The Scottish Government recently announced a new consultation on a Fuel Poverty Strategy for Scotland, leading up to an intended Warmer Homes Bill next year. A core element in this strategy is the proposed adoption of a new definition of Fuel Poverty for Scotland, departing significantly from the established definition and, at the same time, continuing to distance Scotland somewhat from England. In a nutshell, Scotland’s existing definition is that a household is in fuel poverty if its required fuel costs exceed 10% of its net income; in England, a more complex ‘Low Income-High Cost’ definition was adopted following a review led by John Hills in 2012.
I must declare an immediate interest here — I was one of a ‘gang of four’ academics appointed to an independent panel. Our task was to review the definition of fuel poverty in the light of evidence and arguments, and to come up with a recommended approach for the future (our report was published here). We recommended that the definition of fuel poverty should be based on households whose required fuel costs would exceed 10% of their income ‘after housing costs’, and who would be left with less than 90% of the ‘Minimum Income Standard’ (MIS) for their household composition after paying for fuel and housing (for further detail on the MIS, see Hirsch et al 2016 and this report from The Joseph Rowntree Foundation) . This definition is used to measure progress toward government targets of reducing fuel poverty at the national level (a case of conspicuous policy failure recently), while also serving to direct help through various schemes to households most at risk.
It is not always the case that such independent reviews lead to governments adopting their recommendations, but in this case the Scottish Government have adopted the main thrust of our recommendations, while differing slightly on some details of implementation. I want to reflect briefly here on four features of this process, which illustrate some recurring themes in our policy research work, and which often feature in blogs on this site.
My first observation is a comment on process, which is that inter-organisational inertia can be a significant inhibition to change. Organisational actors constantly try to second guess what other actors and organisations will feel or think, and whether they will resist or support change. In this instance, we were constantly warned that ‘the sector won’t like it’, the civil servants didn’t like it, and ‘the minister may not like it’, etc. As it turned out in consultation events and wider presentations the fuel poverty/energy efficiency sector were predominantly receptive and supportive, and the same was true of the minister, with both minister and civil servants canvassing going ‘even further’ than we proposed in some respects.
My second observation concerns the power of consensus, not in relation to the policy process but in relation to the underlying philosophical and evidential base for the recommended approach to the definition. We argued in our report that the most significant and persuasive developments in approaches to measuring poverty were in essence ‘consensual’ approaches – that is approaches which try to define needs and standards through a process which demonstrates consensus, or at least wide and stable agreement. We considered two such approaches – the Minimum Income Standard (MIS) and the Consensual Material Deprivation approach as exemplified by the UK Poverty and Social Exclusion Survey (see also Bramley & Bailey (2017) chapters 1 & 9,) In the end we opted for MIS, in this case on grounds of practical applicability in programme implementation, particularly ‘on the doorstep’ when support agencies need to rapidly assess households’ likely eligibility. Incorporating MIS within the Fuel Poverty definition in Scotland is the first time this methodology, developed by the Joseph Rowntree Foundation, has gained official recognition.
Thirdly, I would claim that this outcome is a modest victory for ‘evidence-based policy’, perhaps the strongest recurring theme in this blog and certainly the main focus of my very first contribution to it. The Panel’s review entailed considerable reviewing of evidence from literature about a number of issues, notably on domestic temperature regimes and their relationship with health and wellbeing, and on issues of vulnerability. However, it was the focus on a certain type of evidence that undoubtedly swung the case for change, and the direction of that change – the potential adverse outcomes of fuel poverty, and the systematic examination of evidence on the relationship of these to different fuel poverty definitions. Across four large-scale datasets and using a dozen distinct outcome measures, it was quite clear that the worst definitions, in terms of showing a poor association with adverse outcomes, were the two existing official definitions (the Scottish and the English), while the MIS-based variants were generally the best of those practically feasible options considered.
Finally, and to balance the above point, it must be conceded that, whatever definition is used, the ‘fit’ to the data remains quite poor. In other words, practical definitions remain crude rules of thumb, while individual homes and households vary very widely, and so certainly do their fuel bills. While some of this may be down to inadequacies in the technical models used to generate ‘standardised’ fuel costs, much of it reflects different priorities, different choices and behaviours of households, but also different abilities to understand and control heating systems, fuel contracts, etc. We certainly felt that this issue presented a microcosm of the global concept of human capabilities (Sen & Nussbaum 1993), which some argue should form a more fundamental basis for assessing need and poverty (Thomson et al. 2017). There are quite a few households in Scotland who are not ‘fuel poor’, in any meaningful sense of the word ‘poor’, but who are using an awful lot of fuel (generating a lot of emissions) and spending an awful lot on it and/or having pretty poor outcomes in terms of thermal comfort. People in this situation need a service that provides them with support in sorting this situation out, even if they don’t need a massive subsidy. Such a service goes beyond a phone call, which often seems to be mainly preoccupied with selling them something which they may or may not need, and requires trusted intermediaries to come into the home and make an overall assessment and provide detailed advice.
In this era when academics are expected to achieve ‘policy impact’ as well do sound research and excellent teaching, it is nice to report a case of fairly direct impact on policy. It is helpful in this instance to be working in a Scottish policy context, which is more receptive to ideas about trying to address poverty and inequality more effectively than in some other jurisdictions. Nevertheless, not all of the measures recommended in previous blogs in this series, based on recent research evidence, have received such a positive reception – see for example the case of the Council Tax, – although we should always live in hope.
Hills, J. (2012). Getting the Measure of Fuel Poverty. Final Report of the Fuel Poverty Review. CASE Report 72. ISSN 1465-3001. Centre for Analysis of Social Exclusion. London School of Economics.
Hirsch, D., Padley, M, & Valadez, L. (2016) A Poverty Indicator Based on a Minimum Income Standard. CRSP Working Paper 656. Loughborough University, Centre for Research in Social Policy.
Bramley, G. & Bailey, N. (eds) (2017) Poverty and Social Exclusion in the UK: vol. 2 –The dimensions of disadvantage. Bristol: Policy Press.
The 2017 Scottish Fuel Poverty Definition Review Panel (2017) A new definition of fuel poverty in Scotland: A review of recent evidence. Edinburgh: Scottish Government http://www.gov.scot/ISBN/9781788512428
Sen, A. & Nussbaum, M. (1993). The Quality of Life. Oxford: Oxford University Press.
Thomson, H., Bouzarovski, S. & Snell, (2017). Rethinking the measurement of energy poverty in Europe: A critical analysis of indicators and data. Indoor and Built Environment, in press.