Policy Heterogeneity in Fuel Poverty Alleviation (Dissertation)

Policy heterogeneity in fuel poverty alleviation A comparative analysis of Germany and the United Kingdom

Andreas Schneller

Dissertation submitted in partial fulfilment of the requirements for the Degree of MSc Environmental Policy and Regulation.

The London School of Economics and Political Science

27th August, 2015

Abstract

Social inequalities persist in the kind of housing conditions that are available to different social groups. An indication is the extent to which people are able to obtain energy for domestic purposes such as the heating of rooms. Accordingly, a primary concern for social policy is the question of how to ensure that everyone has access to sufficient energy for his or her wellbeing. However, from an environmental policy perspective, increasing the supply and consumption of energy, especially from fossil fuels, carries high costs for the environment. Fuel poverty, the inability to maintain an adequate level of warmth at a reasonable cost, is a social phenomenon which exemplifies the difficulties that policy makers face in integrating social and environmental policies. This paper will demonstrate that fuel poverty is prevalent in both, Germany and the UK. However, each nation has adapted highly contrasting responses to alleviate the issue. In order to identify and discuss the varying national patterns and approaches this paper applies the method of a cross-national comparative analysis. The analysis will be informed by a theoretical framework which embodies theories of the policy-making process. Identifying factors of policy heterogeneity in fuel poverty alleviation is the primary aim of this paper. In addition, lessons will be drawn from the UK experience in defining, measuring and targeting fuel poverty in Germany. The analysis offers the following results: (1) Fuel poverty is a pressing issue in Germany which requires political attention and definite policy action. (2) Policy proposals for fuel poverty alleviation are hindered by a lack of political recognition, inadequate problem definition and incomplete measurement. (3) The existing social benefits are not well-targeted and insufficient to cope with rising energy costs.

Acknowledgments

I would like to thank the following people and organisations for their assistance with this research:

Dr. Michael Mason, my supervisor, for his perceptive advice and encouragement;

The Hans-Böckler-Foundation, which equipped me with a generous scholarship to cope with the horrendous costs of studying in London;

The EAGA Charitable Trust for awarding me an M.Sc. dissertation bursary which contributed greatly to the depth and scope of this study;

My parents, friends and family who have been an important and indispensable source of spiritual support;

Lastly, my ever-patient and supportive girlfriend, Lina, has coped with being a book widower with great understanding and tolerance – thank you so much!

III

Acronyms

CWP Cold Weather Payment

DECC Department of Energy and Climate Change

DWP Department of Work and Pensions

EC European Commission

ECO Energy Company Obligation

EEG Renewable Energy Act

EESC European Economic and Social Committee

EU European Union

EWM Excess Winter Mortality kWh Kilowatt Hour

LIHC Low Income High Cost

NEA National Energy Action

SOEP Socio Economic Panel

UK United Kingdom

WFP Winter Fuel Payment

WHD Warm Home Discount

List of Figures

Figure 1: The overlap between income poverty and fuel poverty in the UK ...... 5

Figure 2: The vicious circle of energy inefficiency in fuel-poor households ...... 6

Figure 3: Stylised policy circle ...... 11

Figure 4: Changes in gas, oil, electricity prices and nominal wages in Germany ...... 15

Figure 5: Electricity price in ct/kWh for private consumers between 2002 and 2012 ...... 16

Figure 6: Energy-related expenses as a share of available household income ...... 17

Figure 7: Tenure status of people at risk of poverty ...... 18

Figure 8: Fuel poverty under the LIHC indicator ...... 23

Figure 9: Levels of funding for fuel poverty related policies in 2016 (scenario) ...... 27

Figure 10: Targeting strategies for income-related policies ...... 28

Figure 11: Social benefits, actual costs for various household types and deficit p.a...... 29

List of Tables

Table 1: Comparative methodological choices ...... 10

Table 2: EEG Surcharge from 2008 till 2014 with yearly percentage change ...... 17

Table 3: Proportion of UK households in fuel poverty with LIHC and 10% method ...... 23

Table 4: Fuel poverty measures for Germany ...... 24

Table 5: Selected fuel poverty related policies in the UK; Author´s graphic ...... 26

Table of Contents

I. INTRODUCTION ...... 1

Research objectives ...... 2 Structure ...... 3 Terminology ...... 3

II. LITERATURE REVIEW ...... 4

Analysis of Causes ...... 4 Analysis of Effects ...... 7 Limitations of the Existing Literature ...... 8

III. METHODOLOGY ...... 9

IV. THEORETICAL FRAMEWORK ...... 11

V. COMPARATIVE ANALYSIS ...... 14

Causes and effects of fuel poverty in Germany ...... 14 Problem stream ...... 20 Policy stream ...... 26

VI. CONCLUSIONS ...... 30

BIBLIOGRAPHY ...... 32

I. INTRODUCTION

A social phenomenon which exemplifies the difficulties that policy makers face in integrating social and environmental policy objectives is the prevalent issue of fuel poverty (Fitzpatrick, 2011). A household is usually considered to be fuel poor when they are unable to maintain an adequate level of warmth at a reasonable cost (Boardman, 1991).

As Hills (2012: 8) points out, a general consensus has been established that fuel poverty is caused by a combination of three factors: (1) low incomes; (2) high energy prices; and (3) energy-inefficient housing. However, while rising energy prices and low incomes are constituent factors of fuel poverty, these are also issues for people who are not fuel poor (Boardman, 2010). A shortage of income or wealth is unquestionably fundamental to fuel poverty, but on its own does not explain its prevalence and patterning (Hills, 2012). Based on this observation Boardman (2010: 13) rightly emphasises that “the real differentiating cause is the energy inefficiency of the home as a result of insufficient capital expenditure […] with the result that people who are already affected by poverty have to buy the most expensive warmth in relative terms”.

Incorporating this third dimension of thermal housing and efficiency standards differentiates fuel poverty from poverty. It has to be seen as a distinct policy issue and not just a manifestation of more general forms of poverty (Healy, 2004). The crucial role of insufficient housing and capital stocks determines the need for policies which are specific to the problem (Boardman, 1991).

From this perspective, synergies between the areas of social and environmental policy become feasible. More energy-efficient housing can offer a win-win by creating lower energy bills and reduced emissions (DECC, 2013). One key policy domain should therefore be to ensure that all households can afford sufficient energy services while at the same time mitigating the potential environmental impacts.

Poor energy efficiency also means that many low-income households do not heat their home to an adequate level as a result of high costs (DECC, 2013). Under-heating of a property can affect health, both in terms of mortality and morbidity (Liddell and Morris, 2010). While the prevention of medical conditions became the primary rationale for alleviating fuel poverty (Wilkinson et al., 2007a), many other negative effects are evident such as poor diets, mental health conditions, social exclusion and wider costs for the community (Hills, 2012). In Amartya Sen’s (1985) words, fuel poverty can compromise the capability to achieve a range of valued functionings in everyday life.

Research objectives

Several studies demonstrate that aspects of fuel poverty are manifest across the European Union (e.g. Buzar, 2007; Healy, 2004). A recent EU-wide survey counted a minimum of 50 million fuel poor across the EU (EPEE 2009). Some other estimates are considerably higher (EESC, 2010). Rising prices for electricity and gas, widening wealth disparities and declining net incomes for the poorer population as well as insufficient measures to enhance current building stocks indicate that the issue is unlikely to decline over the next years.

However, the extent to which the constituent factors of fuel poverty are addressed and targeted by political actions differs widely. The UK has a long-standing tradition in academic debate and policy measures devoted to fuel poverty. In contrast, German academia and politics have largely ignored and understated the consequences and are just beginning to catch up with research and policy action. Against this backdrop of fundamental differences in political recognition and definite policy action, the question arises of whether policy learning or transfer could potentially lead to greater outcomes for the alleviation of fuel poverty in Germany.

Thus, the more specific research questions of this paper address theoretical and practical concerns: How does the political recognition of fuel poverty and subsequent policy action differ between Germany and the UK? What lessons can be drawn from the UK experience in defining, measuring and targeting fuel poverty in Germany?

The aim of this paper is not to generate new primary data on fuel poverty in Germany. Its aim is to provide a conceptual discussion on fuel poverty policy based on the rich academic debate and previous experiences in the United Kingdom. This study will deliver an analytical account in support for innovative policy action against fuel poverty in Germany.

Structure

The dissertation will consist of five overarching sections. Initially, this paper will conduct a review of pertinent academic literature on fuel poverty. The following Section III presents the justification behind the use of qualitative methodology. The theoretical framework of analysis in Section IV informs the subsequent comparative analysis which will be presented in Section V. Accompanied by specific policy proposals for the German context, the concluding section will offer final comments and recommendations for future research.

Terminology

The terms ‘energy poverty’ and ‘fuel poverty’ are used interchangeably in this paper. This unification is subject to critique since energy poverty is “commonly used in the development community to refer to poverty that is exacerbated by a lack of access to modern energy sources and end-use technologies” (UNDP, 2012: 2). However, it would be inadequate to refer solely to fuel since a number of households in the UK and Germany still use electricity-driven heating appliances. Consequently, it has to be noted that the term ‘energy poverty’ is only used insofar as it refers to a problem of affordability rather than access, a common practice in relevant EU papers (Ürge-Vorsatz and Tirado Herrero, 2012).

II. LITERATURE REVIEW

The purpose of this literature review is to outline theoretical concepts and empirical findings about fuel poverty. Subsequently, this section identifies several gaps in the academic literature, providing a rationale for this research paper.

The scholarly interest in fuel poverty evolved first in the UK, against the background of the persisting energy crisis of the 1970s which catalysed substantial changes in energy policy (Boardman, 1991; 1988). The idea of cheap and abundant energy supply became obsolete, increasing awareness of the tremendous waste of resources in poorly insulated homes and the necessity for affordable warmth1 (Raho, 2012).

Isherwood and Hancock (1979) have been credited with describing the phenomenon first (Liddell, 2012) and subsequent studies by Brandshaw and Hutten (1983) and Osbaldeston (1984) laid an empirical foundation for further research. Surveys and policy proposals induced by civil society actors such as the National Energy Action Charity (NEA) and the National Right to Fuel Campaign complemented the first attempts to stimulate policy action in the UK.

Almost a decade later, Brenda Boardman’s (1991) seminal work ‘Fuel Poverty: From Cold Homes to Affordable Warmth’ was published, marking the point at which the term finally became manifest in British politics and academics. Boardman’s definition was based on a 10 percent income threshold for fuel expenditures which guided early governmental publications. However, definitions for an accurate measurement of the phenomenon are still highly disputed (Moore, 2012; Liddell et al., 2012; Heindl, 2014). Several other indicators have been developed over the course of the last two decades, whereby the latest officially adopted indicator is constructed in the government-induced ‘Hills report’ (2012).

Analysis of Causes

Looking at the causes of fuel poverty more thoroughly, opinions differ on which constituent factors should be the prime target of future policies.

As Hills (2012) points out, a general consensus has been established that fuel poverty is caused by a combination of three factors: (1) low incomes; (2) high energy prices; and (3) energy-inefficient housing.

1 Equivalent to the ability in keeping the home adequately cool in particular climate conditions. 4

Boardman (2010) confirms the strong correlation of low incomes and fuel poverty in her latest analysis. However she points out that up to a third of households affected by fuel poverty are not on a low income and some income-poor households do not suffer from fuel poverty. Accordingly, several authors claim that combating fuel poverty with a mere emphasis on income distribution would be both expensive and ineffective (Palmer et al., 2008; Boardman, 2010; Hirsch et al., 2011). A study by Palmer (2008: 15) found that there is a strong overlap between fuel poverty and income poverty but this overlap is by no means perfect (Figure 1).

1,1 million 1,9 million households in 1,6 million households households in fuel both fuel poverty and in income poverty but poverty but not in income poverty not in fuel poverty income poverty

Figure 1: The overlap between income poverty and fuel poverty in the UK Author’s graphic, adapted from Palmer et al., 2008: 15

Conversely, Preston et al. (2008) and Roberts (2008) argue that one of the key issues for governing future levels of fuel poverty is the relationship between increases in fuel prices and incomes. Wilkinson et al. (2007b: 974) assert that “recent progress to reduce fuel poverty […] is being reversed by the trend of increasing fuel prices”. Consequently, if the income of the fuel poor increases more slowly than fuel prices, which almost all reasonable scenarios believe will happen, then there is a high risk that fuel poverty will persist.

Another factor which catalyses the development of fuel poverty is energy-inefficient housing. The British government has advocated “improved insulation and efficient appliances since at least 1946” (Boardman, 1991: 21), yet no regulations have been introduced at the policy level for several decades (Boardman, 2010). A number of studies have already validated the positive effects and co-benefits of energy efficiency improvements for fuel-poor households (Green and Gilbertson, 2008; Harrington et al., 2005; Heffner and Campbell, 2011; Howden-Chapman et al., 2012; Ryan and Campbell, 2012).

However, the capital costs of improving energy efficiency in homes are usually high, thus households with lower incomes may find it hard to increase energy efficiency because of budget constraints, e.g. to replace appliances or move to more energy-efficient dwellings (Heindl, 2014: 3). This is reflected by the finding that, in the UK, the energy efficiency of the homes of the richest 25 percent of households is on average over four times as high as that of the homes of the poorest 25 per cent (Huby, 1998: 65).

Nonetheless, as Roberts (2008: 4472) points out, building and energy efficiency measures alone are not sufficient to eliminate fuel poverty. A study by Preston et al. (2008) revealed that these measures would leave some 730,000 households in England, 29.1 per cent of today’s fuel- poor households, in fuel poverty. Thus, it is widely recognised that, especially where households occupy much larger dwellings, some “will remain in fuel poverty even if all reasonable energy efficiency improvements are provided” (DTI, 2001: 12).

Finally, Sunderland et al. (2011) demonstrate that many other constituent factors exist beside these main causes, such as individual consumer behaviour and attitudes, precarious living conditions, under-occupied homes and unfavourable tenant–landlord relationships, resulting in a multi-factorial and multi-dimensional problem in which many causalities have to be considered. According to Proidl (2009), this intertwined combination of factors leads to a self- reinforcing effect which is hard to resolve (Figure 2).

Figure 2: The vicious circle of energy inefficiency in fuel-poor households Author’s graphic, based on Proidl, 2009: 26

Analysis of Effects

A large body of literature explores the negative effects associated with fuel poverty. Fuel poverty can damage people’s quality of life and health, as well as imposing wider costs on the community (DTI, 2001: 7).

A growing number of studies link housing quality to human health. Living in an adequately heated home is commonly viewed as protective for human health, wherefore the WHO recommends a minimum temperature of 21°C in living rooms (Liddell and Morris, 2010: 2987). Households affected by fuel poverty often live in persistently cold and damp homes which are thought to generate significant health risks (Liddell, 2008).

Fuel poverty can contribute to poor health outcomes, both in terms of mortality and morbidity (DECC, 2013). (Braubach et al., 2011) found an excess winter mortality (EWM)2 of 12.8 per 100,000 people across eleven European countries because of insufficient heating. EWM claimed an estimated 24,000 lives in England and Wales in 2011/12 (DECC, 2013: 8). EWM is significant especially in countries with poorly insulated homes (Wilkinson et al., 2007b).

But the correlation between fuel poverty and EWM is disputed. Keatinge’s (1986) early work on the effects of unlimited heating supply on EWM did not display any significant differences in EWM. He concluded that insufficient household warmth does not necessarily lead to an increase of EWM, a point he continues to defend (Keatinge, 2001).

Whilst many studies with ambivalent evidence about EWM exist, this should not be taken to imply that the health effects of fuel poverty are either severe or non-existent. Poverty, bad housing conditions and sickness are often joint products, therefore improved health conditions cannot be related to better housing conditions alone (Braubach et al., 2011). Moreover, the effects of insufficiently heated homes are not limited to physical health conditions. Mental health impacts were found to be high among both adults and adolescents (Liddell and Morris, 2010). This pattern of results corroborates the findings of a previous review by Thomson (2003) in which 9 out of 10 studies indicated significant gains in mental health with housing improvements. In addition, physical and mental health issues can also lead to forms of social exclusion, which can be particularly pronounced for vulnerable groups such as children because of illness-related consequences (Anderson et al., 2012; Howden-Chapman et al., 2012).

2 I.e.: “The surplus number of deaths occurring during the winter season […] compared with the average of the non-winter seasons” (Healy, 2004: 785). 7

Lastly, Hills (2012) asserts that the effects relate not only to the direct impacts of insufﬁcient warmth on physical and mental health and wellbeing, but also to the ability to afford other essentials within the available total household budget. ‘Heat or eat’ situations can occur, whereby households are forced to choose between spending money on energy bills or food (Frank et al., 2006; Beatty et al., 2014).

Limitations of the Existing Literature

Against the background of the current state of knowledge, this dissertation may be able to contribute to several gaps in the literature. While fuel poverty in the UK is a thoroughly discussed issue, few attempts have been made to apply a comparative analysis across national borders. This dissertation may be the first attempt to systematically analyse academic contributions and government reports on fuel poverty in the UK with the intention to apply these findings to the German context.

Up until now, literature on fuel poverty in Germany has been incredibly sparse and underdeveloped. In fact there are only three academic publications available on the subject so far, none of them in a peer-reviewed journal (Tews, 2013; Kopatz, 2013; Heindl, 2014).

These ‘pioneer studies’ struggle with limitations in data availability and a lack of theoretical underpinning. The threefold question as to which definition, measurement technique and policy design might be applicable to the German context has not been answered yet and remains unclear. The low level of empirical engagement with these matters and the lack of efforts to develop theoretical generalisations with cross-country analyses motivates the subsequent chapters’ journey into Germany’s state of affairs with regard to fuel poverty.

III. METHODOLOGY

This paper adopts the method of a cross-national comparative analysis. The research tradition dates back to Lijphart (1971), who established it as a methodology in its own right. At a very general level, ‘comparativists’ are interested in identifying the similarities and differences between macro-social units (Lor, 2011: 2). The macro-social units under investigation in this paper are Germany and the UK, which demonstrate policy heterogeneity in fuel poverty alleviation. Whereas the UK has a wide range of specific policies which aim to alleviate fuel poverty, Germany does not apply any specific measures, treating the issue as a problem of general poverty. This dissertation explores the rationale behind these differences in political recognition and policy action.

Hence, this paper addresses a classical question of comparative policy studies - why governments choose different courses of action or decide not to act at all (Heidenheimer, 1990). This knowledge provides the key to understanding, explaining and interpreting diverse outcomes and processes and their significance for current institutional arrangements (Ragin, 1987: 6). However, Baumgartner et al. (2006: 959) point out that past papers which applied comparative policy analysis have narrowly focused on the political system of the US, leaving many possibilities for new studies in different contexts.

A priori, this paper considers a wide range of methodological choices, outlined in Table 1. It applies qualitative methodology with a case-oriented approach as described by Ragin (1987: 35). The use of ‘thick description’ rather than statistics reflects a greater affinity with interpretivist metatheory, founded on the theoretical belief that reality is socially constructed and fluid (Lor, 2011: 5). The comparative research design represents a national pattern approach which assumes national-level characteristics as a major factor for diverging policies. An a priori hypothesis is not considered due to the qualitative nature of the methodology. The dissertation aims to carry out an inductive approach, potentially generating hypotheses as well as contextualising - that is, analysing socio-economic phenomena in relation to their institutional and socio-cultural settings. Finally, the frame of reference for the comparison is constituted by theories of the policy-making process outlined in the theoretical framework (Chapter IV).

Table 1: Comparative methodological choices Author`s graphic, based on Peter Lor: International and Comparative Librarianship, 2011: 6

Selecting the UK and Germany for comparison follows the idea of a ‘most similar systems’ design, based on John Stuart Mill`s ‘Method of Agreement’ (Mill, 1862), where the selected countries are very similar in many characteristics (e.g. Economic development - GDP/capita; EU membership; Political pluralism; Democratically governed, etc.) except for the particular factors of which we want to study the effect, in this case the highly contrasting responses to the challenges posed by fuel poverty.

The rationale for the UK as a unit of analysis should have become clear via the literature review. The vast number of studies on fuel poverty in the UK as well as the lively political debate there offer the opportunity to identify gaps in research and policy action for the German case while at the same providing the chance to discover useful guidance for defining, measuring and targeting the issue. In addition to the already identified gaps in the literature, another argument for Germany as a unit of analysis will be provided in the analysis, indicating the scale of the problem and its adverse effects at the national level.

IV. THEORETICAL FRAMEWORK

A common way to conceptualise the policy-making process is to break it into a number of discrete stages which comprise a policy cycle (Figure 3).

Figure 3: Stylised policy circle Author’s graphic, based on May and Wildavsky, 1978

The concept highlights how policy-making typically involves multiple stages, decisions and outputs. Due to the limited scope available, this study mainly focuses on the first two stages of the policy process, the (1) agenda setting and (2) policy formulation stages.

However, the policy cycle can be potentially misleading in that it grants the impression that policy-making is a linear, rational and systematic process which can be understood in technocratic terms (Bevir, 2008: 155). In reality public policy-making often follows a ‘messy’ pathway with policy actors who do not always carry out optimal and rational decisions due to bounded rationality, ideological predispositions and normative commitments (Cairney, 2012: 41–43). Furthermore, the concept potentially fails to take into account that particular stages may occur in a different order, overlap with others or be skipped altogether (ibid.). Still, it provides a useful heuristic model for our comparative analysis across two different units of analysis, the national patterns of policy-making in the UK and Germany.

The policy-making process usually begins with the agenda-setting stage, which describes how attention is allocated in political systems (Jones and Baumgartner, 2005: 2). At any one time, there are multiple issues on the agenda, each competing for public attention (ibid.).

The ‘agenda’ refers to a set of issues that receive serious attention in a polity (Cobb and Elder, 1972: 86). Depending on who is giving this attention, several types of agenda may be discerned, such as the political agenda and the public agenda as part of an ‘agenda universe’ (Princen and Rhinard, 2006: 1120; Cobb and Elder, 1972). Consequently, state and non-state actors can be similarly instrumental in putting issues onto the respective agenda.

Political agenda-setting is described as a process whereby issues become policy problems as a result of social processes of claims-making which render them ‘worthy of concern’ (Cairney, 2012: 182). Frequently, claims-making involves elements of defining the nature and magnitude of problems, causal mechanism(s), identity of the actors responsible and possible solutions (ibid: 184).

Claims are constructed and articulated within the context of particular discourses whereby “each discourse rests on assumptions, judgments and contentions that provide the basic terms for analysis, debates, agreements and disagreements” (Dryzek, 1997: 8). A notable academic in the field of discourse analysis, Pierre Bourdieu (2003: 167), assumed that “the different classes and class fractions are engaged in a symbolic struggle, one aimed at imposing the definition of the social world that is best suited to their interests”. According to Bourdieu (2014), the state in particular has considerable power to manipulate discourses by the means of communication in order to constitute specific knowledge. This idea that the state plays a central role in the construction of policy issues concurs with Stuart Hall’s notion of the state as the ‘primary definer’ of social problems and debates (Hall, 1978).

The study of agenda-setting and public attention rarely identifies incremental change but rather rapid, nonlinear processes (Baumgartner and Jones, 1993: 10). Baumgartner et al. (2006: 962) laid out a proposal for a ‘punctuated equilibrium’ theory based on the observation that “rapid, destabilising change is possible even in those policy systems which have been stable for decades”. According to Hall (1993: 291), “long periods of continuity [are] punctuated occasionally by the disjunctive experience of a paradigm shift”.

One cause for policy change are sudden shocks to the political system, so-called ‘focusing events’ which can be large socio-political events which shatter policy routines and force new issues and ideas onto the agenda by their sheer magnitude (Cairney, 2012: 234). They thereby become the focus of public attention and provide new opportunities for groups to mobilise. Focusing events may reveal policy failures which can be exploited opportunistically by particular groups (Birkland, 2006).

In order to discern the different processes which constitute ongoing ‘policy discontinuity’ (Baumgartner et al., 2006: 960) this study adapts one of the most influential theoretical models of agenda setting, Kingdon’s (1995) multiple streams framework (MSF). The MSF emphasises the contingent, chance nature of agenda-setting (Knill and Tosun, 2012) and aims to explain policy-making and the selection of particular ideas over others under conditions of ambiguity: for instance, where there are many different ways of making sense of a particular situation and framing solutions, or where competition means that there is limited space on the decisional agenda and there is limited time to make choices (Cairney, 2012: 233).

Kingdon specifies three variables which ‘flow’ through the policy system:

• Politics stream – “…consists of three elements: the national mood, pressure group campaigns, and administrative and legislative turnover” (Kingdon, 1995: 162).

• Problem stream – “…consists of various conditions that policy makers and citizens want addressed” (ibid: 113).

• Policy stream – “…includes various ideas that compete to win acceptance in policy networks” (ibid: 143).

The three streams must coincide in order for an available solution to a recognised problem to obtain the political attention necessary to be used. Individual ‘policy entrepreneurs’ are required to build acceptance for solutions and to create couplings between these streams of processes “which develop and operate largely independent of one another” (ibid: 88). Policy decisions are made when the three streams coincide during so-called ‘policy windows’. Windows are opened by events in either the problem or political streams, “for a short time when the conditions to push a given subject higher on the policy agenda are right” (ibid: 88).

V. COMPARATIVE ANALYSIS

After constructing the analytical framework and developing a set of propositions, this chapter will analyse the causes and effects of fuel poverty in Germany and compare the policy heterogeneity in fuel poverty alleviation in the UK and Germany.

Causes and effects of fuel poverty in Germany

This chapter will outline empirical findings related to fuel poverty in Germany, providing an overview of its context-specific causes, the scale of the problem and its adverse effects. It is an attempt to assemble widely dispersed evidence, since there is no comprehensive study or governmental report available so far. The analysis is guided by the constituent factors of fuel poverty outlined in the literature review.

Income-related Factors

The most significant cause of fuel poverty is poverty itself. Energy-related expenses can become a severe problem for households with low incomes.

Poverty and rising wealth disparities are not only an issue in developing nations but also in many developed countries (OECD, 2011), including Germany (BMAS, 2013). German officials like to point out that the available income per household has followed an overall positive trend over the last decade. But inequality has grown in the same period. Individuals in the five lowest income deciles suffered on average from an absolute decline in their net income of 8.2 percent between 2000 and 2010 (DIW, 2010). The distribution of wealth followed a similar pattern, with a record high Gini-coefficient of 0.78 in 2012, the highest in the Eurozone (DIW, 2014). The poorer half of the German population possesses only 1.4 percent of total wealth, whereas the upper ten percent holds 66.6 percent (DIW, 2014).

German households with an income of less than 60 percent relative to the median income of the overall population are said to be ‘at risk of poverty’. A disposable income of 952 euros per month is currently set as the threshold for a single household (BMAS, 2013: 460). In 2008 around 16 percent of the German population were affected by relative income poverty (BMAS, 2013: 461), compared to 12 percent in 1999 (Kopatz et al., 2000: 10).

With reference to Boardman’s definition (1991), the number of households that had to spend more than 10 percent of their available income on energy climbed to 6.9 million (17 percent) in 2012, compared to 5.5 million (13.8 percent) in 2008 (Bundestag, 2012: 2). In 2012, almost 4 million households received social benefits for housing and heating costs, worth a total of 15 billion euros (Bundestag, 2014: 6).

Energy Price Dimension

However, low-income households are not only at risk of becoming fuel poor because of declining net incomes and widening inequality, but also through rising energy prices. Following the general upward trend on the world market, significant price increases for all fossil fuels have taken place in Germany over the last couple of years. These were well above the average rate of inflation for consumption goods (Kopatz, 2013: 31) and, more importantly, greatly exceeded the growth in nominal wages (Figure 4).

Figure 4: Changes in gas, oil, electricity prices and nominal wages in Germany Author’s graphic; Data: Bundestag, 2014: 2

While nominal wages increased on average by 17 percent between 2002 and 2012, combined heating costs underwent a rapid growth of 43 percent during the same period (Bundestag, 2014). In addition, markets for fossil fuels remain volatile, leaving low-income households vulnerable to price fluctuations.

Whereas resource prices are determined by the dynamics of global commodity markets, domestic energy prices reflect to a great extent political decisions on energy policy. Analysis of EUROSTAT data by Heindl et al. (Heindl, 2014: 18) found that prices for fossil fuels such as gas were about the same in Germany and the UK in recent years, but electricity prices for private households were considerably higher in Germany than in the UK. Figure 5 illustrates the sharp price increase for private households since 2002.

Figure 5: Electricity price in ct/kWh for private consumers between 2002 and 2012 Graphic: Schultz, 2014; Data: VZBV, 2014

The German Renewable Energy Act (German: Erneuerbare-Energien-Gesetz, EEG) can be identified as one driving force beyond the cost disparity.3 It came into force in the year 2000 and was the initial spark for a significant boost to renewable energy supply in Germany. Subsidies for renewable energy carriers are redirected to electricity consumers via the so-called EEG-Surcharge. Table 2 demonstrates that the surcharge has grown considerably in recent years, leading to higher prices, especially for private households, since businesses benefit from partial exemptions.4

3 Other correlating factors such as market design and lack of competition among energy suppliers are not elaborated on here due to space limitations. 4 In particular, energy-intensive industries are relieved from the surcharge, and in 2014, a total of 2,100 companies received a waiver worth more than 5 billion euros (Döschner, 2015).

Table 2: EEG Surcharge from 2008 till 2014 with yearly percentage change Author’s graphic; Adapted from BDEW, 2014

The surcharge is expected to increase further if the ambitious policy targets of the energy transition are achieved.5 From a distributive justice perspective, this lump sum effectively leads to higher cost burdens for poorer households relative to wealthier ones, which is illustrated for energy-related expenses in figure 6.

Figure 6: Energy-related expenses as a share of available household income The numbers represent the median income share for each group. Author’s graphic, based on Heindl et al., 2014: 512; Data: German SOEP, Version 29

The households with the upper 70 percent of income now spend 2.3 percent more of their available income on energy-related expenses compared to 1998. For the lower 30 percent of households, the increase is almost twice as high: 4.4 percent compared to 1998. Consequently, it can be said that the energy-related cost burden for poorer households has increased significantly in the last decade, indicating the uneven distribution of costs related to the German energy transition.

5 The current target is a share of 80 percent for renewable energy by 2050 (BMUB 2010: 5). 17

Housing and Appliances Dimension

Various studies in the UK have also underlined the close relationship between low incomes and badly insulated, energy-inefficient homes. This relationship has also proven to be true in the German context, where poorly renovated homes are often occupied by low-income residents (Kopatz, 2013). A study by Achtnicht and Madlener (2012) found that the income of house owners is the most significant driver for energy retrofits in Germany, where house owners with lower incomes are less likely to invest in energy retrofits.

Many oil and gas heating appliances in Germany are completely obsolete. 35 percent (2.5 million) are more than 21 years old and 13.5 percent (0.9 million) are more than 29 years old (Bundestag, 2014: 13). The yearly rate of heating system modernisation between 2005 and 2010 was on average 2.8 percent, lagging far behind the actual demand (Bundestag, 2014: 8).

A specific problem in German houses with partial remediation is their often outdated electric night storage heaters. The Federal Cartel Office estimates that 1.6 million households are still using this energy-intensive technology (BkartA, 2010). This form of heating has become even more costly in recent years, as tax exemptions have been cut back and competition among energy providers for special overnight tariffs is now largely absent (BkartA, 2010).

In the case of Germany, it has also been shown that increases of energy efficiency in space heating are more pronounced in owner-occupied dwellings compared to rented dwellings (Heindl, 2014). This finding is especially worrisome since Germany has the lowest level (25 percent) of owner-occupied housing for individuals below the poverty threshold in Europe (Figure 7).

Figure 7: Tenure status of people at risk of poverty Graphic: EPEE, 2009: 35; Data: EUROSTAT, 2012 18

Against the background of the data provided it is hardly surprising that the number of power disconnections in Germany has increased sharply over the past two decades, reaching a peak of 344,000 in 2013, an increase of 33,000 compared to 2011 (BNetzA, 2014: 124).

Furthermore, a great number of households face unpaid energy bills and indebtedness. 71 percent of German energy suppliers see increasing numbers of late payments and outstanding debts as a serious issue (VZ NRW, 2011). Almost seven million default actions by energy suppliers were carried out in 2013 (BNetzA, 2014: 124), an increase of almost one million compared to 2011 (BNetzA, 2014: 124).

In a nutshell, the empirical findings demonstrate that fuel poverty is a pressing issue in Germany which urgently requires more political attention and definite policy action.

Problem stream

Kingdon (1995: 90) considers three main channels through which problems capture the attention of policy-makers. One possibility is that a more or less systematic indicator of a problem affects political recognition. At other times, focusing events or feedback from the operation of existing programs can seize political attention.

Feedbacks, formally via systematic monitoring or evaluation studies of current measures, or informally via complaints (ibid: 101), can be ruled out for our investigation since there is no basis for evaluation in the German case. Feedback effects are not expected to be significant unless particular policies are already in place (ibid: 104).

However, the German Energy Transformation (‘Energiewende’) as a large socio-political ‘focusing event’ might play a decisive role in raising governmental attention in the German case. Rising prices for goods with a limited substitutability and inelastic demand, such as secondary energy, are likely to cause regressive effects. Therefore, the budget of low-income households is expected to be much more affected by rising prices for energy than those of better- off households. In the case of the German Energiewende this fact provokes conflict because it is not only the anonymous allocation mechanism of the market which is responsible for rising energy prices but the political decision to pursue the costly transformation.

The significant increase in energy prices in Germany has led to an intense debate about the potential social drawbacks of the energy transition and the containment of costs. There is a growing risk that low-income households will be left behind and join the coalition of opponents, undermining future progress. As a result, policy change is demanded from both sides: adversaries and advocates of the Energiewende.

Similarly to the UK, where the oil crisis of the 1970s shattered policy routines and triggered the development of new policies against fuel poverty (Boardman, 1991), escalating energy prices in Germany can alter the “national mood” (Kingdon, 1995: 146)6, the public opinion on the necessity for an energy transformation, in the worst case leading to diminished public support for the Energiewende while at the same time provoking reactive political measures.

6 Strictly speaking, the national mood is part of Kingdons ‘politics stream’ which has been ruled out due to space constraints together with the role of ‘policy entrepreneurs’ and ‘claim makers’ in agenda setting. 20

However, the impact of focusing events on political attention should not be exaggerated. Kingdon (1995: 197) points out that “such an event has only transient effects unless accompanied by a firmer indication of a problem”. For this reason we now turn to indicators as the third contributing factor for political agenda-setting.

To define indicators, Kingdon (ibid: 91) distinguishes between routine monitoring of activities or events (e.g. disease rates) and studies conducted on a particular problem. As this paper has already highlighted, Germany does not have much to show in either regard. There have been no specific studies on fuel poverty conducted so far, and neither has a specific indicator for the purpose of fuel poverty monitoring been established. Consequently, as a first step, it is vital for German politics to develop a conceptual framework which accurately captures the potential issue. Unless fuel poverty is properly measured it is difficult to design effective solutions.

How to define and measure fuel poverty

The variety of different indicators for fuel poverty is leading to highly contrasting results about the qualitative and quantitative manifestations of fuel poverty. International comparisons are fraught because of inconsistent techniques. Ireland, for instance, assesses fuel poverty in relation to warmth only, excluding the usage of energy for other domestic purposes, leading to half the number of fuel poor compared to the UK definition (Boardman, 2010: 16).

A large number of measurement proposals have been developed so far. Some definitions focus on fuel expenditures of households relative to income, such as Boardman’s 10 per cent threshold. Others focus exclusively on household expenditures for energy services relative to the median or mean expenditure of the overall population. Some approaches concentrate on the basic needs of different types of households based on minimum income standards (MIS) (Heindl, 2014: 8) or try to combine multiple thresholds such as Hills’s (2012) LIHC indicator.

In order to lay the foundation for a definition and measurement proposal for the German context, two important conceptual questions arise from all approaches: what is to be defined as ‘income’ and what is to set as the threshold for ‘adequate energy services’?

A critical issue for the calculation of incomes are accommodation costs. Two approaches, the ‘basic income’ and the ‘full income’, are distinguishable (Moore, 2012). Whereas the basic income approach includes the net incomes of all household members plus private (e.g. interest rates) and public benefits (e.g. unemployment benefits), it is only the full income approach that also includes incomes generated by the accommodation itself (e.g. housing benefits, rent revenues) (Moore, 2012).

The UK government decided to use the full income approach, since a considerable number of well-off households would be identified as fuel poor without the consideration of rent revenues (DECC, 2014). However it has to be taken into account that housing benefits are often a fraction of the net income and the current rent, which can lead to a paradoxical situation with an increase in the full income value when the rent is rising and income remains constant.

The current definition of ‘full incomes’ is based on a non-equivalised scale, without an adjustment for household size and composition, resulting in an intended bias towards single households, often elderly people, who are classified as ‘vulnerable’ together with minors and disabled or long-term sick individuals (Hills, 2012: 30; Moore, 2012). The UK ‘Fuel poverty strategy’ (DTI, 2001) focused especially on this particular group and proposed that no vulnerable households should remain fuel poor in 2010, a goal which has been flatly missed.

‘Adequate energy services’ were initially calculated by Boardman (1988) based on the data of actual energy consumption per household in the UK. She later revised this technique, now estimating the cost burden based on the average expenditures for heating which are necessary to reach an adequate room temperature of 21°C in accordance with WHO standards.

This idea of a “need to spend” (Boardman, 2010: 23) became the standard for most subsequent publications on fuel poverty. This new approach is more successful in capturing the full extent of fuel poverty because it also includes those individuals who are intentionally reducing their energy consumption below WHO standards in order to save money.

The debate about an appropriate measurement gained further momentum when Boardman’s widely applied 10 percent threshold was considered inappropriate by the DECC-commissioned Hills Report. It stated that “the 10 per cent indicator had painted a misleading picture, understating the scale of the problem when energy prices were low and overstating it when prices were high” (Hills, 2012: 8). The report proposed that a household should be seen as fuel poor if it (1) faces fuel costs above the median level and (2) is left with a remaining income below the 60 percent poverty threshold after expenditures on energy services and housing costs.

One of the key features of this new LIHC indicator is the ability to distinguish more precisely between the extent of the problem (i.e. how many households are fuel poor) and its depth (i.e. what is the severity of the fuel poverty that they face) (DECC, 2013: 9). The fuel poverty gap illustrates the difference between a household’s modelled bill and what their bill would need to be for them to no longer be fuel poor (Figure 8).

Figure 8: Fuel poverty under the LIHC indicator Graphic: DECC, 2014: 6

The new numbers, based on the LIHC indicator, cut 2009 levels of fuel poverty by nearly a third – down from 4.0 million to 2.7 million (for relative terms see Table 3). Even under a relative definition, this estimated fall in fuel poverty since 2004 is questionable, taking into account that for those on the lowest 30 percent of incomes, the average rise in unit fuel costs since 2004 has been 76.2 percent, while the increase in incomes has been only 14.1 percent (Moore, 2012: 26).

Table 3: Proportion of UK households in fuel poverty with LIHC and 10% method Authors graphic, Data: DECC, 2013

The case for Germany

Based on these proposals, Heindl (2014) estimates the number of fuel poor in Germany in a recent paper based on household data from the German socio-economic panel (SOEP), a comprehensive longitudinal data set updated yearly since 1984 (DIW, 2014). Table 4 shows a selection of results from his study based on data from the SOEP 2011.

Type Share of fuel Average Share of fuel poor Average exp. on poor in equivalised income below the poverty energy services sample of fuel poor line (952 EUR) of fuel poor

10% of NEI 25.1% 1.054 45.7% 20.3%

2x mean share of 4.9% 779 74.8% 36.5% energy expend.

MIS (SGB II) 8.8% 721 89.2% 25.0%

LIHC EI 10.5% 907 52.2% 26.3%

Table 4: Fuel poverty measures for Germany Author’s graphic; Data: Heindl, 2014: 20

As shown in Table 4, the share of households potentially subject to fuel poverty is quite high (25.1 percent) if the 10 percent non-equivalised income (NEI) indicator is applied. The figure is considerably higher compared to England, where 15 percent of households fell under the 10 percent threshold in 2011 (DECC, 2013, p.19). The 2x mean share of energy expenditure indicator leads to considerably lower proportions of fuel-poor households (4.9 percent). The MIS poverty line identifies 8.8 percent of households as subject to fuel poverty. These households have the lowest average equivalised income (EUR 721) compared to other poverty lines. About 90 percent of fuel-poor households under the MIS also fall below the income poverty line.

The LIHC poverty line identifies 10.5 percent of households as fuel poor, with an average equivalised income (EI) of EUR 907, which is close to the income poverty line of EUR 952. Fuel-poor households have particularly high shares of expenditure on energy services relative to income, with a mean of 26.3 percent. As in the UK, the LIHC approach delivers a significantly lower share of fuel poor compared to the 10 percent approach (Heindl, 2014: 19).

This paper considers the MIS indicator based on research by Bradshaw et al. (2008) as especially promising for an initial assessment of fuel poverty in Germany. In this context, households are in fuel poverty if they have insufficient residual net income to meet their total required fuel costs after all other minimum living costs have been met (Moore, 2012: 22). German standard welfare rates (according to SGBII) could be applied as a minimum income standard.

SGB II rates are allocated depending on household composition. SGB II rates can therefore be interpreted as an available minimum income on an equivalised scale and could potentially be used as a focal point for a fuel poverty assessment. As a result, the MIS indicator can consistently and accurately measure the affordability of fuel, regardless of whether fuel-related benefits are added to incomes or taken from fuel costs. In principle, it should be readily translatable to other countries with different incomes and minimum living costs, provided that required fuel costs rather than actual fuel expenditure can be determined (Moore, 2012: 22).

However, at this stage of the discussion it is too early to make a final judgement as to which indicator might be suitable for the German context. To start with, further research should be carried out to determine whether the MIS indicator could deliver good guidance for policy action. The previous experience of the UK, where only a few indicators have received considerable scientific and political traction over the past two decades, tells us that many technical difficulties accompany fuel poverty measurement. In any case, an intensified discussion about an appropriate definition for fuel poverty could contribute to heightening political attention on the issue.

Policy stream

The policy stream consists of a “policy primeval soup” where many ideas float around to form combinations and recombinations through a selection process which is hard to predict (Kingdon, 1995: 200). The short list of proposals which remain on the political agenda have to fulfil several criteria: technical feasibility, congruence with community values and the anticipation of future budget constraints as well as acceptance by the general public (ibid: 143).

The UK already carries out a wide range of policies to alleviate fuel poverty. A detailed framework of the current policies is illustrated in Table 5. The policies address all three constituent drivers of fuel poverty as defined by Hills (2012), however the level of total funding available for each policy differs considerably as shown in Figure 9. The policies also target different types of households. Present policies such as CESP, CWP and WHD are focused on low-income households, but not necessarily those with higher than typical energy costs (Hills, 2012: 104). CERT, ECO and WFP schemes are more widely spread over the population.

Fuel poverty Policy Scheme Policy type Conditionality / drivers targeted Eligibility criteria Winter Fuel Payment Non-means tested benefit 60+ aged only (WFP) to households

Income-related Direct benefits

Cold Weather Payment Means-tested benefit to Various (CWP) households

Energy-cost Warm Home Discount Direct energy Related (WHD) bill support Social tariff Various

Warm front scheme (WFS) Grants for heating and Subsidy Households receiving till 2012 insulation improvements particular income-related

benefits

National investment Direct Decent Homes Programme programme for thermal Investment Local authority housing efficiency

Warm Zones Programme NEA administered energy- Thermal- Advisory; Deprived areas only

efficiency efficiency scheme Subsidy

related Energy Efficiency Commitment (EEC);

Thermal efficiency Community Energy Saving ‘priority groups’ measures to reduce carbon Programme (CESP); Statutory of low-income and emissions in the household obligations vulnerable households Carbon Emission Reduction sector Target (CERT)

Table 5: Selected fuel poverty related policies in the UK; Author´s graphic 26

There is a range of means-tested and non means tested benefits that are aimed at increasing household incomes, subsidising energy bills or supporting efficiency measures. A more detailed account of the income-related policies reveals that accurate targeting of fuel poor households is by no means perfect.

The WFP represents a non-means tested benefit paid directly into the bank account of all households with a member aged 60 or over. There is no requirement to spend any part of it on energy and studies found that only 40 percent of WFP are spent on energy (Beatty et al., 2012). Furthermore Hills (2012: 113) estimates that only 10 per cent of the recipients are considered as fuel poor on the basis of the LIHC definition. Consequently, the WFP has attracted heavy criticism, with suggestion that it should be renamed the ‘pension supplement policy’ (Raho, 2012: 33), not only because of its ineffective targeting, but also because it receives the largest share of funding among all fuel poverty policies (Figure 9).

Figure 9: Levels of funding for fuel poverty related policies in 2016 (scenario) Graphic and Data: Hills, 2012: 16

In contrast, the CWP is regarded as an example of a well-targeted policy (Hills, 2012: 113). Its eligibility criteria combine low income (individuals who receive one of a number of means- tested benefits) and vulnerability characteristics. When temperatures remain below a threshold over a defined period of time eligible households have access to funds. 2012 funding was at £431 million, only a fraction of the WFP (Raho, 2012: 33).

Targeting for income related policies is essential to allocate the available funds to those households who need it most. In any case there would be no change to the costs threshold but accurately designed policies (e.g. the CWP) could reduce fuel poverty by lifting households in the fuel poverty quadrant above the income threshold (Figure 10). In contrast if the policy is not well targeted (e.g. the WFP) it will move both, people who do and do not suffer from high energy costs, upwards.

Figure 10: Targeting strategies for income-related policies Author`s graphic; based on Hills 2012: 101

In addition, some criticize that the scale of funding for fuel poverty policies in the UK never matched the scale of the issue. There are estimates that it would cost £18 billion to improve the homes of fuel poor households to the target Band C, while assuming perfect targeting of fuel poor households (FPAG, 2015: 8). In contrast, the current level of investment in energy efficiency measures in households likely to be fuel poor is nowhere near sufficient with only £460 million a year (ibid).

The case for Germany

Unsurprisingly, policies which target fuel poverty directly are largely absent in Germany, based on the widespread perception that heating and energy expenses are sufficiently covered by established social benefit measures. However, a more detailed analysis reveals that this assumption does not hold true.

Roughly five million recipients of unemployment or social welfare benefits receive a minimum of 391 euros per month (so-called Hartz IV basic provision). 33 euros are implicitly allotted to energy expenses (Sell, 2015). Given current energy prices, this is only sufficient to cover 1,350 kWh a year, a level of consumption which is hard to achieve even for very thrifty individuals. A survey by Sell (2015) found out that an average single household spends 42 euros a month, causing a funding shortfall of 116 euros p.a. This considerable gap of 29 percent has to be compensated by savings in the consumption of other essential goods. Finally, figure 11 illustrates gaps for different types of households, based on data by Tews (2013: 21).

In addition, heating costs are only covered if they are considered ‘reasonable’ by the respective municipality. Each federal district has its own calculation and threshold for ‘reasonable’ costs, leading to a situation where the degree of cost absorption differs widely across regions (IFEU, 2010). A nation-wide study (IFEU, 2010) found that only 45 percent of social benefit recipients received a full coverage of heating costs. Social welfare organisations claim, therefore, that the current heating allowances are entirely insufficient (HBS, 2013; DCV, 2013).

Figure 11: Social benefits, actual costs for various household types and deficit p.a. Author`s graphic; based on Tews, 2013: 21; Data: BNetzA, 2014

The findings demonstrate that significant discrepancies exist between the estimated energy costs included in the social benefit rates and the actual costs households are exposed to. In fact, welfare recipients face increasing problems in attempting to cope with their energy bills since the energy prices increase at a much faster rate than the standard rate of welfare benefits.

VI. CONCLUSIONS

The aim of this dissertation was to analyse, from a theoretical viewpoint, how political recognition of fuel poverty and subsequent policy action differ between Germany and the UK, and what practical lessons can be drawn from the UK experience in defining, measuring and targeting fuel poverty in Germany.

The investigation has shown that the fuel poverty agenda is at risk of being overlooked in both political discourse and policy deliberation. For several years it appeared that the UK alone suffered from the problem, but the analysis of the German case has revealed that this seems to have reflected different academic and political concerns, rather than an absence of the problem. Policy inaction, accompanied by rising energy prices, widening wealth disparities and declining net incomes, as well as insufficient measures to enhance current building stocks, have led to a manifestation of fuel poverty in Germany.

Within Germany, fuel poverty has been regarded and treated as a problem of general poverty and has largely failed to be classified as a distinct policy issue. This study reaffirms that fuel poverty is not just another angle on poverty more generally. An adequate definition would clarify the importance of capital investment and correctly reduce the role of income support and energy price subsides. While social policy measures such as heating allowances or a monthly income support can lift a person temporarily out of fuel poverty, they do not address the underlying problem of energy-inefficient housing conditions.

Furthermore, the analysis demonstrated that the choice of an appropriate fuel poverty definition and indicator is decisive for the subsequent assessment. The substantial societal benefits which arise from the concurrent alleviation of all constituent factors of fuel poverty will only be grasped by policy-makers when they understand the full scope of the issue in the first place.

The study showed that the MIS indicator has good practicality when applied to the German context, although it was beyond the scope of this study to engage with technical details in greater depth. In order to attract political attention, more research is necessary to develop an indicator which accurately captures the issue. Moreover, for policy purposes, a move from the paper’s general focus on policy-making in theory to the evaluation of particular programs in the UK could be useful. A further limitation is that the findings in this study cannot be readily be applied to develop broad generalisations for other (European) countries since the comparative method is exposed to the inherent problem of many variables and few cases (Lijphart, 1971). 30

This paper suggests that the impact of environmentally motivated policies on energy prices, such as the EEG surcharge, might be of particular importance with respect to fuel poverty based on the insight that the energy-related cost burden for poorer households has increased significantly in Germany, partly due to subsidies for renewable energy technologies. At the same time, findings confirm that existing social transfers are not well-targeted and insufficient to cope with rising energy costs. Accordingly, this paper argues that alleviating fuel poverty is becoming an increasingly important condition for maintaining the necessary public support for the German Energiewende. The whole project is at stake if escalating energy prices for private households persist. Restricting the EEG exemptions for the energy-intensive industry would be a first step in distributing the cost burden of the energy transition more equally.

A well-designed policy mix against fuel poverty is a chance to bridge the gap between environmental and social policy objectives. Conversely, there is a risk that ill-conceived policy mechanisms may lead to high energy prices, perverse incentives and public dissatisfaction with the German Energiewende at large. Fuel poverty alleviation exemplifies the difficulties and trade-offs policy-makers face in integrating social and environmental policy objectives. An unconditional full coverage of heating expenses could be an incentive for wasteful consumption habits, contradicting attempts to lower environmental impacts. However, the common practice of purposefully leaving households in fuel poverty is unjust and socially unacceptable.

BIBLIOGRAPHY

ACHTNICHT, M. AND MADLENER, R. (2012). Factors Influencing German House Owners’ Preferences on Energy Retrofits. ZEW Discussion Papers: 12–42.

ANDERSON, W., WHITE, V. AND FINNEY, A. (2012). Coping with low incomes and cold homes. Energy Policy 49: 40–52.

BAUMGARTNER, F., GREEN-PEDERSEN, C. AND JONES, B.D. (2006). Comparative studies of policy agendas. Journal of European Public Policy (13:7): 959–974.

BAUMGARTNER, F.R. AND JONES, B.D. (1993). Agendas and instability in American politics. Chicago: Univ. of Chicago Press.

BEATTY, T., BLOW, L. AND CROSSLEY, T. (2014). Is there a ‘heat-or-eat’ trade-off in the UK? Journal of the Royal Statistical Society: Series A (Statistics in Society) 177 (1): 281–294.

BEATTY, T.K.M., BLOW, L. AND CROSSLEY, T.F; O'DEA, C. (2012). Cash by any other name? Evidence on labelling from the UK winter fuel payment. Istanbul: TÜSİAD- Koç University Economic Research Forum.

BEVIR, M. (2008). Key Concepts in Governance. London: Sage Publications.

BIRKLAND, T.A. (2006). Lessons of disaster: Policy change after catastrophic events. Washington, DC.: Georgetown Univ. Press.

BKARTA (2010). Sektoruntersuchung Heizstrom Marktüberblick und Verfahren: Abschlussbericht gemäß § 32e GWB. [Sectoral research on Heating Energy], Bundeskartellamt [Federal Cartel Office], Berlin.

BMAS (2013). Lebenslagen in Deutschland: Der Vierte Armuts- und Reichtumsbericht der Bundesregierung [Living Conditions in Germany. The Fourth Poverty and Wealth Report of the Federal Government], Bundesministerium für Arbeit und Soziales, Berlin.

BMUB (2010). Energiekonzept für eine umweltschonende, zuverlässige und bezahlbare Energieversorgung. [Energy concept for a sustainable, safe and affordable Energy supply], Bundesministerium für Umwelt, Naturschutz, Bau und Reaktorsicherheit [Federal Ministry of Environment]. Available at: https://www.nachhaltigkeit.info/media/1285831803phpz9Rqi2.pdf (accessed 9 August 2015).

BNETZA (2014). Monitoringbericht 2014. [Monitoring Report 2014], Bundesnetzagentur für Elektrizität, Gas, Telekommunikation, Post und Eisenbahn. Available at: http://www.bundesnetzagentur.de/SharedDocs/Downloads/DE/Allgemeines/Bundesnetzag entur/Publikationen/Berichte/2014/Monitoringbericht_2014_BF.pdf?__blob=publicationFi le&v=4 (accessed 25 July 2015).

BOARDMAN, B. (1988). Economic, social and techincal considerations for fuel poverty policy: University of Sussex.

BOARDMAN, B. (1991). Fuel poverty: From cold homes to affordable warmth. London, New York: Belhaven Press.

BOARDMAN, B. (2010). Fixing fuel poverty: Challenges and solutions. London, Sterling, VA: Earthscan.

BOURDIEU, P. (2003). Language and symbolic power. Cambridge: Harvard Univ. Press.

BOURDIEU, P. (2014). On the state: Lectures at the Collège de France, 1989-1992. Cambridge: Polity.

BRADSHAW, J., MIDDLETON, S. AND DAVIS ABIGAIL (2008). A minimum income standard for Britain: what people think, Joseph Rowntree Foundation. Available at: http://www.jrf.org.uk/publications/minimum-income-standard-britain-what-people-think (accessed 15 August 2015).

BRANDSHAW, J. AND HUTTEN, S. (1983). Social Policy Options and Fuel Poverty. Journal of Economic Psychology (3): 249–266.

BRAUBACH, M., JACOBS, D.E. AND ORMANDY, D. (2011). Environmental burden of disease associated with inadequate housing: A method guide to the quantification of health effects of selected housing in the WHO European Region. Copenhagen: World Health Organization, Regional Office for Europe.

BUNDESTAG, D. (2012). Antwort der Bundesregierung: Auf die Kleine Anfrage der Abgeordneten Bärbel Höhn, Markus Kurth, Daniela Wagner, weiterer Abgeordneter und der Fraktion BÜNDNIS 90/DIE GRÜNEN – Drucksache 17/10475 – Energiearmut erkennen und Lösungen anbieten. [Respond by the German Government regarding a proposal by the Green Party concerning the definition and measurement of Fuel Poverty in Germany].

BUNDESTAG, D. (2014). Antwort der Bundesregierung: Auf die Kleine Anfrage der Abgeordneten Bärbel Höhn, Christian Kühn (Tübingen), Oliver Krischer, weiterer Abgeordneter und der Fraktion BÜNDNIS 90/DIE GRÜNEN – Drucksache 18/174. Maßnahmen gegen steigende Heizkosten zur Bekämpfung der Energiearmut. [Respond by the German Government regarding a proposal by the Green Party concerning Energy Poverty in Germany].

BUZAR, S. (2007). Energy poverty in Eastern Europe: Hidden geographies of deprivation. Aldershot: Ashgate.

CAIRNEY, P. (2012). Understanding public policy: Theories and issues. Houndmills, Basingstoke, Hampshire, New York: Palgrave Macmillan.

COBB, R.W. AND ELDER, C.D. (1972). Participation in American politics: The dynamics of agenda-building. Baltimore: Johns Hopkins University Press.

DCV (2013). Eckpunkte und Position des DCV zur Bekämpfung von Energiearmut. [Strategies and opinions by the DCV to alleviate fuel poverty] Deutscher Caritas Verband [German Caritas Association].

DECC (2013). Fuel Poverty: A Framework for Future Action, Department of Energy & Climate Change 13D/149.

DECC (2014). The Fuel Poverty Statistics Methodology and User Manual, 15D/166. Department of Energy & Climate Change, London.

DIW (2010). Polarisierung der Einkommen: Die Mittelschicht verliert. [The polarisation of incomes: The middle class is suffering], Deutsches Insitut für Wirtschaftsforschung [German Insititut for Economic Research]. Available at: www.diw.de/documents/publikationen/73/diw_01.c.357505.de/10-24-1.pdf (accessed 10 July 2015).

DIW (2014). Vermögensverteilung in Deutschland. [Wealth distribution in Germany] Deutsches Insitut für Wirtschaftsforschung [German Insititut for Economic Research]. Available at: http://www.diw.de/documents/publikationen/73/diw_01.c.438708.de/14- 9.pdf (accessed 12 July 2015).

DÖSCHNER, J. (2015). EEG-Reform verfehlt zentrales Ziel: Mehr Unternehmen profitieren von Rabatten. [EEG-reform misses its central goals - More companies benefit from surcharge excemptions] ARD. Available at: https://www.tagesschau.de/wirtschaft/eeg- umlage-industrierabatte-101.html (accessed 10 August 2015).

DRYZEK, J.S. (1997). The politics of the earth: Environmental discourses. Oxford, New York: Oxford University Press.

DTI (2001). The UK fuel poverty strategy, Department for Trade and Industry. Available at: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/42608/1441 -govtresp-warm-front-eligibility.pdf (accessed 20 July 2015).

EESC (2010). Energy poverty in the context of liberalisation and the economic crisis, European Economic and Social Committee TEN/420. Available at: http://www.eesc.europa.eu/resources/ces990-2010_ac_en.doc (accessed 3 August 2015).

EPEE (2009). European Fuel Poverty and Energy Efficiency, European Commission - European Partnership for Energy and the Environment.

FITZPATRICK, T. (2011). Understanding the environment and social policy. Bristol, Portland, OR: Policy Press.

FPAG (2015). Fuel Poverty Advisory Group (for England): 12th annual report 2013-14, DECC. Available at: https://www.gov.uk/government/organisations/the-fuel-poverty- advisory-group (accessed 12 August 2015).

FRANK, D.A., NEAULT, N.B., SKALICKY, A., COOK, J.T., ET AL. (2006). Heat or eat: the Low Income Home Energy Assistance Program and nutritional and health risks among children less than 3 years of age. Pediatrics 118(5): 1293–1302.

GREEN, K. AND GILBERTSON, J. (2008). Warm front: better health: Health impact evaluation of the warm front scheme, Centre for Regional, Economic and Social Research. Available at: www.shu.ac.uk/research/cresr/sites/shu.ac.uk/files/warm-front-health- impact-eval.pdf (accessed 5 August 2015).

HALL, S. (1978). Policing the crisis: Mugging, the state, and law and order. London: Macmillan.

HARRINGTON, B.E., HEYMAN, B., MERLEAU-PONTY, N., STOCKTON, H., ET AL. (2005). Keeping warm and staying well: findings from the qualitative arm of the Warm Homes Project. Health & social care in the community 13 (3): 259–267.

HBS (2013). Hartz IV Regelsatz-Berechnung weiter fragwürdig. [Calculation of social benefit rates is still questionable], Hans-Böckler-Stiftung [Hans-Böckler-Foundation] Düsseldorf.

HEALY, J.D. (2004). Housing, fuel poverty, and health: A pan-European analysis. Aldershot, Burlington: Ashgate Pub.

HEFFNER, G. AND CAMPBELL, N. (2011). Evaluating the co-benefits of low-income energy-efficiency pro-grammes. IEA. Available at: www.iea.org/publications/freepublications/publication/low_income_energy_efficiency.pdf (accessed 13 August 2015).

HEIDENHEIMER, A.J. (1990). Comparative public policy: The politics of social change. 3rd ed. 3: Macmillan.

HEINDL, P. (2014). Measuring Fuel Poverty: General Considerations and Application to German Household Data, SOEP papers 632. DIW, Berlin.

HEINDL, P., SCHÜßLER, R. AND LÖSCHEL, A. (2014). Ist die Energiewende sozial gerecht? [The German energy transition from the perspective of social justice]. Wirtschaftsdienst 7: 508–514.

HILLS, J. (2012). Getting the measure of fuel poverty: Final Report of the Fuel Poverty Review, Centre for the analysis of social exclusion at the London School of Economics and Political Science, London.

HIRSCH, D., PRESTON, I. AND WHITE, V. (2011). Understanding fuel expenditure: Fuel poverty and spending on fuel, Centre for Sustainable Energy. Available at: http://www.consumerfocus.org.uk/files/2011/10/Understanding-fuel-expenditure.pdf (accessed 3 August 2015).

HOWDEN-CHAPMAN, P., VIGGERS, H., CHAPMAN, R., O’SULLIVAN, K., ET AL. (2012). Tackling cold housing and fuel poverty in New Zealand: A review of policies, research, and health impacts. Energy Policy 49: 134–142.

HUBY, M. (1998). Social policy and the environment. Buckingham, Philadelphia: Open University Press.

IFEU (2010). Kommunale Regelungen zur Heizkostenübernahme bei ALG-II- und Sozialhilfehaushalten: Bestandsaufnahme, Handlungsmöglichkeiten und Praxishilfen, Institut für Energie- und Umweltforschung. Endbericht. Available at: http://www.ifeu.de/energie/pdf/Endbericht_Angemessenheit_Heizkosten_IFEU_final.pdf (accessed 12 August 2015).

ISHERWOOD, B. AND HANCOCK, R. (1979). Household Expenditure on Fuel: Distributional Aspects, Economic Adviser’s Office.

JONES, B.D. AND BAUMGARTNER, F.R. (2005). The politics of attention: How government prioritizes problems. Chicago: University of Chicago Press.

KEATINGE, W.R. (1986). Seasonal mortality among elderly people with unrestricted home heating. British Medical Journal 293: 732–733.

KEATINGE, W.R. (2001). Winter Deaths: Warm Housing is not enough. British Medical Journal 323: 166–167.

KINGDON, J.W. (1995). Agendas, alternatives, and public policies. 2. Ed. New York, Harper Collins College Publ.

KNILL, C. AND TOSUN, J. (2012). Public policy: A new introduction. New York: Palgrave Macmillan.

KOPATZ, M. (2013). Energiewende. Aber fair! Wie sich die Energiezukunft sozial tragfähig gestalten lässt. [How to design a socially acceptable energy transformation for the future]. 1. Aufl. München: Oekom Verlag.

KOPATZ, M., SPITZER, M. AND CHRISTANELL, A. (2000). Energiearmut: Stand der Forschung, nationale Programme und regionale Modellprojekte. [Energy poverty, current research projects and regional models], Wuppertal Institut für Klima, Umwelt, Energie GmbH. Available at: www.epub.wupperinst.org/files/3606/WP184.pdf (accessed 12 August 2015).

LIDDELL, C. (2008). The impact of fuel poverty on children, Safe the Children. Policy Brief.

LIDDELL, C. (2012). Fuel poverty comes of age: Commemorating 21 years of research and policy. Editorial. Energy Policy 49: 2–5 (accessed 7 August 2015).

LIDDELL, C. AND MORRIS, C. (2010). Fuel poverty and human health: A review of recent evidence: The Role of Trust in Managing Uncertainties in the Transition to a Sustainable Energy Economy, Energy Policy 38(6): 2987–2997.

LIDDELL, C., MORRIS, C., MCKENZIE, S. AND RAE, G. (2012). Measuring and monitoring fuel poverty in the UK: National and regional perspectives. Energy Policy: 27–32.

LIJPHART, A. (1971). Comparative politics and the comparative method. American political science review 65(3): 682–693.

LOR, P. (2011). International and Comparative Librarianship: Methodology in comparative studies. Available at: https://pjlor.files.wordpress.com/2010/06/chapter-4-draft-2011-04- 20.pdf (accessed 5 July 2015).

MAY, J.V. AND WILDAVSKY, A.B. (1978). The Policy cycle. Beverly Hills: Sage Publications. 37

MILL, J.S. (1862). A system of logic, ratiocinative and inductive: Being a connected view of the principles of evidence, and the methods of scientific investigation. Vol. 1. 5th ed. 5. London: Parker, Son, and Bourn.

MOORE, R. (2012). Definitions of fuel poverty Implications for policy. Energy Policy 49: 19– 26.

OECD (2011). Divided We Stand: When Inequality Keeps Rising. OECD Publishing.

OSBALDESTON, J. (1984). Fuel poverty in UK cities. Cities 1(4): 366–373.

PALMER, G., MACINNES, T. AND KENWAY, P. (2008). Cold and Poor: An Analysis of The Link between Fuel Poverty and Low Income, New Policy Institute. Available at: www.poverty.org.uk/reports/fuel%20poverty.pdf (accessed 13 August 2015).

PRESTON,I.,MOORE,R.,GUERTLER,P. (2008). How much? The costs of alleviating fuel poverty. Report to the EAGA Partnership Charitable Trust. Newcastle Upon Tyne.

PRINCEN, S. AND RHINARD, M. (2006). Crashing and creeping: agenda setting dynamics in the European Union. Journal of European Public Policy (13:7): 1119–1132.

PROIDL, H. (2009). E-Control & Caritas – Pilotprojekt „Energieberatungen von einkommensschwachen Haushalten“. [Energy consultion provision for low-income households]. Endbericht.

RAGIN, C.C. (1987). The comparative method: moving beyond qualitative and quantitative strategies. Berkeley (CA): University of California Press.

RAHO, A. (2012). Fuel poverty related policy: Lessons learned in the UK and other European countries and potentials and possible challenges for fuel poverty policy implementation in Austria. Master-Thesis, University of Exeter. Exeter.

ROBERTS, S. (2008). Energy, equity and the future of the fuel poor. Energy Policy 36 (12): 4471–4474.

RYAN, L. AND CAMPBELL, N. (2012). Spreading the net: The multiple benefits of energy efficiency improvements, OECD / IEA. Available at: http://www.iea.org/publications/insights/insightpublications/Spreading_the_Net.pdf (accessed 11 July 2015).

SCHULTZ, S. (2014). Energiearmut: Im dunklen Deutschland, Der Spiegel. Available at: http://www.spiegel.de/wirtschaft/soziales/energiearmut-reportage-ueber-menschen-ohne- stromversorgung-a-964258.html (accessed 10 July 2015).

SELL, S. (2015). Hartz IV: Teurer Strom, Energiearmut und das ewige Pauschalierungsdilemma. [HartzIV: Energy poverty and the issue of sufficient cost coverage] Hochschule Koblenz. Available at: http://aktuelle- sozialpolitik.blogspot.de/2015/02/16-stromkosten-hartz4.html (accessed 2 August 2015).

SEN, A. (1985). Commodities and capabilities. Amsterdam, New York: North-Holland.

SUNDERLAND, L., CROFT, D. AND GUERTLER, P; WADE, J. (2011). National energy efficiency and energy saving targets, European Council for an Energy Efficient Economy. Available at: hpaba.com/pages/en/energy%20efficiency2.pdf (accessed 10 August 2015).

TEWS, K. (2013). Energiearmut definieren, identifizieren und bekämpfen. Eine Herausforderung der sozialverträglichen Gestaltung der Energiewende. Vorschlag für eine Problemdefinition und Diskussion des Maßnahmenportfolios. [Identifying, defining and fighting energy poverty. A challenge for a socially acceptable energy transformation], Free University of Berlin. Available at: http://www.polsoz.fu- berlin.de/polwiss/forschung/systeme/ffu/aktuelle-publikationen/13-tews- energiearmut/index.html (accessed 12 July 2015).

THOMSON, H. (2003). Health impact assessment of housing improvements: Incorporating research evidence. Journal of Epidemiology & Community Health 57 (1): 11–16.

UNDP (2012). Fuel Poverty: "Improving Energy Efficiency in Low-Income Households and Communities in Romania".

ÜRGE-VORSATZ, D. AND TIRADO HERRERO, S. (2012). Building synergies between climate change mitigation and energy poverty alleviation. Fuel Poverty Comes of Age: Commemorating 21 Years of Research and Policy. Energy Policy 49: 83–90.

VZ NRW (2011). Energie-Grundversorgungsunternehmen – Energieschulden und präventive Lösungsansätze. [Energy-Supplier Survey: Energy debt issues and solution proposals], Verbraucherzentrale Nordrhein-Westfalen [Agency for Consumer Protection]. Available at: http://www.vz-nrw.de/evu-umfrage (accessed 12 July 2015).

WILKINSON, P., SMITH, K.R., BEEVERS, S., TONNE, C., ET AL. (2007a). Energy, energy efficiency, and the built environment. The Lancet (9593): 1175–1187.

WILKINSON, P., SMITH, K.R., JOFFE, M. AND HAINES, A. (2007b). A global perspective on energy: Health effects and injustices. The Lancet (9591): 965–978.