« Previous | Contents | Next »
Listen
Introduction
1. The Scottish House Condition survey ( SHCS) is the only national survey of housing in Scotland. It combines an interview with the household with a physical inspection of its dwelling, to build a picture of Scotland's housing stock. It covers all types of dwellings, whether owned or rented, flats or houses, across the entire country.
2. In 2003 the SHCS moved to a 'continuous' format. Previous surveys were conducted in 1991, 1996 and 2002 and typically covered a sample of around 15,000 paired household interviews and physical surveys of the dwellings.
3. The continuous format was introduced to allow more flexibility of content, to assist in the monitoring of Ministerial targets and to ease the management burden. The 2003/04 survey gathered data from almost 4,000 households and dwellings with paired social and physical data available for around 3,000 of these. The results in this report are based on fieldwork from October 2003 to September 2004.
Background
4. The Scottish Fuel Poverty Statement 1 was published in August 2002. This statement set out the Scottish Executive's definition of fuel poverty, and identified its causes and effects, in particular on vulnerable groups, and considered the circumstances of those in fuel poverty in Scotland. The statement drew on evidence from the 1996 Scottish House Condition Survey, and gave an account of developments since that time, further setting out delivery and outcome milestones with the ultimate aim of eradicating fuel poverty in Scotland.
5. Analysis of the 2002 Scottish House Condition Survey was first published in November 2003 2. This was followed up with a detailed report on Fuel Poverty in Scotland, published in April 2004 3. Key findings from the data for the first year since the survey became continuous, including some analysis of fuel poverty, were published in April 2006 4.
6. The aim of this report is to provide an update of some of the figures published in Chapter 4 of the Fuel Poverty Statement, to show how developments since 1996 have affected the overall make-up of fuel-poor households. Further analysis is included to provide more detail about such households.
Fuel Poverty
7. The term ' fuel poverty' refers to a situation where a household is unable to heat its home at an affordable cost. There is no universally agreed definition of fuel poverty in the academic and practitioner communities, with a number of different definitions being in use in different administrations and at different times.
8. It is important to note that the concept of fuel poverty is based on a theoretical calculation of how much it would cost to heat the dwelling to the specified regime. It does not address how much the household actually spends to heat the dwelling. In practice, each household may choose to heat their home differently.
9. Two definitions have been used throughout the history of the SHCS. The 2002 annual report and Fuel Poverty in Scotland made use of both definitions 5. The analysis in this report is based on the definition provided in the Fuel Poverty Statement ( FPS):
"a household is in fuel poverty if it would be required to spend more than 10% of its income (including Housing Benefit or Income Support for Mortgage Interest) on all household fuel use."
10. "Income" can be more precisely defined as income before housing costs, as derived from the Households Below Average Income ( HBAI) definition set out in the FPS. The HBAI definition has not been used exactly because the information gathered in the SHCS does not disaggregate income sources to this level of detail. However, the crucial point to note is that "income" in this definition is net of council and income tax.
11. The definition of "household fuel use" is derived from the following heating regime as set out in the FPS:
- Nine hours per 24 hour period during the week, with two hours being in the morning and seven hours in the evening;
- 16 hours per 24 hour period at the weekend;
- Living-room heated to 21 degrees Celsius;
- Rest of the house heated to 18 degrees Celsius.
- For elderly and infirm households, the home is continuously heated for sixteen hours per day to a temperature of 23 degrees Celsius in the living-room and 18 degrees Celsius in other rooms. The FPS does not give a precise definition of "elderly" or "infirm". The following definitions of these terms have been applied in the SHCS analyses:
- "Elderly" refers to a household where at least one member is aged sixty or over (male or female);
- "Infirm" refers to a household where at least one member has self-reported as long-term sick or disabled.
12. Note that, because of the changes in definition, any published figures from the 1996 survey will not be directly comparable to published figures for 2003/04. A complete description of both the 1996 and the current method used to calculate whether a household is in fuel poverty is given in Fuel Poverty in Scotland 6.
Who are the fuel poor in Scotland?
13. The 1996 figure of 738,000 households in Scotland (35%) published in the Fuel Poverty Statement is not directly comparable with the 2003/04 figure, because of the different definitions used. However, the 2002 annual report found that there was a significant fall in the number of households in fuel poverty between 1996 and 2002 using the 1996 definition 7.
14. Analysis of the Scottish House Condition Survey for 2003/04 shows that there were around 328,000 households (14.5%) in fuel poverty in Scotland in 2003/04, compared to 286,000 (13%) in 2002. This is not a statistically significant change. Nonetheless, the direction of this change is in line with rising fuel costs during this time.
15. Being in Fuel Poverty is when a household has to spend more than 10% of its income on fuel to maintain the standard heating regime. 'Extreme Fuel Poverty' can be defined as a household having to spend more than 20% of its income on fuel. In 2003/04, around one third (105,000) of households in fuel poverty could be described as suffering extreme fuel poverty ( see Table 1). This is a statistically significant increase on the 2002 figure of 69,000 - about a quarter of those in fuel poverty. While about the same number of households were in fuel poverty, more of those households were in extreme fuel poverty in 2003/04 compared to 2002.
16. Figure 1 (an update of Figure 1 in the Fuel Poverty Statement) and Table 2, show the proportion of Scottish households by presence of claimants and elderly experiencing fuel poverty. 'Claimant' in this context refers to those in receipt of certain welfare benefits 8. They show high rates of fuel poverty among the elderly, both those in receipt of welfare benefits and those not.
17. Figure 2 and Table 3 show the proportion of households classified into types by size and age in Scotland in fuel poverty. They show that single pensioners are the group most vulnerable to fuel poverty, followed by older couples and then single adults.
Fuel poverty and tenure
18. In contrast to 1996, the highest rates of fuel poverty in 2003/04 occurred in the private rented sector, with 22% of households experiencing fuel poverty. A rate of around 13-15% was seen amongst owner-occupiers, and the social rented sector. These trends are shown in Figure 3 (an update of Figure 2 from the Fuel Poverty Statement) 9. 2003/04 figures are provided in Table 4.
Fuel poverty and energy efficiency of homes
19. One important factor affecting domestic fuel use, and hence rates of fuel poverty, is the energy efficiency of homes. Energy efficiency is measured in the Scottish House Condition Survey using the 'National Home Energy Rating' ( NHER). This is based on the total energy costs per square metre of floor area required to achieve a standard heating regime. The NHER total energy costs include costs for space and water heating, lighting and standard domestic appliances. The NHER of a property is assessed on a scale of 0-10, with 0-2 being 'poor', 3-6 'moderate' and 7-10 'good'.
20. Figure 4 and Table 5 show that 94% of households in Scotland live in dwellings with 'moderate' or 'good' energy efficiency ratings. Since the 1996 survey, the NHER profile of dwellings in Scotland has changed. There were fewer households with 'poor' NHER in 2003/04 than households in fuel poverty with 'poor' NHER in 1996. The Warm Deal and the Central Heating Programme, together with the Scottish Housing Quality Standard for social sector housing and the Energy Efficiency Commitment targets for energy suppliers, have all contributed to this change in the energy profile of the housing stock. 40% of households lived in dwellings rated 'good' in 2003/04, compared to 31% in 2002. The number of households in dwellings rated 'poor' also fell, from 8% in 2002 to 6% in 2003/04 10.
21. However, the 6% of households who do live in dwellings with a 'poor' NHER are much more likely to be in fuel poverty. 43% of households in Scotland living in dwellings with a 'poor' NHER rating in 2003/04 were in fuel poverty, compared to 18% of households living in dwellings with a 'moderate' rating, and 8% living in dwellings with a 'good' rating.
Fuel poverty and income
22. Many factors such as dwelling size and age are linked with fuel poverty. They are described starting at Paragraph 24. However, the main factor is obviously low income. In 2003/04, around 92% of households with a weekly income of less than £100 were found to be in fuel poverty, and the majority of these were in extreme fuel poverty. The incidence of fuel poverty among households earning £300 a week or more was very low. This is illustrated in Figure 5 and Table 6.
Marginal fuel poverty
23. An important aspect of fuel poverty is the recognition that it is a dynamic state: households can both enter and be lifted out of fuel poverty due to changes in their own circumstances and changes in fuel prices. It is therefore useful to consider those households who are on the margins of fuel poverty, both just above and just below the fuel poverty threshold. For this purpose, Figure 6 and the tables in the annex look at both households required to spend 8-10% of their income on fuel in order to maintain the heating regime, and households required to spend 10-13% of their income on fuel. Table 1 shows that 155,000 households fall just below the fuel poverty level, being required to spend around 8-10% of their income on fuel, and with a change in circumstances could easily become fuel poor. 103,000 households fall just above the threshold, being required to spend 10-13% of their income on fuel.
Other characteristics of households in fuel poverty
24. Some of the characteristics of households in fuel poverty and the dwellings in which they live have already been covered. However, there are many other factors which are correlated to some degree to the level of incidence of fuel poverty, such as the age, size and type of a dwelling, its external wall construction, the extent of central heating, the presence of double glazing, fuel used for heating, location, the number of dependent children and the health status of residents. These are covered in Tables 7 to 16.
25. Table 7 shows that a greater proportion of disabled people and people in poor health are in fuel poverty than the rest of the population. Table 8 shows that households without children are more likely to be in fuel poverty than families. Table 9 shows that households living in flats are less likely to be in fuel poverty than those living in houses. Tables 10 and 11 show that a greater proportion of households living in older dwellings and larger dwellings are in fuel poverty than those living in newer and smaller dwellings.
26. Table 12 shows that only 14% of households living in dwellings with full central heating are fuel poor, compared to 22% living in dwellings with partial central heating, and 33% with no central heating. Table 13 shows that only 11% of households using mains gas to heat their home are in fuel poverty, compared to 27% of households using other fuels. Table 14 shows that a smaller proportion of households living in dwellings with cavity walls are in fuel poverty than those with solid walls. Similarly, Table 15 shows that a smaller proportion of those with double glazing are in fuel poverty, compared to those with single glazing.
27. Table 16 shows that households living in rural areas are more likely to be in fuel poverty than those in urban areas. One third of households in remote rural areas are in fuel poverty, compared to 12% in large urban areas.
28. Figure 7 shows the relative importance of a number of factors in predicting the likelihood of a certain household to be in fuel poverty 11. The most important factor in predicting fuel poverty is household income. As previously discussed, around 92% of households with an income of less than £100 a week are in fuel poverty.
29. The next most important factor is the size of the dwelling. This is not surprising, as larger dwellings require more heating. The diagram shows that, for households earning between £100 and £200 a week, around 38% are in fuel poverty, but of those in large dwellings (for these purposes, considered to be 260 cubic metres or above) around 83% are in fuel poverty.
30. At the third and fourth levels of the diagram, a number of different factors are shown to be important, depending on the income level and size of the dwelling. For incomes of £100-£200 a week, the energy rating of the dwelling is considered to be the next most important factor. For higher incomes, primary heating fuel and external wall construction both come into play. Even for households with incomes in excess of £300 a week, the diagram shows that fuel poverty can be prevalent, with around 15% of households earning above £300 a week, living in large dwellings, using electricity as their primary heating fuel, being in fuel poverty. It should be noted, though, that the further we drill down into the sub-categories, the smaller the sample size becomes, so the confidence intervals around such estimates become quite wide.
Sampling variability
31. Although the SHCS sample is chosen at random, the selected households will not necessarily be representative of the population. For example, purely by chance, the sample could include disproportionate numbers of certain types of people or dwelling. In general, the smaller the sample size, the greater the likelihood that the estimate could be misleading.
32. The 'paired' (social and physical survey) overall response rate to the survey is about 60%, i.e. for every two households that agree to take part, one household refuses. The survey is reweighted for such non-response but we cannot know authoritatively how the missed population differs from the compliant population.
33. The likely extent of sampling variability can be quantified, by calculating the "standard error" associated with an estimate. By convention, a "95% confidence interval" is used to demonstrate the variability. On average, there is a one in twenty chance that the true value would not fall within the range given by the confidence interval.
34. Table 17 shows the "95% confidence limits" for estimates for a range of percentages calculated from sub-samples of a range of sizes. Note that the confidence limits for estimates of x% and (100-x)% are the same. The interpretation and use of this table are best demonstrated by an example.
35. In Table 3a, 35% of single pensioner households are said to be in fuel poverty. At the bottom of the table it says that the sample size of single pensioner households was 417. Looking at Table 17, and reading across the 400 row to the 35% column, we get a confidence interval of ±5.3% around the estimate of 35%. Thus we can say that we are 95% confident that the true proportion of single pensioner households who are in fuel poverty lies between 29.7% and 40.3% (35% ± 5.3%).
« Previous | Contents | Next »