Rising gas prices bring concerns about energy affordability, with estimated shares of household incomes spent on energy indicating the most serious pressures in a generation. But changes to official definitions of fuel poverty mean that the statistics need to be interpreted with caution.
From 1 April 2022, the energy price cap increased by £693 or £708 for around 22 million customers in Great Britain. This increase is linked to the rising wholesale gas price, which rose from a monthly average of 46p per therm (a unit of heat) in February 2021 to 188p per therm in February 2022 (see Figure 1). These increases are passed on to consumers, raising household bills.
Figure 1: Average monthly wholesale gas prices for gas received the following day in Great Britain, February 2019 to February 2022
The affordability of energy is often framed in terms of fuel (or energy) poverty. Understanding levels of fuel poverty is one way to consider the effects of these rising prices. But a multitude of fuel poverty indicators have been suggested by researchers. In England, three different official definitions have been used, while Scotland and Wales use their own definitions.
Without appreciating the variety of metrics, simple quotations of fuel poverty rates may provide a misleading sense of precision. Whether headline rates of fuel poverty pick up fluctuations in household energy bills over time depends on the statistical definition used.
How should we frame energy affordability and fuel poverty?
Why might we be concerned about energy affordability? When considering this, two things are worth remembering. First, households are concerned about the services that energy provides – in other words, heat and light – rather than gas or electricity itself. Second, a household’s energy expenditure is part of their overall decisions about how to spend their income on different goods and services.
When energy prices increase, concerns about energy affordability boil down to three possibilities:
- Households reduce the energy services that they consume.
- Higher energy expenditures reduce households’ consumption of other products.
- Consumption of energy services and other products is maintained by households incurring debt and/or drawing down savings.
This also assumes that society can agree on a minimum level of energy services (and other products) that it deems necessary or reasonable. Fuel poverty can therefore be said to be occurring when the minimum level of energy services is deemed unaffordable.
Several observations follow from this framing:
- Energy efficiency can improve energy affordability by increasing the quantity of energy services that can be bought with the same expenditure.
- Income is central to energy affordability – lower incomes make it more challenging to pay for a given level of consumption.
- Both high and low levels of energy expenditure could indicate energy affordability challenges.
- How households choose to adjust their consumption following an energy price shock will depend on their individual preferences for energy services and other products.
These four points explain why measuring fuel poverty and energy affordability is more complex than the presence of official fuel poverty statistics may suggest. Different metrics will be stronger at capturing different elements of these four observations.
What is the long-run assessment of energy affordability?
To gain a first impression of energy affordability over the long run, we can look at average (mean) household expenditure on non-motoring fuels, divided by average (median) equivalised household disposable income in the UK between 1977 and 2019/20 (see Figure 2).
Figure 2: Ratio of mean household energy expenditures over median equivalised disposable income, 1977 to 2019/20
Source: Adapted from Deller, 2022
Figure 2 indicates that the proportion of households’ income devoted to energy was much higher in the late 1970s and early 1980s than it has been since the early 2000s. Even when energy costs were last the focus of media attention (in the early 2010s), the proportion of income devoted to energy was only as high as it was in the early 1990s.
Since heat and light are necessities, the quantity of energy consumed responds by a relatively small amount to changes in price or income, at least in the short run. Economists refer to this as energy being both price inelastic and income inelastic (Labandeira et al, 2017; Zhu et al, 2018).
Specifically, when there is an energy price increase, the percentage fall in the volume of energy consumed will be lower than the percentage increase in price. Similarly, when incomes go up, the percentage increase in the volume of energy consumed will be lower than the percentage increase in income. As a result, a higher share of income devoted to energy is an indicator of worsening energy affordability.
Where might energy affordability be later in 2022?
In 2019/20, the average household expenditure on non-motoring fuels was £1,284. The value of the ratio between expenditure on these fuels and median household income (used in Figure 2) was 4.3%.
But Figure 2 is based on survey data from the Office for National Statistics (ONS), which will only become available for 2022 sometime in the future. To gain a more accurate idea of where energy affordability might be later this year, we can use the Office for Budget Responsibility’s (OBR) nominal GDP forecasts. These data can be used to estimate the median equivalised household income in 2022/23 (currently £32,270), and then plug potential energy expenditures into the ratio.
An obvious energy expenditure figure to use is £1,971 – the new level of the default tariff cap for direct debit consumers set by Ofgem, the regulator for gas and electricity markets in Great Britain. This results in a ratio of 6.1% – the highest since 1988.
One projection for Ofgem’s tariff cap for the winter of 2022/23 indicates that it could reach £2,600 (Cornwall Insight, 2022), implying a ratio of 8.1%. Such a value would be heading towards the peaks of the late 1970s and early 1980s. To surpass the 1985 peak of 9.3% would require average annual energy expenditures in the coming year to exceed £2,997.
The conclusion is that this year’s energy affordability pressures are likely to be the most serious for a generation.
What does Ofgem’s default tariff cap show?
A degree of caution with these projections is necessary. Ofgem’s cap is not the actual amount that households will spend on energy, nor is it directly comparable with the average energy expenditures in Figure 2.
Ofgem’s values are generated by assuming that households consume fixed quantities of electricity and gas, labelled ‘typical domestic consumption values’ (TDCVs). Households consuming different quantities to the TDCVs will have different energy expenditures, and not all households have a mains gas supply.
As the TDCVs are fixed quantities, the quoted figures for Ofgem’s cap do not reflect the fact that households will respond to higher energy prices by reducing their energy consumption to some extent. Only future releases of the Living Costs and Food Survey will allow the time series data in Figure 1 to be extended in a fully comparable fashion.
How have official definitions of fuel poverty varied over time?
That consumption responds to price changes highlights one of the trade-offs when measuring energy affordability and fuel poverty.
Figure 2 is based on households’ actual energy expenditures. For identifying households that cannot afford energy, a concern with using a high ratio of actual energy expenditure to income is that it may fail to pick up households that respond to affordability pressures by limiting their energy expenditure.
To address this issue, the official fuel poverty statistics in England are based on an engineering model that estimates the energy expenditures required to achieve a pre-specified temperature. The trade-offs from this approach are that differences in households’ temperature preferences are ignored and that the estimates are only as valid as the engineering model’s own assumptions. While using required energy expenditures is done for valid reasons, it also implies a potential gap between official fuel poverty statistics and the energy bills that households actually pay.
In England, there have been three official fuel poverty definitions. These changes in definition, as well as the complexity of calculating required energy expenditures, mean that there are no time series data on the official fuel poverty rate comparable to Figure 2.
Table 1 provides the core definitions of the official fuel poverty metrics (for the full definitions of income etc., see bre, 2020 and bre, 2022). Figure 3 charts the official data that are available.
Table 1: Official fuel poverty definitions in England
|Statistic||Definition||Adopted||Period of available data|
|10%||(Required fuel costs / income ) >0.1||2001||2003-11|
|Low income — high cost (LIHC)||(i) Required fuel costs are above the national median and |
(ii) Income after the deduction of required fuel costs is below the official poverty line (60% of median equivalised disposable income)
|Low income — low energy efficiency (LILEE)||(i) Live in a property with a fuel poverty energy efficiency rating (FPEER) of band D or below and |
(ii) Income after the deduction of required fuel costs is below the official poverty line ( 60% of median equivalised disposable income)
In Wales, the 10% metric continues to be used, while Scotland is turning to a two-part metric that combines the 10% threshold with the requirement that a fuel-poor household has an income – after housing, fuel and childcare costs – that is below the level needed for an acceptable standard of living.
Figure 3: Official headline rates of 10%, LIHC and LILEE fuel poverty in England (available data)
Source: Combines data from Department for Business, Energy and Industrial Strategy, BEIS, 2021; BEIS, 2022 and BEIS, 2013
Figure 3 shows how the different fuel poverty definitions have a noticeable impact on the percentage of households identified as fuel-poor, as well as the time trend for fuel poverty.
In particular, the rate of low income-high cost (LIHC) fuel poverty shows very little variation over time. This is because it is a ‘relative’ metric – what constitutes high energy costs is measured relative to median energy costs.
A sharp rise in energy prices will lead to a sharp increase not only in energy costs for individual households, but also in median energy costs. As a result, the ability of the headline rate of LIHC fuel poverty to represent fluctuations in energy bills and energy affordability over time is limited.
The 10% fuel poverty metric has a time trend that shows some similarity to the relevant section of Figure 2. This is because it is based on a related ratio and, being a fixed threshold, the number of households exceeding the 10% threshold will increase as energy prices rise.
Low income-low energy efficiency (LILEE) fuel poverty shows a continuously declining trend, which is likely to be related to the average energy efficiency of dwellings increasing over time. Households in homes with a high energy efficiency rating are excluded from LILEE fuel poverty by design.
An important consideration is the extent to which the LILEE fuel poverty rate will rise with the recent increase in energy bills. Higher energy prices will result in more households being captured by the second part of the LILEE definition – that income after the deduction of required fuel costs is below the official poverty line – but this will be tempered if the proportion of energy efficient homes increases over time.
Growing numbers of energy efficient homes is especially likely if some households respond to price increases by installing energy saving technologies. It seems plausible that the increase in LILEE fuel poverty over the coming year will be lower than if fuel poverty were still measured by the 10% metric.
The takeaway: fuel poverty statistics need to be interpreted with caution
The relationship between energy bills and rates of fuel poverty all depends on the indicator being used. Policy-makers basing decisions on the fuel poverty rate need to understand the implications of the definition of the statistic that they are observing.
For example, under the LILEE indicator, someone living in an energy efficient dwelling but struggling to afford energy due to a very low income would not be identified as fuel-poor. They would simply be identified as income-poor. In contrast, using the 10% metric, the same individual could be identified as fuel-poor.
Identifying the best fuel poverty metric is likely to depend on the intended purpose. A 10% metric would be good at identifying changes in energy affordability linked to fluctuating energy prices, as we are seeing at the moment. On the other hand, the LILEE metric is reasonable for tracking the roll-out of energy efficiency upgrades in low-income households.
An alternative approach to assessing fuel poverty would be to measure energy services directly and identify the households that are dissatisfied with the services that they can afford.
For example, if a primary concern is households living in the cold, it would seem desirable to measure the temperatures achieved in homes, as well as households’ temperature preferences (Deller et al, 2021).
While such an approach would require investment in new data and methods, it would directly tie assessments of fuel poverty to households’ views of their lived experience.
Where can I find out more?
- Official English fuel poverty statistics: a comprehensive set of data and reports with lots of detailed breakdowns.
- Energy Economics special issue – Energy poverty: trends and perspectives – the latest economics research on energy (fuel) poverty from a global perspective.
- Fairness in retail energy markets? Evidence from the UK: the final report of a UK Energy Research Centre project combining economic, legal, sociological and political science perspectives.
- Ofgem data portal: a comprehensive set of charts on the functioning of British energy markets.
- National Energy Action: the leading UK charity campaigning on fuel poverty.
Who are experts on this question?
- Catherine Waddams
- Andrew Burlinson
- Stefan Bouzarovski
- Carolyn Snell
- Brenda Boardman
- Benjamin Sovacool
- Harriet Thomson