Tijn Croon

The in-depth study of energy poverty as a social phenomenon commenced in the late 19th century through the works of British social researchers Booth and Rowntree (O’Connor, 2016). This era was characterised by significant social and economic transformation, and these scholars were troubled by the living conditions and welfare of impoverished urban populations, who were residing in congested and unsanitary environments. Throughout the 20th century, poverty in policy contexts became quite narrowly defined as a lack of income. However, it was another social concern in the UK that led to the development of concepts like ‘fuel poverty’ or ‘energy poverty’ a century after Booth and Rowntree.[i] Following the 1973 oil crisis, the Child Poverty Action Group took the initiative to address how increasing energy costs were affecting low-income households in the UK (Johnson & Rowland, 1976). As essentials like heating, electricity, and fuels became necessary for maintaining a decent standard of living in modern British society, this advocacy group pushed for government financial support. Later, Bradshaw and Hutton (1983) introduced a narrower definition of energy poverty: “the inability to afford adequate heat in the home”. Since then, studies on energy poverty have typically excluded motor fuels, as they fall under transport poverty, a related but separate area of study (Mattioli et al., 2017). Energy poverty, as defined by Bouzarovski and Petrova (2015, p. 33), refers to "the inability to secure or afford sufficient domestic energy services that allow for participation in society." Although the precise boundaries of relevant domestic energy usage are still debated, this definition expands beyond mere heating as it encompasses energy used for cooling, which is particularly relevant in warmer climates (Thomson et al., 2019). Moreover, it enables a socially and culturally dependent understanding of what it means to participate in society (Middlemiss et al., 2019). On 13 September 2023, the European Union (2023) officially defined energy poverty as “a household’s lack of access to essential energy services, where such services provide basic levels and decent standards of living and health, including adequate heating, hot water, cooling, lighting, and energy to power appliances, in the relevant national context, existing national social policy and other relevant national policies, caused by a combination of factors, including at least non-affordability, insufficient disposable income, high energy expenditure and poor energy efficiency of homes”. The doctoral thesis and subsequent book by Brenda Boardman, Fuel Poverty: From Cold Homes to Affordable Warmth (1991), marked a significant breakthrough in energy poverty research. She emphasised the detrimental impact of energy-inefficient housing on health and quality of life. In the decades that followed, substantial literature confirmed her qualitative findings (Thomson et al., 2017). Notably, studies have demonstrated the adverse effects of living in energy poverty on physical health (Liddell & Morris, 2010), mental health (Liddell & Guiney, 2015), stress levels (Longhurst & Hargreaves, 2019), social isolation (Harrington et al., 2005), and absenteeism (Howden-Chapman et al., 2007). Boardman’s work introduced an indicator that has remained influential to this date, although it was not the first attempt to operationalise the concept of fuel poverty (Isherwood & Hancock, 1979). Her ‘2M’ indicator categorises a household as energy poor if it needs to allocate twice the median share of its budget for energy expenses to heat its home adequately. Boardman calculated this threshold to be 10% at that time. Due to its simplicity and ease of comprehension, many governments directly adopted this 10% threshold without considering specific contextual circumstances. Since the early nineties, numerous attempts have been made to develop alternative indicators. Highly influential ones include ‘Low Income High Cost’ (LIHC) by John Hills (2012), ‘Low Income Low Energy Efficiency’ (LILEE) that subsequently became the official British indicator (BEIS, 2022), and a 'hidden' energy poverty indicator by (Meyer et al., 2018). Critiques of these indicators focus, amongst other things, on their simplicity and perceived 'technocratic' approach (Croon et al., 2023; Middlemiss, 2017). This marked the beginning of significant government commitment, initially in the UK and later in other countries to address energy poverty. Although certain forms of cold weather payments had already been introduced by the UK's Conservative administrations, it was under the successive governments of Blair and Brown, following the publication of Boardman's work, that programmes such as the Winter Fuel Payment and Warm Home Discount were implemented (Koh et al., 2012). The UK examples highlight bipartisan support for addressing energy poverty, with both the Conservatives and Labour backing these efforts. This policy objective has also gained momentum in various legislative contexts, leading the EU to incorporate energy poverty alleviation as a fundamental pillar of the European Green Deal and a specific goal of its landmark Social Climate Fund (European Commission, 2021). Over the last three decades, public interest in energy poverty as a 'wicked' problem has surged, particularly during the recent energy crisis. This crisis began in 2021 when energy markets tightened due to a post-pandemic economic rebound, and it worsened dramatically after Russia's invasion of Ukraine in February 2022 (IEA, 2023). Extensive research on the impact of this price surge on energy poverty levels has been carried out throughout Europe and globally (Guan et al., 2023; Simshauser, 2023). Consequently, energy poverty has become a significant focal point in discussions related to the 'just transition,' especially within the realm of energy justice, as it serves as a valuable concept for targeting policies towards a specific vulnerable group in this context (Carrosio & De Vidovich, 2023).     [i] ‘Fuel poverty' and 'energy poverty' are used interchangeably, with the former being more common in the UK and the latter in mainland Europe (Bouzarovski & Petrova, 2015). Previously, scholars in the UK used 'energy poverty' to denote a lack of access to energy and 'fuel poverty' when affordability was the concern (Li et al., 2014). However, this distinction is no longer maintained.

The shift from ‘government’ to ‘governance’ has been debated since the early 1970s. Whilst state interventionism had been widely embraced within western societies during the post-war decades, governments gradually moved from exercising constitutional powers to acting as facilitators and cooperative partners (Rhodes, 1997). Over the course of a few decades, this resulted in governance as ‘interactive social-political forms of governing’ (Nag, 2018, p. 124).  Hira and Cohn (2003, p. 12), influenced by Keohane (2002), define governance as “the processes and institutions, both formal and informal, that guide and restrain the collective activities of a group”. Its decentralised and flexible nature could still include public actors but would also leave space for private and third-sector parties to provide services in hybrid and temporary institutional arrangements. To formulate one single definition of ‘housing governance’ as a particular mode of governance is however difficult due to its multilevel character. Housing could relate to either a family home, a housing association, or a complete local/national housing governance framework. On a household level, Wotschack (2005, p. 2) defines governance as managing “the daily time allocation of spouses by household rules and conflict handling strategies”. The work of Wijburg (2021) indicates that local/municipal governance entails a set of public interventions, strategies, policies and provisions used to provide local needs (e.g. housing supply). On the national level, Yan et al. (2021) define public rental housing (PRH) governance as “a structure of a wide range of government and non-governmental actors that act in all its phases of PRH provision from policy design to implementation and realisation”.[1] This specific definition on PRH combines the domestic definition of governance with Wijburg’s understanding of governance on the local level. Within the Chinese context, the national government provides policies and creates nationwide operational methods, whilst local governments implement and formulate the policies locally (Yan et al., 2021). Critics point out that a more decentralised governance structure complicates the public accountability of housing provision. Peters and Pierre (2006, p. 40) distinguish problems concerning the ‘isolation’ and ‘enforcement’ of accountability. The former refers to demarcation, as it is easier to measure the performance of a government housing agency directly responsible for new build and operations, than those from the private sector in an indirect role trying to stimulate and facilitate other actors and contracting out construction and operations (Shamsul Haque, 2000). The latter relates to the accountability deficit that arises when responsibility is transferred from democratically governed municipal agencies to actors without a representative institutional arrangement, and thus without control mechanisms for tenants or the wider population (Mullins, 2006). Throughout history, understanding of governing has evolved together with the role of government. The state plays a different role in capitalism, corporatism and socialism, which has varying effects on local and/or (inter)national levels. Whilst the above paragraphs describe housing governance within a democratic governance regime, transferring the conceptual debate to autocratic or hybrid regimes would pose difficulties. Thus, finding a unique definition of housing governance applicable in all spheres remains a challenge, and the specific context must be carefully considered. Important challenges remain, and as housing provision mechanisms evolve, further exploration of housing governance, especially on a municipal level, are likely to gain importance (Hoekstra, 2020). [1] “Housing provision is a physical process of creating and transferring a dwelling to its occupiers, its subsequent use and physical reproduction and at the same time, a social process dominated by the economic interests involved” ibid.

Justice theory is as old as philosophical thought itself, but the contemporary debate often departs from the Rawlsian understanding of justice (Velasquez, Andre, Shanks, & Meyer, 1990). Rawls (1971) argued that societal harmony depends on the extent to which community members believe their political institutions treat them justly. His First Principle of ‘justice as fairness’ relates to equal provision of ‘basic liberties’ to the population. His Second Principle, later referred to as the ‘Difference Principle’, comprises unequal distribution of social and economic goods to the extent that it benefits “the least advantaged” (Rawls, 1971, p. 266).1[1] As this notion added an egalitarian perspective to Rawlsian justice theory, it turned out to be the most controversial element of his work (Estlund, 1996). The idea of a ‘just transition’ was built on these foundations by McCauley and Heffron (2018), who developed an integrated framework overarching the ‘environmental justice’, ‘climate justice’ and ‘energy justice’ scholarships. The term was first used by trade unions warning for mass redundancies in carbon-intensive industries due to climate policies (Hennebert & Bourque, 2011), but has acquired numerous interpretations since. This is because the major transition of the 21st century, the shift towards a low-carbon society, will be accompanied by large disturbances in the existing social order. In this context, a just transition would ensure equity and justice for those whose livelihoods are most affected (Newell & Mulvaney, 2013). A just transition implies that the ‘least advantaged’ in society are seen, heard, and compensated, which corresponds with three key dimensions conceptualised by Schlosberg (2004): distributive, recognitional, and procedural justice. Distributive justice corresponds with Rawls’ Difference Principle and comprehends the just allocation of burdens and benefits among stakeholders, ranging from money to risks to capabilities. Recognitional justice is both a condition of justice, as distributive injustice mainly emanates from lacking recognition of different starting positions, as well as a stand-alone component of justice, which includes culturally or symbolically rooted patterns of inequity in representation, interpretation, and communication (Young, 1990). Fraser (1997) stressed the distinction between three forms: cultural domination, nonrecognition (or ‘invisibility’), and disrespect (or ‘stereotyping’). Procedural justice emphasises the importance of engaging various stakeholders – especially the ‘least advantaged’ – in governance, as diversity of perspectives allows for equitable policymaking. Three elements are at the core of this procedural justice (Gillard, Snell, & Bevan, 2017): easily accessible processes, transparent decision-making with possibilities to contest and complete impartiality. A critique of the just transition discourse is that it preserves an underlying capitalist structure of power imbalance and inequality. Bouzarovski (2022) points to the extensive top- down nature of retrofit programmes such as the Green New Deal, and notes that this may collide with bottom-up forms of housing repair and material intervention. A consensus on the just transition mechanism without debate on its implementation could perpetuate the status quo, and thus neglect ‘diverse knowledges’, ‘plural pathways’ and the ‘inherently political nature of transformations’ (Scoones et al., 2020). However, as Healy and Barry (2017) note, understanding how just transition principles work in practice could benefit the act of ‘equality- proofing’ and ‘democracy-proofing’ decarbonisation decisions. Essentially, an ‘unjust transition’ in the context of affordable and sustainable housing would refer to low-income households in poorly insulated housing without the means or the autonomy to substantially improve energy efficiency. If fossil fuel prices – either by market forces or regulatory incentives – go up, it aggravates their already difficult financial situation and could even lead to severe health problems (Santamouris et al., 2014). At the same time, grants for renovations and home improvements are poorly targeted and often end up in the hands of higher income ‘free-riding’ households, having regressive distributional impacts across Europe (Schleich, 2019). But even when the strive towards a just transition is omnipresent, practice will come with dilemmas. Von Platten, Mangold, and Mjörnell (2020) argue for instance that while prioritising energy efficiency improvements among low-income households is a commendable policy objective, putting them on ‘the frontline’ of retrofit experiments may also burden them with start-up problems and economic risks. These challenges only accentuate that shaping a just transition is not an easy task. Therefore, both researchers and policymakers need to enhance their understanding of the social consequences that the transition towards low-carbon housing encompasses. Walker and Day (2012) applied Schlosberg’s dimensions to this context. They conclude that distributive injustice relates to inequality in terms of income, housing and pricing, recognitional justice to unidentified energy needs and vulnerabilities, and procedural injustice to inadequate access to policymaking. Ensuring that the European Renovation Wave is made into a just transition towards affordable and sustainable housing therefore requires an in-depth study into distributive, recognitional and procedural justice. Only then can those intertwining dimensions be addressed in policies.   [1] To illustrate his thesis, he introduces the ‘veil of ignorance’: what if we may redefine the social scheme, but without knowing our own place? Rawls believes that most people, whether from self-interest or not, would envision a society with political rights for all and limited economic and social inequality.  

The concept ‘prebound effect’ refers to the phenomenon where actual energy consumption in buildings is significantly lower than the calculated energy consumption needed to maintain comfortable indoor temperatures (Sunikka-Blank & Galvin, 2012). This is typically observed in older, less energy-efficient homes, where occupants often under-heat their homes due to financial constraints, leading to lower actual energy usage than predicted by models (Galvin & Sunikka-Blank, 2016). This contrasts with the more commonly used term ‘rebound effect’, which occurs when energy consumption increases following efficiency improvements, offsetting some of the anticipated savings (Teli et al., 2015). The difference between the two is clearly visualised in Figure 1, developed by Sunikka-Blank and Galvin (2012). Energy savings discrepancies The prebound effect leads to significant discrepancies in energy savings estimates for retrofitting projects. Standard models often overestimate energy savings because they do not account for the lower baseline consumption caused by the prebound effect. This discrepancy is crucial because it implies that thermal retrofits may not yield the anticipated reductions in energy use and carbon emissions, which are often overstated in policy and financial analyses (Galvin, 2023). A detailed explanation of the performance gap is available in the vocabulary entry by Furman (2024). For instance, in Hungary, research shows that energy consumption predictions are often based on technical data alone, ignoring actual under-consumption behaviours, leading to overestimates in expected savings (Gróf et al., 2022). Similarly, in Italy, studies have highlighted that prebound effects cause a gap between theoretical energy models and actual energy usage, complicating the accurate forecasting of energy savings from retrofits (Giuliani et al., 2016). The study by Van den Brom et al. (2019) also supports this, showing that building characteristics and occupant behavior significantly influence energy consumption, with discrepancies between expected and actual savings due to prebound effects being notable in both the Netherlands and Denmark. Additionally, Gróf et al. (2022) emphasised that the prebound effect is influenced by the financial status of households, with households having limited financial means more likely to under-heat their homes, resulting in a greater prebound effect and further complicating the prediction of energy savings. Inherently linked to energy poverty Therefore, the prebound effect is closely linked to energy poverty, as it often reflects the behaviour of households that cannot afford to heat their homes adequately. Households experiencing fuel poverty are more likely to exhibit high prebound effects due to their efforts to save money on heating (Galvin & Sunikka-Blank, 2016). For instance, social housing tenants in the UK showed significant prebound effects, leading to under-heated homes and reduced energy consumption prior to retrofitting (Teli et al., 2015). Addressing the prebound effect involves recognising and mitigating the financial constraints that lead to under-consumption, thus improving both the accuracy of energy savings predictions and the living conditions of households experiencing energy poverty (Galvin, 2024a, 2024b). Moreover, hidden energy poverty is a critical aspect of this issue. Households often are not categorised as energy poor based on expenditure-based indicators because they do not dare to put their heating on due to financial constraints (Cong et al., 2022; Eisfeld & Seebauer, 2022). These households minimise their heating to the point where their energy expenditures are deceptively low, masking the severity of their situation. However, more recent indicators that focus on energy efficiency rather than high costs would identify them as energy poor (Betto et al., 2020). Addressing hidden energy poverty requires a shift in assessment criteria, also focusing on the energy efficiency of homes rather than just the high cost of energy bills. This approach ensures that households struggling with inadequate heating due to financial constraints are accurately identified and supported (Croon et al., 2023).

Contemporary scholars generally accept the multidimensional understanding of sustainability - social, political, economic, cultural and environmental amongst other dimensions – but the concept used to be defined more narrowly as the ‘conservation of natural resources’ and the ‘restoration of ecological balance’ (Meadows et al. 1972). While the ‘Brundtland Report’ was instrumental in broadening the definition and bridging the environmental and economic dimensions (WCED 1987), it was Elkington who stressed the social dimension in the ‘triple bottom line’ of ‘people, planet, profit’ (1998). However, the role of community participation as an elementary part of social sustainability was only established after the turn of the millennium by Giddings et al. (2002). They emphasised the participation aspect of procedural equity “so that people are able to shape their own futures” (ibid., p.194). Dempsey et al. (2011) drew upon this contribution when they considered urban sustainability from a community approach and concluded that communities thrive upon social interaction between community members, organisational initiative through collective groups and networks, the relative stability of a neighbourhood in terms of net migration and turnover, a positive identification or sense of place and the level of trust that follows from a perception of safety. These factors are summarised by Dixon and Woodcraft (2013, p.475) as “the extent to which a neighbourhood supports individual and collective well-being (…) It combines design of the physical environment with a focus on how the people who live in and use a space relate to each other and function as a community”. While most community participation researchers look into social sustainability on the neighbourhood level, Putnam’s book ‘Bowling alone’ (2000) described how a lack of social capital, here understood as strong civic participation and localised empowerment, could prevent collective action and undermine democracy on the macro-level. 

The scholarly discourse on targeting versus universalism in social protection has significantly influenced the evolution of global welfare states over the past century and a half. Recently, surging energy prices across Europe have reignited the debate over choosing between universal household support or targeted relief, mirroring a similar discussion on the trade-offs between efficiency and equity in providing renovation subsidies. This vocabulary entry delves into the advantages and disadvantages of each approach, while exploring various attempts to find middle ground between the two (e.g., ‘targeted universalism’), particularly in the context of affordable and sustainable housing. Historic context In the 1880s, Germany established one of the earliest forms of the modern welfare state with its Bismarckian system (Manow, 2020). It introduced social insurance programmes to address workers' issues such as healthcare, insurance, and pensions for the elderly. While revolutionary for its time, the system was modest in scope and initially limited to workers and their dependents. In the decades that followed, social welfare programmes across Western Europe expanded to cover a wider range of risks, gradually incorporating non-working populations into social protection schemes. The Great Depression catalysed the recognition of the state's crucial role in social welfare, a stance prominently advocated by economists such as Keynes. Following WWII, there was a widespread consensus on the importance of improving working-class living conditions to avert future disasters. The Beveridge Report, authored by Sir William Beveridge (1942), played a crucial role in reshaping the British welfare state and influencing welfare policies worldwide. It proposed an expansive social security system with universal coverage in areas like health and unemployment insurance. Many of the Beveridge Report's recommendations were implemented in the UK and significantly shaped the post-war social landscape after WWII under the government of Clement Attlee (Reeves & McIvor, 2014). Although the Report appeared to favour universalism, it did incorporate means-tested elements for specific benefits, creating a nuanced approach that combined universalism with targeted support to address poverty and meet specific needs. In the latter half of the 20th century, the evolution of different welfare states led to varying housing outcomes. The UK's liberal regime shifted to a more targeted dualist rental system with an increasing focus on homeownership, in contrast to corporatist and particularly social democratic regimes with a unitary rental system that aligned more with universalism (Esping-Andersen, 1990; Kemeny, 2001). Interestingly, housing presented a unique challenge in Beveridge's pursuit of universalism. He dedicated nine pages to the 'problem of rent', acknowledging significant geographical variations in rent levels while recognising that they were often not a matter of choice (Lund et al., 2021). Beveridge considered covering actual rent costs but realised this conflicted with his principle of flat-rate benefits. His solution was to recommend a national average rent allowance, differing only for working and pensioner households. Given the rent disparities, larger families' higher rent needs, and poor housing conditions, Beveridge's plan risked falling short of providing universal subsistence. To address this, he suggested post-war housing improvements in later publications (Beveridge, 1949, 1952), including increased supply and better quality of housing in order  to reduce rent disparities across the UK. He advocated for the creation of New Towns and incentivising voluntary housing associations, hoping that enhancing housing quality would lead to  improvements in relative affordability. Advantages and disadvantages of approaches Beveridge’s ‘problem of rent’ mirrors a core dilemma central to the debate on universalism and targeting. While a universal approach aims for inclusivity by covering all households, ensuring that no one is left out, the support offered in a universal system may not be as substantial as in a targeted approach. This could potentially fall short of providing adequate relief to those with the most significant needs. The universalist approach offers numerous advantages (Thompson & Hoggett, 1996). It not only promises inclusivity but also enjoys greater public acceptance. Furthermore, its simplicity in administration streamlines implementation and reduces administrative complexities. By eliminating the need for means testing or eligibility criteria, it simplifies the delivery of benefits. However, universalism also comes with its share of drawbacks. The cost implications of providing benefits universally can be substantial, potentially straining government budgets, without necessarily providing sufficient support to those in need. Additionally, responding to crises and price shocks with universal support, often referred to as ‘helicopter money,’ can cause substantial inflationary pressure. In the context of affordable and sustainable housing, there is a particular disadvantage to this approach. Universal energy subsidies may diminish the incentive for high-income homeowners, who typically spend the most on energy, to reduce consumption or invest in energy efficiency (Lausberg & Croon, 2023). Simultaneously, low-income households in energy-inefficient dwellings may hesitate to use heating due to leakages and financial concerns (Betto et al., 2020), potentially making them ineligible for support. In his influential essay on the ‘Political Economy of Targeting’, Amartya Sen (1998) describes the pros and cons of targeting. A major benefit is cost-effectiveness, directing resources efficiently to the most vulnerable and ensuring maximum impact for the targeted groups. However, the approach is impaired by potential exclusion errors, wherein genuinely disadvantaged households might be overlooked due to stringent eligibility criteria. Additionally, the stigma associated with being a beneficiary can lead to negative social consequences. Furthermore, the administrative complexity involved in identifying and reaching the right households is a significant hurdle, requiring comprehensive and accurate systems. This final complication is especially evident in the government's response to the energy crisis in Europe, as policymakers have cited a shortage of data, time constraints, and rigid social compensation mechanisms to explain why they could not effectively assist those who are most vulnerable (Natili & Visconti, 2023). Finding middle ground Universal and targeted approaches coexist in most social welfare systems (Jacques & Noël, 2021). Many European countries have implemented a universally accessible pension system, ensuring entitlement to benefits for everyone of a certain age, supplemented by targeted support for elderly individuals lacking private income. Moreover, several European countries approach housing renovation in a comparable manner, offering general subsidies to all residents while also implementing means-tested subsidies for households considered vulnerable. Another branch of public policy emphasises ‘targeted universalism,’ focusing on designing support policies and interventions that prioritise and address the specific needs of disadvantaged households while still providing benefits to everyone (Powell et al., 2019). In other words, these policies are available to all but particularly important to some. One example would be the reduction in the cost of public transport, a service that is available to everyone but predominantly used by households with lower incomes. Other examples include childcare subsidies and job training programmes, which are particularly beneficial for low-income families struggling to access quality childcare and for disadvantaged individuals looking to improve their job prospects, respectively (Coote & Percy, 2020). While the European Commission places a strong emphasis on targeting in its criteria for funding from the Social Climate Fund (SCF), Member States could consider strategies to implement policies that embody the principles of targeted universalism. For instance, in many Western European countries, a potential approach could involve prioritising the renovation of social housing. Given that energy poverty is prevalent among social tenants in these housing regimes, it would be logical to extend financial support or fiscal incentives to their landlords, such as tax deductibility of renovation costs or government guarantees for low-interest loans (Seebauer et al., 2019). On the other hand, the majority of energy poor households in the Mediterranean and Eastern Europe own their homes. In this context, an effective strategy that combines targeted and universalistic elements would entail investing in ‘one-stop-shops’ that offer comprehensive renovation services (Bertoldi et al., 2021). These establishments would provide bundled services such as energy audits, renovation works, and financing. By leveraging economies of scale and specialised expertise, they can effectively reduce costs and offer affordable renovation solutions to disadvantaged groups. Although one-stop-shops are accessible to all, they can be targeted at households experiencing specific information deficits, such as those confronted with language barriers, digital illiteracy, or limited expertise within their social networks. The EU explicitly states that Member States can use the SCF for “targeted, accessible and affordable information, education, awareness and advice on cost-effective measures and investments, available support for building renovations and energy efficiency” (European Parliament, 2023, p. 16)