In part 1 of this series, Harriet Thomson discussed the methods and early outputs of a recently completed research project. In this follow up post, Harriet will now discuss the availability and quality of existing data for measuring fuel poverty at the EU-level, as well as the key outcomes of a pilot survey, before ending with recommendations for improving future measurement of European fuel poverty.
As outlined in the previous article, fuel poverty, energy poverty and energy vulnerability have been the focus of numerous European academic and policy conferences over previous years. One the unrelenting key messages across all these events has been the need for good quality standardised statistical data concerning housing stocks, energy efficiency, energy consumption (actual and modelled), individual energy needs, and health. Whilst a number of comparative studies of fuel poverty have been conducted (Thomson and Snell, 2012; Buzar, 2007; Healy and Clinch, 2002; EPEE, 2009; Whyley and Callender, 1997), none of these studies has been able to provide a comprehensive level of detail about fuel poor households in Europe, thus hindering efforts to produce robust policy frameworks.
Available data for measuring fuel poverty
At the European level, there is no dedicated survey of fuel poverty, energy poverty, or energy affordability, and an absence of standardised household micro data on fuel expenditure. At present, the main two sources of data are:
- The EU Statistics on Income and Living Conditions (EU SILC)
- Household Budget Surveys (HBS)
The EU SILC offers comparable statistics on income, living conditions and social exclusion, with an annual sample of approximately 100,000 EU households. Three indicators from the EU SILC dataset have been widely used to measure aspects of EU fuel poverty, namely, inability to keep adequately warm, living in a damp home, being in arrears on utility bills. Data from 2010 and 2011 shows that across the EU as a whole, nearly 10% of the population are unable to keep their home adequately warm, almost 16% live in homes that are damp, rotting or leaking, and around 9% are behind on payments for utility bills.
However, such proxy indicators are an imperfect measure of fuel poverty, with commentators criticising the subjective nature of the indicators, the potential for error of exclusion whereby respondents do not identify as fuel poor (Dubois, 2012), and the focus on adequate warmth, rather than on all energy services in the home. An additional concern is the poor overlap between consensual measures of fuel poverty derived from EU SILC and expenditure measures, with different populations identified as fuel poor under each measure. Finally, as binary yes/no response variables are used, the EU SILC indicators cannot distinguish between households that intermittently experience problems and those that persistently experience difficulties (Thomson, 2014a). Despite these flaws, EU SILC is the largest standardised dataset currently available at EU level and as such is the best available data.
By comparison, Household Budget Survey (HBS) are conducted in all EU countries to collect data on household expenditure on goods and services, including household energy. The main purpose of HBS is to compile weights for Consumer Price Indices and national accounts (Eurostat, 2014). Unlike EU SILC, which is an annual survey, HBS surveys are conducted irregularly across Europe, with the latest data reference years ranging from 2005 onwards.
Using HBS, the average (mean) household expenditure on energy as a proportion of income has been calculated for each country. The EU27 average is 7 – 8%, and the highest average expenditure is found in Slovakia (14.5%), whilst the lowest is found in Malta (1.8%) (European Commission, 2010). Figure 2 below displays the percentage of households spending twice the national average for energy. As can be seen, around a fifth of households in Slovakia, the United Kingdom and Estonia were spending twice the national mean for energy between 2005 and 2008 when the data was collected. By comparison only around 6% of households in Latvia, Bulgaria, Cyprus and Malta were spending twice the national mean.
However, as with EU SILC, there are numerous weaknesses to HBS data. Firstly, the data is not harmonised across Europe, with variation in sampling methods, variable design and how often Member States conduct HBS, ranging from annually to every five years (Eurostat, 2014). An additional concern is that actual expenditure on fuel is a poor indication of fuel poverty as low income households often spend significantly less on fuel than would be required to maintain a warm home (Moore, 2012). Because of these issues, the results displayed above should be viewed with caution.
Pilot survey of fuel poverty
Given the issues of data availability and quality highlighted above, this project set out to develop a pilot survey to test the wording, translation and format of new fuel poverty indicators, with a view to making recommendations for changing existing data collection. As the main purpose of developing a pilot survey was to pre-test fuel poverty variables rather than to collect representative data, non-probability convenience sampling was used via an online survey of individuals from all over Europe.
Some of the key findings of the survey testing process were:
- Using a multi-item scale variable rather than a binary-response variable is more suitable for subjective questions as the former can capture households that are intermittently fuel-poor as well those that are persistently fuel poor.
- A number of survey respondents stated they were part of a district heating network, which has previously been ignored in EU SILC central heating variables. Given that district heating users can be exposed to unique heating issues (Tirado Herrero and Ürge-Vorsatz, 2012), it is important that district heating be incorporated as a response option.
- Some respondents reported difficulties maintaining an adequately cool indoor temperature during summer months. The issue of summertime fuel poverty is underexplored and needs further examination.
- A range of energy sources were listed by respondents, including coal, wood, oil and bottled gas. This suggests a need to broaden the focus of survey questions and EU policy away from just mains gas and electricity.
Based on discussions with the expert project steering group, and the research findings produced throughout the project, a number of practical recommendations for improving the measurement of fuel poverty in the EU have been outlined in a final project report. The final report and other outputs are available at: http://www.eagacharitabletrust.org/index.php/projects/item/european-fuel-poverty-measurement-pilot-project-western-europe
Overall, there are four aspects to the recommendations: amending EU SILC, harmonising HBS, harmonising the monitoring of cold-related morbidity and mortality, and launching a new pan-EU survey of fuel poverty.
- Amend EU SILC:
The key criteria was that the proposed amendments should not deviate too substantially from the existing formats, in order to maintain comparability across survey years for longitudinal analysis and to avoid complex and costly changes. Three of the eleven suggested amendments are listed below:
- Harmonise HBS to produce a pan-EU dataset of actual fuel expenditure across Europe
Despite the issues associated with actual expenditure data, harmonised HBS data on energy expenditure would be useful for exploring seasonal and annual variations in energy expenditure, and for investigating differences in expenditure between different household types.
- Harmonise pan-European monitoring of cold-related morbidity and mortality.
As well as measuring the incidence of fuel poverty, it is important to monitor the ongoing health and wellbeing impacts of fuel poverty. During a recent policy conference hosted by the European Economic and Social Committee, Liddell (2014) cited the EuroMOMO project as an example of best practice for standardising the measurement of mortality across Europe.
- Create a new dedicated pan-EU household survey of fuel poverty
Given the limitations of existing data, and the difficulties associated with amending current surveys, it is suggested that consideration be given to creating a new pan-EU household survey of fuel poverty. The key advantage of this approach is that it would allow stakeholders to develop an evidence-based dataset that is relevant and appropriate for monitoring fuel poverty trends and developing rigorous policy responses.
Buzar, S. (2007) Energy Poverty in Eastern Europe: Hidden Geographies of Deprivation. Aldershot: Ashgate
Dubois, U. (2012) From targeting to implementation: The role of identification of fuel poor households. Energy Policy, 49: 107-115
EPEE, (2009) Tackling Fuel Poverty in Europe: Recommendations Guide for Policy Makers. http://www.fuel-poverty.com/files/WP5_D15_EN.pdf
European Commission (2010) Commission Staff Working Paper: An Energy Policy for Consumers. European Commission, Brussels.
Eurostat (2014) Glossary: Household budget survey (HBS). Available: http://epp.eurostat.ec.europa.eu/statistics_explained/index.php/Glossary:Household_budget_survey_%28HBS%29 (last accessed 1st April 2014).
Healy, J. D., and Clinch, P. (2002) Fuel poverty in Europe: A cross-country analysis using a new composite measure. Environmental Studies Research Series, University College Dublin.
Moore, R. (2012) Definitions of fuel poverty: Implications for policy. Energy Policy, 49: 19-26.
Thomson, H. (2014a) ‘The perception and incidence of fuel poverty across the European Union’. Presented at the ‘Energy Vulnerability in Europe’ workshop hosted by the Vulnerable Consumer Working Group of the European Commission, February 2014, Brussels.
Thomson, H. (2014b) The measurement of fuel poverty at the EU level: a comparison of approaches. Manuscript in preparation.
Thomson, H., and Snell, C. (2012) Quantifying the prevalence of fuel poverty across the European Union. Energy Policy, http://dx.doi.org/10.1016/j.enpol.2012.10.009
Tirado Herrero, S., and Ürge-Vorsatz, D. (2012) Trapped in the heat: A post-communist type of fuel poverty. Energy Policy, 49: 60-68.
Whyley, C. and Callender, C., (1997) Fuel poverty in Europe: evidence from the European Household Panel Survey. National Energy Action, Newcastle upon Tyne.