasing the price of goods and services or reducing access to them are, therefore, less likely to find political support. Conversely, more egalitarian societies, associated with higher levels of mutual trust and more civic behaviour, are more likely to demand environmental policy. In this latter approach, environmental protection is not associated with any one social group. According to Roemer (1993), dominant groups can impose social norms. In this context, rich persons' interests influence poorer persons' preferences. Boyce (2003) reproduces this idea and integrates it into his classification of forms of power under the heading of value power, i.e. the power of a dominant social group to influence the values of other social groups. The demand for environmentally protective policies is then conflicted by adverse political demand. This confrontation has different outcomes depending on the type of political system in question. 2.2.3. How is Political Conflict in This Area Resolved and What is the Level of Environmental Policy Adopted? Authors tend to envision two main types of political systems in this respect: Power Weighted Social Decision Rules (PWR); and Simple Majority Rules (SMR). Roemer (1993) is the first to use this distinction to account for public good issues. In the first case, political decisions tend to reflect dominant social groups' political demands when inequality is high. In the second situation, political decisions translate demands emanating from the median segment of the electorate. Boyce (1994) and Scruggs (1998) are tenants of the former system; and Magnani (2000) of the latter. According to Boyce (1994), an environmental degradation can be generated when those that benefit from it can impose the choice to degrade the environment on those that suffer the consequences. Apart from cases where a lack of information exists, or where the degradation will only affect future generations, the dominant social groups require the use of power to impose their choices. This power comes in around five distinct forms13 and two are crucial in the present context: the power to put environmental issues to public debate (agenda power) and the power to make decisions (decision power). Here, affluent actors have a significant role in public decision-making as long as they can pay the transaction costs required to influence political decision-making (thus, power and economic inequality are highly correlated). Furthermore, affluent people, as they are small groups in societies characterised by high inequality, can coordinate more easily in order to defend their interests (Olson, 1965). Consequently, the less the income difference between affluent and disadvantaged households, the greater the latter can express an interest in increased environmental protection, thus the greater this protection. However, starting with the same type of political system, Scruggs (1998) comes to the opposite conclusion, alleging that the more affluent social groups are the ones demanding environmental protection. Thus, in this case, higher inequality levels should lead to higher levels of environmental protection. Adopting SMR framework, Magnani (2000) hypothesises that a wealthier median segment, which is close to average income (meaning a less unequal society), should opt for a more stringent level of environmental protection. As for Roemer (1993), who accepts both SMR and PWR as mechanisms, environmental protection is considered as a concern of the poor. Fig. 3 summarises the different channels linking inequality and environmental policies in situations where inequality is on the rise.14 The idea here is that inequality can have contradictory effects on environmental policies depending on the analysis of individual preferences and the political system which enables to resolve the political conflict. We would note that analyses by Boyce (1994) and Magnani (2000) adopt opposite views about both preferences and political systems but come to identical conclusions. 13 Boyce (2003) defines these five forms of power as: Purchasing power, Decision power, Agenda power, Value power and Event power. 14 The sign of this relationship is the opposite in cases where inequality declines.

195

2.3. Overview of Theoretical Structure The literature reviewed suggests a wide variety of theoretical mechanisms, sometimes contradictory, that assesses how variations in inequality might influence the level of environmental pressure. Fig. 4 offers an overview of the relationships dividing the corpus (numbered from 1 to 5). We identify two channels: one related to economic behaviour and the other to political choices. Regarding the first channel, the effect of inequality depends on (1) the shape of the curve connecting individual income and individual environmental pressure; (2) inequality's possible effect on the social norms affecting these pressure. The effect of inequality on environmental protection depends on (3) whether environmental protection coincides with the interests of a social group; (4) whether the group's interest leads to a corresponding political demand which depends on political ideas (themselves adopted to varying degrees by social groups depending on the social norms and the short-term material constraints they are subjected to); (5) how these political demands complement or compete with one another within the framework of a particular political system. Note that these two channels are not entirely disconnected. This is because environmental policies influence the price of environmental pressure. The phenomenon has two consequences. Firstly, if social groups' voting reflects their interests in terms of environmental policy implementation, this vote depends on the level of environmental pressure to which they are being subjected. Conversely, this also means that the adoption of an environmental policy might lead to variations in actors' economic behaviour, hence to changes in the level of environmental pressure incorporated therein. This theoretical structuring elucidates the idea that empirical testing must be sufficiently detailed to enable understanding of which mechanisms are best at explaining the causal relationship between economic inequality and environmental pressure. The problem is that in most cases, the methods used are only partially successful at achieving this objective. 3. Do Existing Empirical Studies Discriminate Between These Different Theoretical Mechanisms? Literature discussing the causal relationships between inequality and environmental deterioration has produced a large number of empirical tests since the late 1990s. After building a table summarising the results of these tests (3.1), we suggest studying possible statistical patterns in order to achieve an initial discrimination between the theoretical mechanisms identified in the second Section (3.2). In the end, because they mask the intermediary stages between inequality and environmental pressures, these tests are unable to validate any particular theoretical explanation. We then propose research directions capable of accounting for the wide variety of theoretical mechanisms in further empirical analyses (3.3). 3.1. Summary of Empirical Studies Table 1 provides an overview of findings from 14 econometric studies focused on the role of economic inequality in environmental pressure or policy outcomes.15 These 14 econometric studies test 15 These analyses have been identified using a three-step process. First, the authors have reviewed all the studies cited by at least one of the articles that we consider as central in the theoretical part of the paper (Boyce, 1994; Heerink et al., 2001; Scruggs, 1998; Magnani, 2000; Wilkinson and Pickett, 2010). Second, the authors have reviewed all the studies citing the five aforementioned papers through Scopus until August 2014. Finally, the authors have inspected the reference sections of the papers selected through the two preceding steps. At each of the three steps, studies were included if they conformed to the following criteria: 1) the paper must be published in a French or English language peer-reviewed scientific journal; 2) the paper must include an econometric study; 3) the endogenous variables used must be measures of environmental pressure or policy; 4) the central exogenous variables mobilised must be measures of economic inequality at regional or national scales.

196

A. Berthe, L. Elie / Ecological Economics 116 (2015) 191–200

Rising inequality, ceteris paribus

Who ask for environmental quality?

Poor people

Affluent people

Which is the rule for social decision?

SMR

PWR

SMR

PWR

What is the impact on environmental policies?

+

-

-

+

Roemer (1993)

Boyce (1994), Roemer (1993)

Magnani (2000)

Scruggs (1998)

Fig. 3. From inequality to environmental policies: splits in the literature. Source: authors.

different endogenous variables that lead to 41 different econometric tests. The analyses generally suggest testing the suitability of a particular theoretical explanation and are carried out using a wide variety of endogenous variables and spatio-temporal scales. The endogenous variables tested here relate to different environmental pressures, ranked in the table from the most global to the most local, and environmental policies. The period of analysis extends

from the 1960s to the late 2000s. The exogenous variables play out at the regional and national levels. 31 of the 41 tests involve international comparisons, with authors generally mobilising a reduced form of the environmental Kuznets curve's characteristic relationship (including average per capita income and the quadratic form thereof) and supplement this with a variable capturing economic inequality, measured by the ratio 80/20 or the Gini index. Other studies

Fig. 4. Channels linking inequality and environmental pressures. Source: authors.

Table 1 Summary of empirical findings from studies analysing economic inequality's effect on environmental pressures and policies.*. Source: Authors. Dependent variable

Geographic scale

Period of analysis

Inequality measure

Relationship inequality/dependent variable

N

Authors

180 countries 17 OECD countries 17 OECD countries

1961–2001 1980 1980

Gini Gini Ratio 80/20

− (2/3) + (1/2) NS

2043 17 17

Torras et al. (2011) Scruggs (1998) Scruggs (1998)

64 countries 83 developing or transition countries United States

1985 1988–2003 1967–2008

Gini Gini Gini

− (1/1) NS + (long-run) + (short-run) (1/1)

64 226 46

Heerink et al. (2001) Clément and Meunié (2010b) Baek and Gweisah (2013)

18–52 cities/19–42 countries Chinese provinces 18–52 cities/19–42 countries 31 countries 83 developing or transition countries 83 developing or transition countries 25–29 countries (stations) 38 countries 25–29 countries (stations) 18–52 cities/19–42 countries

1977–1991 1996–2008 1977–1991 1985 1988–2003 1988–2003 1979–1990 1985 1979–1990 1977–1991

Gini Gini Gini Gini Gini Gini Gini Gini Gini Gini

+ (low income) NS (high income) (1/1) − (1/1) + (low income) − (high income) (1/1) NS NS NS NS NS − (1/1) − (low income) NS (high income) (1/1)

405 402 1188 31 43–310 205 183 38 160 854

Torras and Boyce (1998) Jun et al. (2011) Torras and Boyce (1998) Heerink et al. (2001) Clément and Meunié (2010a) Clément and Meunié (2010b) Scruggs (1998) Heerink et al. (2001) Scruggs (1998) Torras and Boyce (1998)

Chinese provinces 25–29 countries (stations) 287 stations/58 countries 25–29 countries (stations) 288 stations/58 countries 83 developing or transition countries 83 developing or transition countries 82 countries 79 countries 46 countries 33 countries

1996–2008 1979–1990 1977–1991 1979–1990 1977–1991 1988–2003 1988–2003 1994 1994 1985 1985

Gini Gini Gini Gini Gini Gini Gini Gini Gini Gini Gini

− (1/1) − (1/1) + (low income) NS (high income) (1/1) NS NS (low income) − (high income) (1/1) − (2/6) + (1/1) − (low income) NS (high income) (1/1) NS (low income) NS (high income) (1/1) + (1/1) + (1/1)

402 185 1931 148 1484 41–259 182 82 79 46 33

Jun et al. (2011) Scruggs (1998) Torras and Boyce (1998) Scruggs (1998) Torras and Boyce (1998) Clément and Meunié (2010a) Clément and Meunié (2010b) Torras and Boyce (1998) Torras and Boyce (1998) Heerink et al. (2001) Heerink et al. (2001)

16 countries in Subsaharan Africa 16 countries in Subsaharan Africa

1983 1983

Gini Gini

− (1/1) − (1/1)

16 16

Heerink et al. (2001) Heerink et al. (2001)

52 countries 50 countries 45 countries 87 countries 45 American States

1961–1986 1980–1984 2004 (iv: 1989) 2007 (iv: 1990) 1966–2005 (iv: 1969)

Gini Gini Gini Gini Gini

+ (1/1) + (3/3) + (3/3) NS + (3/3)

52 50 45 87 45

Heerink et al. (2001) Holland et al. (2009) Mikkelson et al. (2007) Pandit and Laband (2009) Mikkelson et al. (2007)

19 OECD countries 19 OECD countries OECD countries OECD countries Europe/USA/Canada OECD countries 24 European countries

1980–1991 1980–1991 1980–2000 1980–2000 1999–2004 1980–2000 1996

Gini Ratio 80/20 Gini Gini Gini Gini Gini

NS + then − beyond income threshold (3/6) − (2/3) − (1/3) − (1/1) − (6/9) NS

17–52 16–49 109 109 36 111 24

Magnani (2000) Magnani (2000) Vona and Patriarca (2011) Vona and Patriarca (2011) Vona and Patriarca (2011) Vona and Patriarca (2011) Bimonte (2002)

Environmental performance

A. Berthe, L. Elie / Ecological Economics 116 (2015) 191–200

General performance Ecological footprint Emission score (fertilisers, 4 pollutants) Emission score (fertilisers, 4 pollutants) Climate Per capita CO2 emissions Per capita CO2 emissions Per capita CO2 emissions Air Air pollution (smoke) Industrial gas emissions SO2 Per capita SO2 in urban zones SO2 emissions SO2 Urban concentrations of SO2 Urban concentrations of particulate matter Particle pollutants Heav particles Water Emission of industrial wastewater Dissolved oxygen Dissolved oxygen Faecal coliforms Faecal coliforms Organic water pollution Water pollution Access to safe water (% pop.) Access to sanitation (% pop.) Urban population without sanitation Population without safe water Soil Soil depletion (nitrogen) Soil depletion (phosporus) Biodiversity Disappearance of forests, in pct. Species under threat (plants, vertebrates) Species under threat (plants, vertebrates) Species under threat (plants, vertebrates) Decline of residential bird species Environmental policies Public spending on environmental R&D Public spending on environmental R&D Per capita public green R&D Public green R&D/total R&D Per capita eco-industry revenues Number of environmental PCT/100 inhab. Percentage of protected zones

*Regarding the table's annotations: iv: instrumental variable; +/−/NS: positive/negative/non-significant correlation; (../..): number of models tested as significant/total number of models 197

198

A. Berthe, L. Elie / Ecological Economics 116 (2015) 191–200

add complexity to the functional form of the relationship by including additional control variables.16 3.2. Contributions and Limitations of Statistical Patterns for the Detection of Dominant Theoretical Mechanisms Section 2 identifies and structures the different theoretical channels envisioned by authors in this field in their bid to link economic inequality and environmental pressure. These channels describe explanations postulating a negative or positive relationship between these two variables. From an econometric perspective, identifying statistical patterns reflecting the nature of this relationship helps to detect the dominant explanations. Yet analysis of the empirical studies grouped in Table 1 does not support the idea of a systematic connection between economic inequality and the environment. The findings largely depend on which endogenous variables are mobilised, with no outright trend being identified for the following environmental pressures: CO2 emissions, air pollution and water.17 On the other hand, a positive relationship tends to emerge between inequality and biodiversity loss. Similarly, environmental policy levels seem to increase as inequality decreases. Thus, an initial contribution of these empirical studies is to show that the negative effects of inequality on biodiversity and environmental policy seem stronger than the positive effects. In that case, the theoretical explanations suggested by Boyce (1994), Magnani (2000) or Wilkinson and Pickett (2010) dominate the ones suggested by Scruggs (1998) and Heerink et al. (2001). In particular, the identification of the negative effect of the inequality on environmental policy raises questions about the empirical strength of the idea, suggested by Roemer (1993) and Scruggs (1998), that inequality strengthens environmental protection (Fig. 3). A second contribution of these empirical studies, and of the wide variety of results found, is to indicate that the effects of inequality differ depending on which environmental pressure is being studied, meaning that reality is more complex than usually portrayed in theoretical literature. Existing studies tend to provide few explanations for variations in findings between different environmental fields since they tend to look at the environment in its globality.18 According to Heerink et al. (2001), the negative impact of inequality on environmental protection is larger in the case of local problems. Working within a framework formulated by Boyce (2008), Clément and Meunié (2010a) come to the conclusion that this negative effect on environmental pressure is greater where pressures with localised effects are concerned, given affluent populations' ability to isolate themselves through territorial segregation. This might explain the convergence of the findings formulated in biodiversity literature. In addition, the negative impact of inequality on biodiversity and environmental policies might also be explained by its more direct influence on these dimensions compared to the others. Indeed, a wider variety of determinants tend to affect environmental deterioration if they are more global in nature. For instance, global environmental pressures are 16 These variables can be classified into four types, with the first three applying to all of the types of endogenous variables. The categories include: biophysical and demographic pressures: urbanisation (Jun et al., 2011), population density (Scruggs, 1998; Torras and Boyce, 1998), population (Mikkelson et al., 2007; Pandit and Laband, 2009); human capital, level of information and technology: number of years of education for each inhabitant (Jun et al., 2011), number of newspapers sold annually (Bimonte, 2002), R&D/GDP spending (Jun et al., 2011); indicators for level of freedom and concentration of power (Clément and Meunié, 2010a,b; Torras and Boyce, 1998; Torras et al., 2011); specific influences on particular variables: measures of biodiversity to explain the pressures faced at this level (Mikkelson et al., 2007, Holland et al., 2009; Pandit and Laband, 2009), energy consumption explaining CO2 emissions (Baek and Gweisah, 2013), industrial output explaining industrial pollutant waste (Jun et al., 2011), type of geographic area (Torras and Boyce, 1998) explaining air pollution, green R&D explaining number of green technology patents (Vona and Patriarca, 2011). 17 We note one study of soil depletion. Based on few observations and focusing on a very particular context (sub-Saharan Africa), we cannot derive any particular conclusions from it regarding this type of environmental pressure. 18 Which may seem paradoxical given that empirical tests almost never imply summative environmental pressure indicators (aside from analyses by Scruggs, 1998, and Torras et al., 2011).

affected by the local system of production, by the pressures coming from neighbouring countries and by climatic and geographic factors. On the other hand, local environmental pressures and policies are associated with a narrower and more geographically limited range of determinants, intimating that inequality's role is possibly paramount in this case. This can be tested by analysing inequality's effects on variables situated further upstream in the causal chain featured in Fig. 4. For instance, Franzen and Vogl (2013) develop a study assessing inequality's effect on environmental attitudes, notably through the use of multi-level models. The aforementioned findings should be assessed carefully given the statistical limits that sometimes characterise studies whose samples are so small that they raise questions about the findings' reliability. This is particularly true for studies by Scruggs (1998), Heerink et al. (2001), Bimonte (2002), and Magnani (2000). The limits might be explained by a relative lack of perspective at present regarding environmental data, especially at the time the studies were being conducted. In addition to this initial shortcoming, Wilkinson and Pickett (2010) resort to an approach using descriptive statistics that amongst other problems do not control for other variables' influence over their findings. Lastly and more generally, the control variables used in this body of literature are highly heterogeneous, creating problems in terms of the findings' comparability (see footnote 16). In general, empirical studies in this field do not try to demonstrate the suitability of one particular theoretical channel compared to other possibilities but focus solely on identifying the general direction of causality between inequality and environmental deterioration. This method makes it possible to detect several dominated theoretical explanations but does not suffice to ascertain the best way of describing reality. In the present sub-section, we saw that the empirical findings, whenever these are significant, tend to validate the hypothesis that inequality drives environmental deterioration, as conveyed by Boyce (1994), Magnani (2000) and Wilkinson and Pickett (2010). At the same time, these authors justify the hypothesis using diverse and potentially rival theoretical mechanisms. Boyce and Magnani adopt opposite reasons to explain actors' preferences and political systems. Wilkinson and Pickett's approach is situated in a different register, with inequality being conceptualised as a determinant of social norms. As such, they make no hypotheses regarding the political system and assume instead homogeneous individual preferences irrespective of the actors' position in the social structure (see Section 2). The question here is which of these reasonings is best at explaining reality. 3.3. Possibilities for Directly Testing These Theoretical Mechanisms The theoretical mechanisms behind each of the three aforementioned reasonings should be empirically tested directly. An initial way of doing this consists of testing the interactions between inequality and other variables of interest in order to reveal inequality's potentially differentiated effect depending on the level of each variable of interest. For instance, integrating an interaction term between inequality and political system makes it possible to ascertain whether inequality's effects vary depending on the type of political system. With this in mind, and even though their findings do not reveal any dominant trend, Torras and Boyce (1998) propose isolating the effects of inequality according to the level of wealth in a particular country.19 Similarly, Clément and Meunié (2010a) use a discrete index of political rights to isolate the effect of inequality according to national political regimes. Their findings do not demonstrate with any certainty that the impact of inequality on environmental pressure levels is dependent on the political regime. The method suggested by Boyce et al. (1999) offers another way to identify the most suitable theoretical channel. This study focuses on the connection between economic inequality, power and the determinants of health in the United States. The authors suggest a recursive 19 Here Torras and Boyce (1998) hypothesise that power inequality is higher in lowincome countries.

A. Berthe, L. Elie / Ecological Economics 116 (2015) 191–200

procedure enabling an empirical examination of each stage of the theoretical model. This procedure consist in testing four successive econometric equations: 1) determinants of the distribution of power, including economic inequality; 2) determinants of environmental policy including the distribution of power; 3) determinants of environmental stresses, including environment policy; 4) determinants of health, including environmental stress and the distribution of power. Note that the procedure applied here makes it possible to view the distribution of power as endogenous, contrary to the methods applied by Torras and Boyce (1998) and Clément and Meunié (2010a). In our opinion, however, the study has several limitations. Firstly, the analysis covers a very small number of observations from crosssectional data and only relates to U.S. states, undermining the findings' generalisability. Secondly, the authors employ a questionable distribution of power indicator. This is because the criteria distinguishing between the distribution of power determinants and the variables comprising this indicator are not explained. Some variables comprising the aforementioned index might be considered as determinants thereof (e.g. the level of educational attainment). The opaque way in which the power index is constructed and the choice of its constituent variables make it difficult to isolate the effect of economic inequality on environment policy. The mobilisation of other power indicators might help to verify the model's reliability. Thirdly, the article only envisions a political channel. Fourthly, the authors do not test the hypothesis that poorer persons have a net interest in preserving the environment, and that this interest is mechanically transformed into corresponding political demands (cf. Section 2.2). Nevertheless, the study does have the merit of demonstrating that it is probably necessary to produce recursive empirical studies to deepen understanding in this field. 4. Conclusion and Perspectives The paper reviews theoretical explanations linking economic inequality and environmental deterioration. We demonstrate that some rival explanations come to the same conclusions regarding the nature of this relationship. Two channels, economic and political, are identified, each containing their own internal diversity. The first relates to aggregate environmental deterioration due to the economic behaviour of individuals, the second to the implementation of public policy aimed at protecting the environment. The development of an analytical framework comparing existing studies in this field enhances understanding of the phenomenon by identifying similarities and dissimilarities between the explanations of the relationship between inequality and environmental pressure. In addition, this framework allows us to show that most empirical studies do not adopt a methodological approach enabling an assessment of different theoretical pathways' suitability but instead only tested a global hypothesis about the influence of inequality on environmental pressures or policy. From this perspective, two areas seem to confirm the hypothesis that inequality causes environmental deterioration: environmental policy and biodiversity loss. Lastly, the shortcomings of empirical analyses might, in fact, be overcome by the adoption of a recursive approach. Moreover, data that is now available could help to overcome the constraints that different studies have faced in this field. To further advance understanding of the mechanisms at work in this area of research, there is a need to compile and supplement existing analyses of various hypotheses about the different channels evoked here. This could become the objective of the future research programme whose purpose would be to answer the following five questions: From an economic channel perspective: 1. What is the relationship between individual income and individual environmental pressure? 2. How does inequality affect the aggregate environmental pressure related to economic behaviours, particularly through the variation of social norms?

199

From a political channel perspective: 3. What are different social groups' overall interests in terms of protecting the environment? 4. How have these interests been transformed into political positions? 5. Are these positions reflected in political decisions?

Answering these questions should enable a better understanding of the extent to which (and under what conditions) the social crisis fosters the environmental crisis. This is an important challenge, since it could appear that policies aimed at reducing inequality are a prerequisite for the adoption of environmental friendly policies and individual behaviours. Acknowledgment The authors would like to thank colleagues, and especially Nathalie Gaussier, as well as the four anonymous reviewers and the associate editor for their insightful comments on earlier versions of this paper. The research leading to these results was funded under the European Community's Seventh Framework Programme (FP7/2007–2011) under grant agreement n225349 (ICaTSEM project) and under the MOVIDA program of the French Ministry of Ecology, Sustainable Development and Energy. All errors and omissions are the responsibility of the authors. References Ala-Mantila, S., Heinonen, J., Junnila, S., 2014. Relationship between urbanization, direct and indirect greenhouse gas emissions, and expenditures: a multivariate analysis. Ecol. Econ. 104, 129–139. http://dx.doi.org/10.1016/j.ecolecon.2014.04.019. Baek, J., Gweisah, G., 2013. Does income inequality harm the environment? Empirical evidence from the United States. Energy Policy 62, 1434–1437. http://dx.doi.org/10. 1016/j.enpol.2013.07.097. Baiocchi, G., Minx, J., Hubacek, K., 2010. The impact of social factors and consumer behavior on carbon dioxide emissions in the United Kingdom. J. Ind. Ecol. 14, 50–72. http:// dx.doi.org/10.1111/j.1530-9290.2009.00216.x. Baland, J.-M., Bardhan, P.K., Bowles, S., 2007. Inequality, Collective Action, and Environmental Sustainability. Princeton University Press, Russell Sage Foundation, Princeton, NJ, New York. Bimonte, S., 2002. Information access, income distribution, and the Environmental Kuznets Curve. Ecol. Econ. 41, 145–156. http://dx.doi.org/10.1016/S0921-8009(02)00022-8. Boyce, J.K., 1994. Inequality as a cause of environmental degradation. Ecol. Econ. 11, 169–178. http://dx.doi.org/10.1016/0921-8009(94)90198-8. Boyce, J.K., 2003. Inequality and environmental protection. Working Paper 52. Political Economy Research Institute, University of Massachusetts, Amherst. Boyce, J.K., 2007. Inequality and environmental protection. In: Baland, J.-M., Bardhan, P.K., Bowles, S. (Eds.), Inequality, Collective Action, and Environmental Sustainability. Princeton University Press, Russell Sage Foundation, Princeton, NJ, New York, pp. 314–348. Boyce, J.K., 2008. Is inequality bad for the environment? Res. Soc. Probl. Public Policy 15, 267–288. Boyce, J.K., Klemer, A.R., Templet, P.H., Willis, C.E., 1999. Power distribution, the environment, and public health: a state-level analysis. Ecol. Econ. 29, 127–140. http://dx.doi. org/10.1016/S0921-8009(98)00056-1. Büchs, M., Schnepf, S.V., 2013. Who emits most? Associations between socio-economic factors and UK households' home energy, transport, indirect and total CO2 emissions. Ecol. Econ. 90, 114–123. http://dx.doi.org/10.1016/j.ecolecon.2013.03.007. Bullard, R.D., 1990. Dumping in Dixie: Race, Class, and Environmental Quality. Westview Press, Boulder, CO. Clément, M., Meunié, A., 2010a. Is inequality harmful for the environment? An empirical analysis applied to developing and transition countries. Rev. Soc. Econ. 68, 413–445. http://dx.doi.org/10.1080/09644016.2013.755391. Clément, M., Meunié, A., 2010b. Inégalités, développement et qualité de l'environnement : mécanismes et application empirique. Mondes Dév. n° 151, 67–82. http://dx.doi. org/10.3917/med.151.0067. Cox, A., Collins, A., Woods, L., Ferguson, N., 2012. A household level environmental Kuznets curve? Some recent evidence on transport emissions and income. Econ. Lett. 115, 187–189. http://dx.doi.org/10.1016/j.econlet.2011.12.014. Csutora, M., 2012. One more awareness gap? The behaviour–impact gap problem. J. Consum. Policy 35, 145–163. http://dx.doi.org/10.1007/s10603-012-9187-8. Duarte, R., Mainar, A., Sánchez-Chóliz, J., 2012. Social groups and CO2 emissions in Spanish households. Energy Policy 44, 441–450. http://dx.doi.org/10.1016/j.enpol.2012.02. 020. Dynan, K.E., Skinner, J., Zeldes, S.P., 2004. Do the rich save more? J. Polit. Econ. 112, 397–444. http://dx.doi.org/10.3386/w7906. Ekins, P., 1999. European environmental taxes and charges: recent experience, issues and trends. Ecol. Econ. 31, 39–62. http://dx.doi.org/10.1016/S0921-8009(99)00051-8.

200

A. Berthe, L. Elie / Ecological Economics 116 (2015) 191–200

Espey, J.A., Espey, M., 2004. Turning on the lights: a meta-analysis of residential electricity demand elasticities. J. Agric. Appl. Econ. 36, 61–81. Fleurbaey, M., Kartha, S., Bolwig, S., Chee, Y.L., Chen, Y., Corbera, E., Lecocq, F., Lutz, W., Muylaert, M.S., Norgaard, R.B., Okereke, C., Sagar, A.D., 2014. Sustainable development and equity. In: Edenhofer, O., Pichs-Madruga, R., Sokona, Y., Farahani, E., Kadner, S., Seyboth, K., Adler, A., Baum, I., Brunner, S., Eickemeier, P., Kriemann, B., Savolainen, J., Schlömer, S., von Stechow, C., Zwickel, T., Minx, J.C. (Eds.), Climate Change 2014: Mitigation of Climate Change. Contribution of Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Franzen, A., Vogl, D., 2013. Acquiescence and the willingness to pay for environmental protection: a comparison of the ISSP, WVS, and EVS. Soc. Sci. Q. 94, 637–659. http://dx.doi.org/10.1111/j.1540-6237.2012.00903.x. Girod, B., De Haan, P., 2010. More or better? A model for changes in household greenhouse gas emissions due to higher income. J. Ind. Ecol. 14, 31–49. http://dx.doi.org/ 10.1111/j.1530-9290.2009.00202.x. Heerink, N., Folmer, H., 1994. Income distribution and the fulfillment of basic needs: theory and empirical evidence. J. Policy Model 16, 625–652. http://dx.doi.org/10.1016/ 0161-8938(94)90012-4. Heerink, N., Mulatu, A., Bulte, E., 2001. Income inequality and the environment: aggregation bias in environmental Kuznets curves. Ecol. Econ. 38, 359–367. http://dx.doi.org/ 10.1016/S0921-8009(01)00171-9. Holland, T.G., Peterson, G.D., Gonzalez, A., 2009. A cross-national analysis of how economic inequality predicts biodiversity loss. Conserv. Biol. 23, 1304–1313. http://dx.doi. org/10.1111/j.1523-1739.2009.01207.x. Inglehart, R., 1990. Culture Shift in Advanced Industrial Society. Princeton University Press, Princeton. IPCC, 2013. Summary for policymakers. In: Stocker, T.F., Qin, D., Plattner, G.-K., Tignor, M., Allen, S.K., Boschung, J., Nauels, A., Xia, Y., Bex, V., Midgley, P.M. (Eds.), Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Jun, Y., Zhong-kui, Y., Peng-fei, S., 2011. Income distribution, human capital and environmental quality: empirical study in China. Energy Procedia 5, 1689–1696. http://dx. doi.org/10.1016/j.egypro.2011.03.288. Kahn, M.E., 1998. A household level environmental Kuznets curve. Econ. Lett. 59, 269–273. http://dx.doi.org/10.1016/S0165-1765(98)00035-4. Kerkhof, A.C., Nonhebel, S., Moll, H.C., 2009. Relating the environmental impact of consumption to household expenditures: an input–output analysis. Ecol. Econ. 68, 1160–1170. http://dx.doi.org/10.1016/j.ecolecon.2008.08.004. Keynes, J.M., 1936. The General Theory of Employment, Interest and Money. MacMillan & Co., London. Laurent, E., 2011. Issues in environmental justice within the European Union. Ecol. Econ. 70, 1846–1853. http://dx.doi.org/10.1016/j.ecolecon.2011.06.025. Lenzen, M., Murray, S.A., 2001. A modified ecological footprint method and its application to Australia. Ecol. Econ. 37, 229–255. http://dx.doi.org/10.1016/S0921-8009(00)00275-5. Lenzen, M., Wier, M., Cohen, C., Hayami, H., Pachauri, S., Schaeffer, R., 2006. A comparative multivariate analysis of household energy requirements in Australia, Brazil, Denmark, India and Japan. Energy 31, 181–207. http://dx.doi.org/10.1016/j.energy.2005.01.009. Liu, W., Spaargaren, G., Heerink, N., Mol, A.P.J., Wang, C., 2013. Energy consumption practices of rural households in north China: basic characteristics and potential for low carbon development. Energy Policy 55, 128–138. http://dx.doi.org/10.1016/j.enpol. 2012.11.031 (Special section: Long Run Transitions to Sustainable Economic Structures in the European Union and Beyond). Magnani, E., 2000. The Environmental Kuznets Curve, environmental protection policy and income distribution. Ecol. Econ. 32, 431–443. http://dx.doi.org/10.1016/S09218009(99)00115-9. Martinez-Alier, J., 2002. The Environmentalism of the Poor: A Study of Ecological Conflicts and Valuation. Edward Elgar Publishing.

Mikkelson, G.M., Gonzalez, A., Peterson, G.D., 2007. Economic inequality predicts biodiversity loss. Plos One 2, e444. http://dx.doi.org/10.1371/journal.pone.0000444. Millenium Ecosystem Assessment, 2005. Ecosystems and Human Well-being: General Synthesis. Island Press, Washington, DC. Ng, Y.-K., Wang, J., 1993. Relative income, aspiration, environmental quality, individual and political myopia: why may the rat-race for material growth be welfare-reducing? Math. Soc. Sci. 26, 3–23. http://dx.doi.org/10.1016/0165-4896(93)90008-7. OECD, 2008. Household Behaviour and the Environment: Reviewing the Evidence. OECD publishing, Paris. OECD, 2011. Divided We Stand: Why Inequality Keeps Rising. OECD publishing, Paris. OECD, 2014. Rising Inequality: Youth and Poor Fall Further Behind. OECD publishing, Paris. Olson, M.J., 1965. The Logic of Collective Action. Harvard University Press, Cambridge. Olsson, L., Opondo, M., Tschakert, P., Agrawal, A., Eriksen, S.H., Ma, S., Perch, L.N., Zakieldeen, S.A., 2014. Livelihoods and poverty. In: Field, C.B., Barros, V.R., Dokken, D.J., Mach, K.J., Mastrandrea, M.D., Bilir, T.E., Chatterjee, M., Ebi, K.L., Estrada, Y.O., Genova, R.C., Girma, B., Kissel, E.S., Levy, A.N., MacCracken, S., Mastrandrea, P.R., White, L.L. (Eds.), Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part A: Global and Sectoral Aspects. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp. 793–832. Pandit, R., Laband, D.N., 2009. Economic well-being, the distribution of income and species imperilment. Biodivers. Conserv. 18, 3219–3233. http://dx.doi.org/10.1007/ s10531-009-9638-y. Piketty, T., 2014. Capital in the Twenty-First Century. Harvard University Press, Cambridge, MA. Roca, J., 2003. Do individual preferences explain the Environmental Kuznets curve? Ecol. Econ. 45, 3–10. http://dx.doi.org/10.1016/S0921-8009(02)00263-X. Roca, J., Serrano, M., 2007. Income growth and atmospheric pollution in Spain: an input– output approach. Ecol. Econ. 63, 230–242. http://dx.doi.org/10.1016/j.ecolecon.2006. 11.012. Roemer, J.E., 1993. Would economic democracy decrease the amount of public bads? Scand. J. Econ. 95, 227–238. http://dx.doi.org/10.2307/3440408. Scruggs, L.A., 1998. Political and economic inequality and the environment. Ecol. Econ. 26, 259–275. http://dx.doi.org/10.1016/S0921-8009(97)00118-3. Serret, Y., Johnstone, N. (Eds.), 2006. The Distributional Effects of Environmental Policy. Organisation for Economic Co-operation and Development, Paris. Steffen, W., Grinevald, J., Crutzen, P., McNeill, J., 2011. The Anthropocene: conceptual and historical perspectives. Phil. Trans. R. Soc. A 369, 842–867. http://dx.doi.org/10.1098/ rsta.2010.0327. Torras, M., Boyce, J.K., 1998. Income, inequality, and pollution: a reassessment of the environmental Kuznets Curve. Ecol. Econ. 25, 147–160. http://dx.doi.org/10.1016/ S0921-8009(97)00177-8. Torras, M., Moskalev, S.A., Hazy, J.K., Ashley, A.S., 2011. An econometric analysis of ecological footprint determinants: implications for sustainability. Int. J. Sustain. Soc. 3, 258–275. http://dx.doi.org/10.1504/IJSSOC.2011.041267. Veblen, T., 1899. The Theory of the Leisure Class: An Economic Study in the Evolution of Institutions. The Macmillan Company, New-York. Vona, F., Patriarca, F., 2011. Income inequality and the development of environmental technologies. Ecol. Econ. 70, 2201–2213. http://dx.doi.org/10.1016/j.ecolecon.2011. 06.027. Weber, C.L., Matthews, H.S., 2008. Quantifying the global and distributional aspects of American household carbon footprint. Ecol. Econ. 66, 379–391. http://dx.doi.org/10. 1016/j.ecolecon.2007.09.021. Wilkinson, R., Pickett, K., 2010. The Spirit Level: Why Equality is Better for Everyone. Penguin books, London.