Unemployment and out-of-work support
Five numbers, five different questions. Only one of them is the unemployment rate.
The measures
4.9%
Unemployment rate, 16+
Not the same as: A count of benefit claimants. It is the internationally agreed ILO definition, measured by survey.
20.9%
Economic inactivity rate, 16–64
Not the same as: People refusing to work. It includes students, carers, retired people and the long-term sick.
75.1%
Employment rate, 16–64
A note on reliability. ONS advises caution with short-term changes in Labour Force Survey estimates. Response rates and sampling have been through significant change, some volatility remains, and the ONS asks users to focus on long-term movements rather than month-to-month shifts.
Five numbers, five questions
These are measured by different bodies, on different definitions, over different areas and on different dates. Swapping one for another is the commonest way a true statistic becomes a false claim.
- Unemployment rate (ILO)4.9%
People without a job who were available to start and had recently looked for work, or were waiting to start one. The internationally agreed definition, measured by survey. UK, March–May 2026.
Does not mean: Everyone receiving an out-of-work benefit. It is a survey measure of labour-market status, not a count of claimants.
- Economic inactivity rate20.9%
People aged 16–64 who are neither employed nor unemployed on the ILO definition — including students, carers, retired people and the long-term sick. UK, March–May 2026.
Does not mean: People who are able to work but choose not to. Most inactive people are students, carers, retired or too ill to work.
- Total Universal Credit caseload8.4m
Everyone on Universal Credit across all circumstances, including people in work. Great Britain, January 2026.
Does not mean: Eight point four million unemployed people. UC supports working households, disabled people, carers and families with housing and children.
- UC “searching for work” group1.6m
The administrative conditionality group required to look for work. Great Britain, January 2026.
Does not mean: The same thing as ILO unemployment. It is a benefit rule category, counted differently and on a different date.
- JSA caseload71,000
People receiving Jobseeker’s Allowance. Great Britain, November 2025.
Does not mean: The number of jobseekers. Most people looking for work are now supported through Universal Credit, not JSA.
Who is actually on Universal Credit
The 8.4 million figure gets quoted as though it were an unemployment count. DWP’s own conditionality breakdown shows what it is: the largest group has no work requirements at all.
- No work requirements (4.2m) — 50%Too ill to work, caring, or already earning above the threshold
- Searching for work (1.6m) — 19%The group required to look for work
- All other groups (2.6m) — 31%Working with requirements, planning for work, preparing for work
Great Britain, January 2026. Source: DWP Universal Credit statistics. Percentages are rounded and may not total exactly 100.
The largest single group — 4.2 million — has no work requirements at all. That includes people too ill to work, carers, and people already earning above the relevant threshold. The group required to look for work is 1.6 million.
What out-of-work support pays
| Payment | Rate, 2026/27 |
|---|---|
| New Style JSA, under 25 | £75.65/week |
| New Style JSA, 25 or over | £95.55/week |
| UC standard allowance, single under 25 | £338.58/month |
| UC standard allowance, single 25+ | £424.90/month |
| UC standard allowance, couple both under 25 | £528.34/month |
| UC standard allowance, couple one or both 25+ | £666.97/month |
The standard allowance is not the award.A household’s Universal Credit can include housing, child, disability and carer elements on top — and can be reduced by earnings, capital, deductions, sanctions or the benefit cap. Multiplying the standard allowance by the caseload does not estimate the cost of Universal Credit.
How much of previous pay does it replace?
OECD modelling for its 2024 UK review estimated that income support after six months of unemployment came to about 14% of average earnings under its standard assumptions. That is a modelled figure for a specific hypothetical household, not what any real claimant receives.
Any international comparison of unemployment support has to state the model household, prior earnings, housing-cost assumption, tax treatment and duration of unemployment. The OECD indicator commonly assumes prior earnings of 67% of the average wage and a specified family type. Change any of those and the answer moves. Source: OECD, Benefits in unemployment: share of previous income.
Claim checks
“8.4 million people are claiming unemployment benefit”
IncorrectWhat is known: 8.4 million people were on Universal Credit in Great Britain in January 2026. 1.6 million were in the group required to search for work, and 4.2 million had no work requirements at all.
“Everyone economically inactive could work but chooses not to”
IncorrectWhat is known: Economic inactivity covers students, carers, retired people and the long-term sick — anyone aged 16 to 64 neither employed nor actively seeking work. It describes labour-market status, not willingness.
“The claimant count is the unemployment rate”
IncorrectWhat is known: They are different measures. The claimant count is administrative and tied to benefit conditionality; the unemployment rate is a survey measure using the ILO definition. They move differently and are not interchangeable.
Related: the benefits system, PIP — a different benefit for a different purpose — and jobs and pay.