Who bears the brunt? Distributional effects of climate change mitigation policies

TL;DR: This article assess the claim that climate change policies have regressive effects by comparing different types of mitigation policies and argue that many of these are indeed likely to have a regressive distributional implications, but that there are several policy options to counteract these effects.

Abstract: Climate change scholars generally urge that CO2 emissions need to be cut rapidly if we are to avoid dangerous risks of climate change. However, climate change mitigation policies are widely perceived to have regressive effects – that is, putting a higher financial burden as a proportion of household income on poor than on rich households. This is one of several major barriers to the adoption of effective mitigation policies. They would also have considerable social justice implications requiring significant welfare state responses. We assess the claim that climate change policies have regressive effects by comparing different types of mitigation policies. We will argue that many of these are indeed likely to have regressive distributional implications but that there are several policy options to counteract regressive effects.

Summary

Introduction

• There is long-standing interest in the distributional effects of climate change mitigation policies in the literatures on climate and energy policy, and economics.
• The authors shall argue that the distributional effects of climate change mitigation policies will depend to a large extent on their exact design and on the area of emissions that they target (for example, emissions from home energy use as opposed to emissions from transport).
• It needs to be stressed here that so-called ‘market-based’ climate change mitigation policies which define an overall cap on emissions for the economy (on any scale, e.g. nationally or globally) impose a strict ‘straitjacket’ on the market, in effect creating a highly regulated type of market.
• The second section summarizes the findings regarding the distributional effects of different types of mitigation policies.

Types of mitigation policies

• The two main alternative market-based tools are carbon taxes and cap and trade schemes which both put a price on energy use or greenhouse gas emissions in order to include ‘negative externalities’ resulting from economic activities (e.g. pollution) in the pricing mechanism.
• Mitigation instruments can apply at different levels of economic activity: up-, mid- or downstream in the chain of production running from natural resource extraction down to the end user.
• Within cap and trade schemes, several options exist as to how emission permits are allocated to the participants – all of which have different distributional impacts.
• This approach is called ‘grandfathering’ in the literature.

Distributional effects

• This expectation also carries through to various types of cap and trade schemes.
• There is a general consensus that downstream taxses on home energy use are regressive if the revenue from those taxes or charges is not redistributed to the citizens (Baranzini et al., 2000; Barker and Köhler, 1998; Dresner and Ekins, 2006).
• This means that low income households spend much higher shares of their income on home energy than richer households (Dresner and Ekins, 2006; Druckman and Jackson, 2008; Wier et al., 2005).
• Several authors argue that environmental policies will be more effective if distributional aspects are dealt with separately (e.g. Johnstone and Serret, 2006: 5; Metcalf and Weisbach, 2009).
• Studies using data from the United States report regressive effects of motoring taxes even if the whole population is included, because of high car dependency (Wadud et al., 2008; Walls and Hanson, 1999). at Freie Universitaet Berlin on April 24, 2015csp.sagepub.comDownloaded from.

Impact of methodological choices

• The distributional effects of climate change mitigation policies presented in different studies also depend on several methodological choices, for example the type of measurement of distributional effects and whether or not behavioural responses are included in the model.
• Behavioural models also estimate the distribution of financial burdens but include estimates of changes in people’s consumption behaviour in response to price increases.
• In a Cap and Share or Cap and Dividend scheme, any individual who consumes less than the capped level of emissions will financially gain from the rebate/revenue (AEA and Cambridge Econometrics, 2008; Boyce and Riddle, 2007).
• If poorer households gain more in absolute terms than richer households, the distributional effect will be strongly progressive in relative terms.

Socio-demographic factors

• The previous section was confined to a consideration of the distributional effects of mitigation policies across income bands.
• An analysis of distributional effects that only focuses on income groups disguises considerable variations of the distribution of burdens and benefits within income groups, because of such additional factors.
• The trend towards smaller household sizes is therefore a driving factor behind increasing per capita emissions in many rich countries, particularly if indirect emissions are included (Druckman and Jackson, 2008: 3184).
• So far, relatively few studies use multivariate regression analysis to examine the relative impact of different socio-economic factors (Baiocchi et al., 2010; Brand and Preston, 2010; DEFRA, 2008a; Lenzen et al., 2006; Weber and Matthews, 2008).

Conclusions

• The literature exhibits a broad consensus that climate change mitigation policies usually have regressive effects in the absence of compensatory revenue recycling.
• Study results also depend on methodological choices, such as the measure that is used to examine distributional effects and whether or not behavioural responses are included in the model.
• Policy options of mitigating those effects need to be further investigated.
• Of course, revenue can only arise from carbon reduction policies as long as there are emissions left that can be reduced – once net-emissions have reached zero, the revenue stream from those policies will run dry.

Figures

Figure 1 Estimated distributional effect of a carbon tax on home energy use of UK households Based on Dresner and Ekins (2006: 55). In this model, the carbon tax is modelled as 0.43p/kWh for electricity and 0.19p/kWh for gas.

Full text

285
© The Author(s), 2011. Reprints and permissions: http://www.sagepub.co.uk/journalsPermissions.nav
Critical Social Policy, 0261-0183 101; Vol. 31(2): 285–307; 396036 10.1177/0261018310396036
http://csp.sagepub.com
 M I L E N A B Ü C H S
University of Southampton

N I C H O L A S B A R D S L E Y
University of Reading

S E B A S T I A N D U W E
Free University Berlin
Who bears the brunt? Distributional effects of climate
change mitigation policies
Abstract
Climate change scholars generally urge that CO
2
emissions need to be cut
rapidly if we are to avoid dangerous risks of climate change. However,
climate change mitigation policies are widely perceived to have regres-
sive effects – that is, putting a higher financial burden as a proportion of
household income on poor than on rich households. This is one of several
major barriers to the adoption of effective mitigation policies. They would
also have considerable social justice implications requiring significant
welfare state responses. We assess the claim that climate change policies
have regressive effects by comparing different types of mitigation poli-
cies. We will argue that many of these are indeed likely to have regressive
distributional implications but that there are several policy options to
counteract regressive effects.
Key words: emissions, environment, fairness, inequality, taxation
Introduction
There is long-standing interest in the distributional effects of climate
change mitigation policies in the literatures on climate and energy
policy, and economics. Hitherto, though, this debate has not impacted
much on mainstream social policy debates (however, see Gough et al.,
2008). This needs to change because climate policy will have an imme-
diate impact on social policy through various channels, in particular
at Freie Universitaet Berlin on April 24, 2015csp.sagepub.comDownloaded from

286 C R I T I C A L S O C I A L P O L I C Y 3 1 ( 2 )
redistribution effects. Those effects are important for two reasons. First
of all, they raise questions about fairness. For instance, how do the bur-
dens of climate change mitigation policy on citizens relate to household
income? Are such burdens proportional to the impact on the environ-
ment of different lifestyles? And how can climate change policies be
designed such that unjust distributional effects are avoided? Secondly,
the public acceptability of such policies will influence the likelihood
that governments adopt them. Little empirical research has been done
in this area so far but it seems plausible that the (perceived) fairness
of those policies will play an important role for public acceptability
(Bristow et al., 2010). The potential power of adverse public reactions
to policies is evident in the UK, for example, in the abandonment of the
poll tax and more recently the fuel tax escalator. This does not neces-
sarily imply that governments will adopt policies which are fairest and/
or most accepted by the broader public. Existing power asymmetries
in society and politics mean that governments frequently respond to
considerable pressure from corporate interests to adopt designs favour-
able to them as evidenced in relation to the European Union Emissions
Trading scheme (Michaelowa and Butzengeiger, 2005). However, the
fairness of climate change mitigation policies remains important from
a normative point of view and their public acceptability will be one
of several factors that governments take into account in the course
of policy-making as, for example, DEFRA’s recent research into the
public acceptability of a Personal Carbon Trading scheme illustrates
(DEFRA, 2008b).
This article
1
analyses arguments and evidence concerning regarding
the distributional impacts of a range of climate change mitigation poli-
cies. We will conclude that a range of currently debated and applied
economic policy tools to mitigate climate change are likely to have
regressive effects, meaning that households in lower income brackets
bear a (considerably) higher burden of the cost as a percentage of their
income than those in higher brackets.
However, we shall argue that the distributional effects of climate
change mitigation policies will depend to a large extent on their exact
design and on the area of emissions that they target (for example, emis-
sions from home energy use as opposed to emissions from transport).
As we are particularly interested in the question of how regressive
effects of emissions reduction policies can be avoided for reasons that
we explained above, we will compare different policy options that can
be used to mitigate or at least limit those effects.
at Freie Universitaet Berlin on April 24, 2015csp.sagepub.comDownloaded from

B Ü C H S E T A L . — W H O B E A R S T H E B R U N T ? 287
This article focuses on ‘economic instruments’ rather than ‘pure’
regulation because they are generally regarded as more ‘efficient’ in
the literature on climate change mitigation (see section below). How-
ever, it needs to be stressed here that so-called ‘market-based’ climate
change mitigation policies which define an overall cap on emissions for
the economy (on any scale, e.g. nationally or globally) impose a strict
‘straitjacket’ on the market, in effect creating a highly regulated type of
market. This does not mean that there are no ethical concerns in rela-
tion to economic emission reduction instruments. One common point
of contention is that economic instruments put a price on a commons –
‘propertizing’ the earth’s atmosphere. Another is that some rich people
will be able to maintain their high carbon lifestyles as they are able to
pay a higher price for their consumption. However, within schemes that
set a strict overall cap on emissions this will not be possible for the gen-
erality of the rich. This is because the bulk of emission reductions would
need to be based on a cutback of their consumption, which is currently
responsible for the high emission levels of rich countries. In short, we
acknowledge considerable ethical concerns about, and potential short-
comings of, market-based mitigation policies. However, we find it
plausible that they form a part of any viable plan to avoid dangerous
climate change, since emissions cannot be regulated away overnight.
The first section of this article provides a brief overview of different
types of economic mitigation policies. The second section summarizes
the findings regarding the distributional effects of different types of
mitigation policies. Here we will focus on emission and energy taxes
and different types of cap and trade schemes. We will also discuss the
impact that different methodological choices have on study results. The
third section describes the different policy options for using revenues
stemming from mitigation policies, and their distributional effects.
The fourth section discusses the distributional impact of mitigation
policies across a wider range of socio-economic factors. The conclusion
summarizes the findings and identifies points for further discussion.
Types of mitigation policies
Mitigation policies can take a variety of forms, with different advan-
tages and disadvantages [...] from environmental, economic and distri-
butional perspectives. Generally, one can distinguish regulation, taxes
or charges, subsidies and trading schemes (Helm, 2005; OECD, 1994).
at Freie Universitaet Berlin on April 24, 2015csp.sagepub.comDownloaded from

288 C R I T I C A L S O C I A L P O L I C Y 3 1 ( 2 )
Regulatory instruments set legally binding emission, energy use or
efficiency targets which all addressees have to adhere to – or otherwise
face financial and possibly penal penalties. It is frequently argued that
regulation is less cost-efficient than market-based instruments because
it does not reflect differing abatement costs of the various polluters.
That is, it does not exploit the fact that some entities will be able to
make large emissions savings at relatively low cost whilst others will
find it very difficult and costly to adjust to lower emission targets.
This is one reason why economic instruments – which include vari-
ous regulatory aspects – have gained prominence in the literature on
environmental policy.
2
These instruments essentially use price incen-
tives and market mechanisms to generate environmentally beneficial
behaviour.
The two main alternative market-based tools are carbon taxes and
cap and trade schemes which both put a price on energy use or green-
house gas emissions in order to include ‘negative externalities’ result-
ing from economic activities (e.g. pollution) in the pricing mechanism.
In theory, both schemes create financial incentives to switch to low-
impact lifestyles and production methods. The key difference between
the two is that environmental taxes – in economics named ‘Pigovian
taxes’ – establish a fixed price for environmentally damaging behav-
iours, whereas cap and trade fixes the amount of pollution by establish-
ing an emissions cap.
Environmental taxes are therefore believed to enhance economic
security, but at the expense of environmental security, because the
environmental improvement depends on actors’ responses to the tax.
The opposite holds true for cap and trade schemes which fix the maxi-
mum amount of pollution but let the market set the price. Consider
petrol for cars, for example. Both instruments would influence the price
of petrol. In the case of a tax, a fixed charge is added to the ‘regular’
price of every litre of petrol, whilst the additional cost under a cap and
trade scheme depends on the emissions cap and the overall demand for
petrol. With the cap it is certain that only the amount of petrol associ-
ated with the allowed pollution target is sold in the economy, whilst
the amount of petrol sold under a fuel or carbon tax may still increase
if consumers are prepared to pay a higher price.
Mitigation instruments can apply at different levels of economic
activity: up-, mid- or downstream in the chain of production running
from natural resource extraction down to the end user. An upstream
scheme would apply a tax or emissions cap to the production or import
at Freie Universitaet Berlin on April 24, 2015csp.sagepub.comDownloaded from

B Ü C H S E T A L . — W H O B E A R S T H E B R U N T ? 289
of fossil fuels into the economy, thus achieving broadest coverage
whilst minimizing the number of actors included in the scheme and the
related administrative costs. Examples are the proposals for upstream
carbon taxes (Hansen, 2009), Cap and Dividend (Barnes, 2003), Cap
and Share (AEA and Cambridge Econometrics, 2008; FEASTA, 2008)
or the Kyoto2 scheme (Tickell, 2008). A mid-stream scheme would
apply to companies in specific economic sectors; the largest existing
cap scheme, the European Union Emissions Trading Scheme (EU ETS)
is an example. Downstream schemes apply to individuals, and in some
variants businesses, who would have carbon accounts and trade permits
themselves (DEFRA, 2008a; Fleming, 2007).
Within cap and trade schemes, several options exist as to how emis-
sion permits are allocated to the participants – all of which have dif-
ferent distributional impacts. Initial emission budgets can be allocated
to the participants in the scheme free of charge, through auctioning,
or through a mix. For example, in the European Union Emissions
Trading Scheme (EU ETS), emission permits have largely been given
away for free to companies in the participating sectors, depending on
their previous and estimated future emissions. This approach is called
‘grandfathering’ in the literature. It is widely believed that this type
of ‘grandfathering’ leads to windfall profits for companies as they can
pass the additional costs on to customers or sell a considerable volume
of their allocated permits. In other words, ‘grandfathering’ is likely to
have regressive effects (Shammin and Bullard, 2009; Sijm et al., 2006).
In contrast, auctioning the permits makes the polluters pay whilst the
distributional effects depend on the capabilities of the targeted indus-
tries to pass on the cost to the consumer and the availability of alterna-
tives to these goods for consumers. Furthermore, auctioning emission
permits to the participants creates a revenue stream for the government
or organization that issues the permits. We will discuss below how
those revenues can be used to counter-balance possible regressive effects
of mitigation policies.
Distributional effects
Regressivity is a general feature of taxes on consumption, and therefore
one would expect carbon taxes to be regressive. This expectation also
carries through to various types of cap and trade schemes. Overall, the
literature on the distributional effects of mitigation policies confirms
at Freie Universitaet Berlin on April 24, 2015csp.sagepub.comDownloaded from

Frequently asked questions

Q1. What have the authors contributed in "Who bears the brunt? distributional effects of climate change mitigation policies" ?

The authors assess the claim that climate change policies have regressive effects by comparing different types of mitigation policies. 

Q2. What are the future works in "Who bears the brunt? distributional effects of climate change mitigation policies" ?

However, there are several issues related to this which require further debate and research: i. Further research is required to identify the characteristics of low income groups at risk of losing out under lump-sum recycling arrangements. One possibility could be the application of ‘ contraction and convergence ’ ( Meyer, 2000 ) frameworks which envisage equal per capita allowances at a later point in time, thus implying less immediate redistributions between rich and poor nations. The literature indicates that, for a variety of reasons, there will still be a substantial proportion of lower income losers owing to high direct or indirect fossil fuel energy use under lump-sum recycling arrangements. 

Q3. What are the main factors that influence the study results?

Study results are often influenced by the choice of country as climatic conditions, energy efficiency of the housing stock, level of car-ownership, public transport infrastructure, levels of poverty and income inequality and so on all make a difference to potential distributional effects. 

Q4. What is the main reason for higher emissions in rural areas?

DEFRA (2008a: 4) claims, for example, that less access to mains gas (which has lower emissions per energy unit than any other fossil fuel) and less well insulated houses in rural areas are the main reasons for higher rural emissions, rather than greater dependency on cars. 

Q5. How can initial emission budgets be allocated?

Initial emission budgets can be allocated to the participants in the scheme free of charge, through auctioning, or through a mix. 

Q6. How many income groups would lose out under a cap and share scheme?

Several studies also report that even with full and equal per capita revenue recycling, that is an equal per capita payment to all citizens financed through the revenue from the emissions reduction policy, a substantial proportion of low income households would still lose out under such a scheme (DEFRA, 2008a: 3; Dresner and Ekins, 2004: 4; 2006: 55). 

Q7. What is the main reason for increasing per capita emissions in many rich countries?

The trend towards smaller household sizes is therefore a driving factor behind increasing per capita emissions in many rich countries, particularly if indirect emissions are included (Druckman and Jackson, 2008: 3184). 

Q8. What is the definition of revenue neutral?

If the entire revenue is earmarked to decrease/remove other taxes, the tax reform is termed ‘revenue neutral’, meaning that the costs of the new source of revenue are completely compensated through the reduction of other taxes or charges. 

Q9. What is the main argument that regulation is less cost-efficient than market-based instruments?

It is frequently argued that regulation is less cost-efficient than market-based instruments because it does not reflect differing abatement costs of the various polluters. 

Q10. What is the potential power of adverse public reactions to policies?

The potential power of adverse public reactions to policies is evident in the UK, for example, in the abandonment of the poll tax and more recently the fuel tax escalator. 

Q11. How much of the UK’s emissions are indirect?

For indirect emissions comprise a considerable share of households’ overall emissions, in some cases more than 50% (Druckman and Jackson, 2009: 2074; Reinders et al., 2003) whilst overall expenditure including consumer goods generally increases less than proportionally with income (see, for example, ONS, 2009, table A9, for the UK case). 

Q12. What is the effect of auctioning emission permits?

auctioning emission permits to the participants creates a revenue stream for the government or organization that issues the permits. 

Q13. Why would international equal allowance and lump-sum rebate schemes be regressive?

This is because international equal allowance and lump-sum rebate schemes would be regressive in rich countries, and trigger an enormous transfer of resources from the developed to the less developed world. 

Q14. What are the main reasons for the lack of public support for lump-sum rebates?

Whilst lump-sum rebates are likely to attract wide public support, they are likely to be opposed by (high emitting) wealthier people and businesses.