A new definition of fuel poverty in Scotland: review of recent evidence

Executive Summary

On 24 October 2016, the Scottish Fuel Poverty Strategic Working Group ( FPSWG) published a report entitled A Scotland Without Fuel Poverty is a Fairer Scotland . This included the recommendation that a review of the current definition of fuel poverty in Scotland should be commissioned, in light of concerns that the current definition is too broad and impedes targeting assistance towards those in most need.

The Scottish Government accepted this recommendation and, in early 2017, established a Panel of four independent academics

• to conduct a review of the current definition of fuel poverty in use in Scotland;
• to make evidence-based recommendations for whether the definition should be retained;
• and if not, to indicate any changes that should be made.

This report comprises the result of the Panel's deliberations.

The Panel had five months in which to complete a Review, and we hope that it is read in that context. By contrast, Boardman's 1991 definition of fuel poverty took more than three years to develop. The LIHC indicator developed by John Hills in 2012 had a 12 month timeline, as did the 2016 report by the FPSWG.

Chapter 1 lays out the scope and remit of the task which the Scottish Government asked the Panel to address. It begins with a near-verbatim description of the remit we were given. Drawing on the FPSWG's work in 2016, as well as that of the Scottish Rural Fuel Poverty Task Force (which also delivered a report in 2016), we explored whether a rebalance of focus in the definition was required; for example whether the definition might need shifting towards a greater emphasis on its interrelationship with poverty and deprivation. We concluded that some of the adverse outcomes associated with fuel poverty were at risk of being de-emphasised in the increasing policy focus on energy efficiency and building fabric. While still recognising the multifaceted nature of the issue, and the range of relevant policy concerns, we also concluded that these adverse outcomes belonged at the heart of how fuel poverty should be defined in Scotland. We were aware too that a great deal of high-quality evidence had been published in recent months, and that this was not part of the evidence which FPSWG or the Scottish Rural Fuel Poverty Task Force had been able to take into account. The Panel felt that this evidence merited detailed scrutiny, and also full representation. For that reason, and as a consequence of the broad scope of our remit, the Review is lengthy.

Chapters 2 to 8 contain our analysis of the evidence which underpins the development of our thinking. In selecting evidence to include, the Panel were in agreement that the debate about fuel poverty has been particularly fast-moving in the past 3 years. For that reason, the Review is dominated by data and evidence published between 2015 and June 2017, and the Panel has relied heavily on this when reaching its conclusions.

That being said, on occasion the Panel also examined historical evidence, particularly in the context of indoor temperatures. As the Review indicates, and contrary to what many have surmised, there was a great deal of early thinking and discussion around indoor temperature regimes, dating back to 1936. A strong inter-disciplinary consensus can be seen in this early work. We viewed the long timeline of evidence as very important when reaching decisions about what should constitute an affordable heating regime.

Chapter 2 considers the content and role of definitions, and gives an account of the different ways in which fuel poverty has been defined in Europe, with particular emphasis on the UK. It illustrates the broad range of definitions which have been in play since the millennium:

• some of these have focused on calculating the how many and who domains of fuel poverty (the technical definitions);
• whilst others have tried to interpret the meaning and significance of living in fuel poverty (the consensual definitions).

The latter are definitions largely based on how ordinary people respond when asked if they are experiencing fuel poverty, and if so what the lived experience of being fuel poor is like. The Panel has given equal weight to the contributions made by both, but saw consensual definitions as increasingly vital complements to the more longstanding technical definitions which focus on energy costs, income, and thresholds.

In this context, the Panel felt that the consistency with which studies have reported mental wellbeing benefits associated with alleviating fuel poverty lent strong support to the argument that being fuel poor leads to discomfort, stress and anxiety. Encompassing hardship and burden, we recognised, would draw the issue of fuel poverty more squarely into a socio-political arena in which energy justice and equality would have a greater presence. This, we also recognised, would have the potential to reframe the concept in quite profound ways.

Chapter 3 explores the additional and sometimes confusing issue of Vulnerability. Part of the confusion which surrounds this term stems from the different ways it is used:

• ' households that are vulnerable to fuel poverty' can mean households likely to be in it;
• 'vulnerable households' sometimes refer to households that have been unable to develop the capacities for avoiding fuel poverty, and are therefore less likely to be in a position to avoid being fuel poor;
• at other times, ' vulnerable households' refers to the collective of households where someone has a health condition or disability that makes them especially prone to suffering the adverse effects of being fuel poor - these include people with limited mobility, cardiovascular and respiratory conditions, as well as people with dementia.

The Panel concluded that this last-mentioned definition, namely vulnerable to the adverse health outcomes of fuel poverty, should ideally prevail in the Scottish context. In this way, the focus on fuel poverty as a condition of hardship and burden could be sustained, and the protection of human health and wellbeing could remain central to how fuel poverty was addressed in policy and practice.

There was, however, also merit in retaining the idea that energy-related skills and capabilities could help protect households from falling into fuel poverty, although perhaps through the use of a term other than ' vulnerable'. In this way, the broader toolkit for alleviating fuel poverty (encompassing energy advice, support in tariff-switching, debt management, etc.) could become more central to how fuel poverty programmes were designed and delivered.

As a consequence of the lack of scientific consensus about health conditions and age groups most vulnerable to adverse impacts of fuel poverty, the Panel recommended that a short piece of further work on vulnerability criteria should be undertaken, as an integral, focused, component of planned Government consultation. The work should be done by a specialist group representing public health practitioners, local health and social care partnerships, and the social security team. The terms of reference should be narrow, so that the group confines its deliberations to the issues related to vulnerability as these affect a definition of fuel poverty.

Also in Chapter 3, the Review Panel sets out the working assumptions made in relation to age groups, long term ill health and disabilities, and the resulting adjustments made to recommended indoor temperatures and Minimum Income Standards. The additional work of the practitioner group should test the validity and robustness of these assumptions, consider their connectivity with vulnerability criteria used in other domains of Scotland's social security strategy, and recommend a set of vulnerability criteria, and consequent adjustments to income standards and/or energy needs, to be used in the context of fuel poverty.

Given the dominance of the Boardman-based and Hills definitions, Chapter 4 provides a more in-depth analysis of the common ground which they share, as well as their respective strengths and weaknesses. For the first time, it focuses specifically on how each of these definitions performs in the context of Scottish fuel poverty data. Neither definition emerges as ideal.

• A major drawback of the Boardman-based definition is that households which have quite high incomes can be classified as fuel poor - in Scotland especially, this group represents more than half of all those in fuel poverty, making this a very substantive problem.
• When considering the Low Income High Cost ( LIHC) indicator devised by Hills, almost the opposite problem emerges: people living on very low incomes may be excluded from fuel poverty if their energy costs are lower than the national median cost. Their energy bills may be just as burdensome, probably more so, when compared with households who have higher energy costs but can afford them. In light of Chapter 2's conclusion that any definition of fuel poverty should capture issues of hardship and energy burden, this drawback made the LIHC equally problematic when compared with Boardman's definition.
• A further difficulty with the LIHC lies in its insensitivity to fuel prices; estimates of the impact of changes in fuel price on fuel poverty are only accessible through a second-tier analysis of the fuel poverty gap, since these impacts are not reflected in the headline data concerned with prevalence.

Taking these issues into consideration, we concluded that neither of the two definitions currently operational in the UK were suitable for Scotland in a future fuel poverty context.

Chapter 5 focuses on optimal indoor temperatures, which have been a recurring issue for many years; precisely where temperature thresholds are set has implications for the prevalence of fuel poverty in Scotland (higher recommended temperatures will increase prevalence). But much more importantly it has implications for people's health, wellbeing, and thermal comfort. At present, Scotland recommends that able-bodied and healthy households have their living rooms set at 21° C, and all other rooms set at 18° C. These recommended temperatures are based on longstanding World Health Organisation ( WHO) guidelines, and in the absence of evidence to the contrary, the Panel took the view that these guidelines should continue to be adhered to.

For households where someone's poor health or disability makes them vulnerable to the adverse effects of cold and damp, WHO guidelines simply recommend that both temperature thresholds (living room and all other rooms) are increased by 2° C. At present only half of this recommendation is adhered to in Scotland, with living room temperatures of 23° C stipulated for vulnerable households. But bedroom temperatures have been left the same as for all other households at 18° C. The Panel thought the gap between 23° C (living room) and 18° C (all other rooms in the house) might demand too great a physiological adjustment for people who were suffering from ill health or disabilities. Consequently, we recommended 23° C for living rooms and 20° C for all other rooms in the homes of vulnerable people. This brings the overall recommended temperature regime more fully in line with longstanding WHO recommendations.

Given our proposed reorientation of the fuel poverty definition to more adequately relate to wider poverty and deprivation concerns, Chapter 6 deals specifically with our approach to developing and justifying our preferred concepts of poverty and energy affordability, and how we believe these might best be measured in future. We draw on insights from mainstream poverty research, and make the case for a more sophisticated understanding of relevant deprivations and hardships that moves beyond crude and 'arbitrary' income poverty thresholds. We take up some suggestions in the FPSWG (2016) report, particularly on the case for moving to an 'after housing costs' ( AHC) measure of income. We also develop the argument for using 'Minimum Income Standards' ( MIS) (Hirsch et al., 2016) in our revised definition of fuel poverty; this reflects the Panel's commitment to consensual and more participatory approaches that command majority public support in this field.

In Chapter 7 we focus on the association between

• six alternative ways of defining fuel poverty;

and

• some of the adverse outcomes of being fuel poor.

We consider a strong association between definition and adverse outcomes to be a key test of a good definition. We take the two established official definitions, Boardman and LIHC, and compare their performance with that of modified versions of each of these, and then also with two new definitions based on the Minimum Income Standards concept ( MIS). We use four distinct datasets to interrogate the performance of these different definitions in terms of their ability to predict, or discriminate between, households who report a range of adverse outcomes and those who do not. This leads to consistent findings which further support a move away from the status quo. We offer a revised definition, which emerges from these analyses, for scrutiny and comment. It is as follows:

Households in Scotland are in fuel poverty if:

• they need to spend more than 10% of their AHC income on heating and electricity in order to attain a healthy indoor environment that is commensurate with their vulnerability status;
• and if these housing and fuel costs were deducted, they would have less than 90% of Scotland's Minimum Income Standard as their residual income from which to pay for all the other core necessities commensurate with a decent standard of living.

Chapter 8 outlines the demography and geography of fuel poverty under different options for a revised definition. Which types of households are most likely to be in fuel poverty alters when using these different options, as does the energy efficiency profile of dwellings. Under our preferred option, the prevalence rate of fuel poverty in Scotland (2015) is broadly similar to the rate under the current Boardman definition (31% under Boardman and 32% under our preferred option). However, older age groups are less often deemed to be in fuel poverty, as are owner occupiers; a larger proportion of fuel poor households live in relatively energy efficient dwellings, highlighting the extent to which almost any level of energy cost is a significant economic burden for households on lowest incomes. Groups which are more often in fuel poverty under the preferred definition include those in rented accommodation, both social and private. Households where someone is living with a long-term illness or disability remain at relatively greater risk of fuel poverty than other households. Whilst not all rural households have a greater likelihood of being fuel poor, those in remote rural areas are at greater risk.

Under the preferred option, we also examined the fuel poverty gap - a means by which the severity of fuel poverty can be better understood. Groups with the largest gap (and likely to be experiencing more severe fuel poverty as a consequence) include elderly couples, private renters, and households located in remote rural areas. The Panel concluded that the combination of data on prevalence and severity was essential for a fuller understanding of how alleviation measures could be most effectively targeted.

Chapter 9 contains a summary list of the Panel's 37 Key Conclusions, as these appeared in the previous Chapters.

In the time available to us, we have proposed a new definition of fuel poverty which is rooted in an objective and impartial scrutiny of the current evidence base. We believe that this evidence base should serve as a foundation on which some core decisions about how fuel poverty should be defined in Scotland can rest.

The proposed revision retains the classic focus on issues of income and required energy cost, but it takes additional cognisance of the meaning and significance of being fuel poor in two key ways:

• in the manner in which income is measured, with the MIS being a democratic and participatory metric; and
• in testing our proposed definition (and 5 of its rivals) against core adverse outcomes that people say they experience as a consequence of being fuel poor.

During Panel discussions, we frequently considered how a new definition might affect professionals and practitioners working on doorsteps and communities in Scotland. We were unanimous in thinking that any revised definition should have the highest regard for the challenges they would experience in working with it. We are mindful of their importance in the months to come, as consultation and exchanges of views are worked through. At the very least, we hope that a revised definition will show a greater synergy between definition, Strategy, policy and practice than has hitherto been possible.

The Review concludes with a list of Sources and an Annex.