How poverty and low income are measured in the UK

Poverty in the UK is not measured by counting who cannot afford food or heating. It is measured by comparing household income with the income of typical households across the country. The official poverty line is set at 60% of the median UK household income, and a person is counted as being in poverty if the income of their household, after adjusting for its size and composition, falls below that line¹.

Because the line is a share of what typical households have, it moves as the country gets richer or poorer. That produces two main measures: relative poverty, where the line moves each year with current incomes, and absolute poverty, where the line is fixed to a base year and then adjusted only for inflation. Both are published on two income bases, before housing costs and after housing costs, and the difference between those two bases is large in some parts of the country, above all in London³.

The figures come from a single source: the Family Resources Survey, an annual government survey of around 20,000 UK households, which feeds the Department for Work and Pensions' Households Below Average Income series, the UK's primary source of poverty estimates⁵.

The UK poverty line: 60% of median household income

The starting point for every official poverty figure is the median: the income of the household in the middle of the distribution, with half of households above it and half below. A household is below the poverty line if its income is below 60% of that median. The Welsh Government's statistics, for example, define a person as living in relative income poverty if they live in a household where total household income from all sources is less than 60% of the average UK household income, as given by the median¹⁰.

The median is used rather than the mean because it is not distorted by very high incomes at the top. A small number of very wealthy households would pull a mean upwards and make everyone else look poorer by comparison, while the median stays anchored to the household in the middle.

Household income is adjusted, or "equivalised", for the size and shape of the household before the comparison is made. A single person needs less income to reach the same living standard as a couple with children, so the thresholds differ. Scotland's official statistics show how this works in cash terms: in 2022-25, the relative poverty threshold after housing costs, at 60% of the UK median, was £211 per week for a single person, £364 for a couple without children, £437 for a single person with two children aged 5 and 14, and £589 for a couple with two children of those ages⁷.

Alongside the 60% line, the official statistics also publish figures for households below 50% and 70% of the median, which show how sensitive the headline number is to where the line is drawn¹¹. A household just above 60% of the median is not materially different from one just below it; the threshold is a convention, not a cliff edge.

Relative and absolute poverty: how each one is measured

Relative poverty is measured against the UK median income in the year in question. Northern Ireland's official statistics state it plainly: an individual is in relative poverty if they live in a household with an equivalised income below 60% of UK median income in the year in question¹². If typical incomes rise, the line rises with them, so a household can be pulled out of absolute poverty by rising living standards while remaining in relative poverty.

Absolute poverty uses the same 60% threshold, but the median it is measured against is fixed to a base year and then adjusted for inflation. For years before 2021/22, the base year is 2010/11¹³. From the March 2026 Northern Ireland publication, the absolute reference year changes to 2024/25 for the years where administrative data linking has been applied, and it remains 2010/11 for prior years¹¹. Because the absolute line only moves with prices, it shows whether the poorest households are better or worse off in real terms than they were in the base year, while the relative line shows whether they are keeping pace with everyone else.

The two measures tell different stories about the same households. In 2022/23, 21% of individuals in the UK were in relative poverty after housing costs and 18% were in absolute poverty after housing costs; among children, 25% were in absolute poverty after housing costs¹⁴. Absolute rates have generally fallen further than relative ones over the long run, because incomes have grown in real terms even while the poorest have not gained ground against the middle.

The headline numbers are large in absolute terms. Relative poverty in the UK increased by one million people in 2021/22 to reach 14.4 million, or 22% of the population, according to analysis by the TUC¹⁵. The Joseph Rowntree Foundation's dashboard, drawing on Households Below Average Income 2023/24, puts overall relative poverty after housing costs at 21%, within a band of 20% to 22% that has held since 2004/05³.

Before and after housing costs: why rent and mortgages change the figures

Every poverty figure is published twice: once before housing costs are deducted and once after. Housing costs can include rent or mortgage payments, building insurance and water rates⁴. The after housing costs measure asks a more practical question: once the household has paid for the roof over its head, how much income is left to live on?

The two measures can diverge sharply, and the size of the gap depends on where a household lives. In 2023/24, the gap between the before and after housing costs poverty rates was 11 percentage points in London, compared with 4 percentage points in the rest of England⁴. A separate Trust for London indicator puts the rest of England gap at 3 percentage points for the same year, so the two sources differ slightly on the figure outside London, but they agree that London's gap is several times larger¹⁶. The proportion of people in poverty in London almost doubles when housing costs are taken into account, rising from 15% before housing costs⁸.

Housing costs have been rising for many households, not only in London. In October 2023, one in six UK households (16%) had seen their rent or mortgage go up by £100 or more per month; among households with housing costs, excluding those without, the figure was 26%¹⁷.

The burden falls hardest on those already on low incomes. In the rest of England, households in poverty spent 32% of their income on housing costs, compared with 9% for those not in poverty, on a three-year average to 2024/25¹⁸. This is the core reason the after housing costs measure is usually treated as the more meaningful one for comparing places: two households with the same income can be left with very different amounts to live on.

Material deprivation: measuring what people go without

Income measures say nothing about what a household can actually buy. Material deprivation fills that gap by asking whether people have access to a list of goods and services that most people would regard as necessities, from a warm winter coat to a holiday away from home.

For children, the official combined low income and material deprivation measure counts a child as poor if their household has an income below 70% of the contemporary UK median and a material deprivation score of 25 or more¹⁹. The UK government's Child Poverty Strategy adds a deep material poverty measure, defined as lacking at least 4 out of 13 essential material deprivation items²⁰.

Wales collects its own material deprivation data through the National Survey for Wales, and the results show how deprivation clusters with other disadvantages. In 2022-23, 25% of people in households with children aged 16 or under were materially deprived, compared with 11% of people in child-free households; 23% of people with no qualifications were deprived; and 14% of people in employment and 17% of economically inactive people were in material deprivation²¹. People who had bought or received second-hand items in the last 12 months were more likely to be in material deprivation, at 17%, than those who had not, at 10%²¹.

The Welsh survey also links deprivation to wellbeing: in 2021-22, 52% of materially deprived people reported high life satisfaction, and 10% of non-deprived people reported loneliness²². One caution applies to these figures: the National Survey for Wales results are not comparable with the overall proportions produced by the Family Resources Survey, because the questions and methods differ²².

Pensioner material deprivation: 4 or more of 19 items lacked

Pensioners have their own material deprivation measure, and it works differently from the children's version. A pensioner is in material deprivation if they lack 4 or more items¹¹. The questions were updated in 2023/24: there are now 19 questions, of which 8 are pensioner items and 11 are household items, whereas up until 2022/23 respondents were asked about 15 goods and services¹¹. Before the change, a pensioner was counted as deprived if they had a final material deprivation score of 25 or more²³.

The measure is only available for pensioners aged 65 or over¹¹. Northern Ireland's statisticians present it as a separate measure rather than combining it with low income, on the grounds that for pensioners the concept of material deprivation is broad and very different from poverty¹¹.

The change in questions means figures from before and after 2023/24 cannot be directly compared. Trust for London notes a similar caution in its own data: 45% of children in households in poverty in London were in material deprivation in 2023/24 under the previous measure, but the measure has changed and should not be directly compared with later figures⁸.

Where the figures come from: the Family Resources Survey

All the low income statistics in the UK are built on one survey. The Family Resources Survey is a continuous household survey, run by the Department for Work and Pensions, collecting information from a representative sample of around 20,000 private households in the UK each year, of which approximately 2,000 interviews take place in Northern Ireland⁵. It was launched in Great Britain in October 1992 and extended to Northern Ireland in April 2002²⁷.

The survey aims to interview all adults aged 16 or over in each selected household, and covers household composition, tenure, mortgage and rent details, household insurance, income, benefits, tax credits, pension schemes, savings and investments, and social deprivation²⁷. Its primary purpose is to give the DWP data to inform the development, monitoring and evaluation of social welfare policy, and it is used for tax and benefit policy modelling by HM Treasury and HM Revenue and Customs, as well as by academics and research institutes²⁷.

The FRS estimates underpin the Households Below Average Income series, which is the UK's primary source of poverty estimates²⁶. HBAI figures are published as three-year moving averages, rounded to the nearest whole number, with breakdowns for the countries of the UK and the regions of England, covering April 1994 to March 2025¹. The FRS datasets and published information are accredited official statistics⁵.

The survey is being transformed. Benefit linkage has been incorporated, replacing survey-reported benefit receipt with administrative records to improve accuracy and reduce historical underreporting, with earnings linkage and an updated grossing regime planned to follow¹¹. National Insurance Numbers can now be matched for more than 95% of respondents in Great Britain, up from just over 50% previously³⁰.

What counts as income in the poverty statistics

The income being measured is total household income from all sources: earnings from work, benefits and tax credits, pensions, and other income, after deducting taxes. The Welsh definition spells this out: a person is in relative income poverty if their household's total income from all sources is less than 60% of the UK median¹⁰.

Because benefits count as income, the poverty statistics capture the full effect of the benefit system. Income from employment accounted for 72% of gross household income across the UK in 2024/25, which means the remaining share comes largely from benefits, pensions and other sources⁵.

Work does not guarantee being above the line. In Wales, an estimated 50,000 working-age adults were in relative income poverty despite living in households where everyone worked full-time, and 17% of children in households where all the adults worked were in poverty, compared with 51% of children in households where some but not all adults worked¹⁰. Research by the Joseph Rowntree Foundation found that in 2011/12, for the first time, more than half of people in poverty lived in a working family, and that over half of adults in working families in poverty were paid below the living wage once self-employed people were excluded³¹.

Disability also raises the risk. In Wales, 28% of children in families where no-one was disabled were in relative income poverty, and the rate was markedly higher where someone in the family was disabled¹⁰. Among people in relative income poverty in Wales, 50% were in families with children and 34% were in families without children¹⁰.

Limits of the data and breaks in the series

The poverty statistics have known limits, and several of them matter when comparing figures over time.

The most serious break concerns 2020/21. Data for that year has been excluded from the Households Below Average Income analysis due to concerns with bias in the sample, collected during the pandemic⁹. It is also not available in the DWP's Stat-Xplore tool, because of data quality concerns affecting many of the HBAI estimates²⁸. Averages that would normally span 2020/21 are calculated over two years instead of three¹⁶.

A second break comes from the transformation of the survey itself. Key low income measures changed with the move to administrative data linking, creating a break in the data series from 2022, and users should not make direct comparisons of the low income measures across this break¹. Trust for London's analysis notes the same: there is a break in the series after 2021/22, meaning values after this point should not be directly compared with those before¹⁶.

Coverage is a third limit. The survey is designed to be representative of private households, but some individuals are not covered: students in halls of residence and residents of nursing homes are excluded⁵. The Household Finances Survey, used for some related income statistics, likewise excludes people in institutionalised households such as care homes and hostels, and people who are homeless²⁶. In Northern Ireland, the sample is smaller because households with spouses living away from home are excluded¹¹.

Finally, the survey is a sample, not a census. The total non-response rate is typically around 50%, which the statisticians do not consider unreasonable for a survey of this kind, and the results are weighted, using software called CALMAR, to reconcile the sample with population totals²⁵. The sample is stratified and clustered to produce robust regional estimates, but it is not suitable for analysis below regional level⁵. Confidence intervals of plus or minus two standard errors are used to state, with 95% cent confidence, the range within which the true value lies²⁵.

Low income in London: work, housing costs and the poverty premium

London's poverty figures are shaped by two forces: high housing costs and the extra costs of being poor.

On the income measure, the capital looks better off than it is. London's poverty rate is 15% before housing costs, but the proportion of people in poverty almost doubles when housing costs are taken into account⁸. The gap between the before and after housing costs rates was 11 percentage points in 2023/24, against 4 percentage points in the rest of England⁴.

Housing affordability compounds this. An average-priced home in London cost the equivalent of 25 years of a low-income household's income in the financial year ending 2024, compared with 13 times a lower-income household's income in England as a whole and 7.8 times in Scotland³². Households in poverty in the rest of England spend 32% of their income on housing, against 9% for those not in poverty¹⁸.

On top of this sits what researchers call the poverty premium: the extra costs low-income households pay for the same goods and services, such as paying for energy through prepayment meters or being unable to pay annually for insurance. Trust for London's 2026 analysis found that around 97% of households in poverty in London are affected by the poverty premium in some way, at an average cost of £451 a year per affected household, which is 9% higher than the next highest region, the West Midlands. Across London, the poverty premium costs households in poverty a combined £364 million a year³⁴.

The food-related premium alone affects 44% of London's households in poverty and costs them a combined £109 million a year, and the energy-related premium costs affected London households an average of £161 a year³⁴.

How household income, savings and housing differ across the UK

Poverty rates vary across the four nations, and the differences are consistent across recent reporting periods. Between the financial years ending 2022 and 2024, 22% of people in England and 22% in Wales were in relative income poverty, with England slightly lower at 20% in the later period ending 2025¹⁰. For working-age adults in the latest period, the rate was 19% in England, 18% in Scotland and 13% in Northern Ireland²⁰. For pensioners, it was 17% in England, 15% in Scotland and 13% in Northern Ireland¹⁰.

Savings are thin at the bottom of the income distribution. The Family Resources Survey found that 18% of UK families had no savings in 2024/25, a reduction of two percentage points from the previous year, and that 46% of UK families, equivalent to 17 million families, received no state support⁵. The Joseph Rowntree Foundation reports that 22% of adults in households with annual incomes below £15,000 had no savings at all in 2024, compared with 10% of UK adults overall³⁵, and the FCA's 2024 Financial Lives survey similarly found that one in ten people in the UK have no savings at all³⁶.

Housing tenure has shifted only slowly. The percentage of households in the private and social rented sector combined has remained broadly the same over the last seven years, at 36%²⁹. Wealth, by contrast, has moved unevenly: median household wealth in the North East fell by 7% in real terms between April 2018 and March 2020 compared with the previous period, according to the ONS³⁷.

These differences are why the poverty statistics are published separately for each nation, and why after housing costs figures are the ones most often quoted when comparing places. The same income goes much further in one part of the country than another, and the official measures are designed to show that. For how typical incomes and prices are tracked, see household incomes across the four nations, private rents and inflation.