
The Department for Work and Pensions’ official statistics show that more than 1,266,000 foreign nationals are now claiming universal credit, the highest number ever recorded. According to an analysis of the data by Reform UK, the total has passed the previous record of 1,165,000 set under the Conservatives, and it has done so on Labour’s watch.
The timing could hardly be more awkward. It comes just weeks after the Home Secretary, Shabana Mahmood, withdrew her plan to make foreign care workers wait 15 years for indefinite leave to remain, preserving a five-year route for qualifying members of the existing cohort and their dependants. Settlement removes immigration restrictions on access to public funds, and the benefit bill attached to that system has now reached a record high in absolute terms. To be clear about the sequence: today’s claimant record reflects arrivals and status grants accumulated over years, not the care worker concession, which has not yet even come into force. But both developments raise the same question, and it is the question this investigation pursues: why has Labour announced the detailed direction of its settlement concessions without yet publishing the full impact assessment necessary to assess their long-term fiscal consequences?
WHAT THE DWP’S OWN FIGURES ACTUALLY SHOW
The DWP’s official quarterly statistics for August 2026 count 8.4 million people on universal credit. Its new Immigration and Nationality section, published as statistics in development, breaks the caseload down by immigration status: 84.6 per cent are British, Irish or hold right of abode, 8.8 per cent are covered by the EU Settlement Scheme, 2.8 per cent hold indefinite leave to remain, and 1.6 per cent are refugees. Applied to the caseload, those proportions produce the totals behind the headlines: roughly 736,000 EUSS claimants, a record 236,000 ILR holders, up more than 50 per cent from 153,000 in July 2024, and 131,691 refugees, up 19 per cent. Excluding the EUSS, the foreign national total stands at a record 530,000.
One point of precision belongs here. The DWP does not publish a single “foreign nationals” count: the 1,266,000 figure is the total of claimants recorded under immigration statuses other than British, Irish or right of abode, and the mapping to nationality is imperfect in both directions. The DWP acknowledges the ILR figure may include naturalised British citizens who never notified the Department, and that a data error briefly mis-categorised up to 34,000 refugees. “Foreign nationals” is the description used by the analysts who compiled the figure, and it is the fairest shorthand available, but this article does not pretend it is an exact nationality count. Nor is it a count of the unemployed: the same release shows 56.9 per cent of EUSS claimants and 39.7 per cent of ILR holders were in employment, and universal credit exists precisely to top up low-paid work. But the published statistics also confirm the trajectory that matters for settlement policy: the ILR group’s share has risen even as the EUSS share falls, and these are the cohorts whose status removes the bar on means-tested support. The fact that this breakdown is still labelled “statistics in development” rather than a routine official series is itself part of the story.
£11.9 BILLION: THE SPENDING PICTURE
The spending numbers are larger still. Figures obtained by the Centre for Migration Control under freedom of information laws show that households containing at least one foreign national received a record £11.9 billion in universal credit payments in 2025, up from £9.5 billion in 2024 and £7.5 billion in 2023, around one in six of all universal credit payments made that year.
Three caveats belong in print alongside the figure, because accuracy is the difference between journalism and propaganda. This is gross household expenditure, not a net fiscal cost of immigration. It measures households, so it includes payments to British partners and children living alongside a foreign national. And only people in the UK legally with recourse to public funds can claim, with many recipients in work and paying tax. None of that softens the trend: a 25 per cent rise in a single year is a 25 per cent rise, and gross spending is what the Treasury pays out. Every pound of public expenditure demands proper scrutiny, particularly when ministers are asking taxpayers to trust that their immigration reforms will reduce the long-term welfare bill. What is missing is not the total but the detail: there is no routine, published reporting of universal credit expenditure broken down by immigration status, and no published Government modelling of how future settlement decisions will feed into it. The public learns these figures through FOI requests and newspaper reporting. That is no way to run fiscal scrutiny of a policy area worth billions.
THE U-TURN, IN ITS ACTUAL NUMBERS
On 2 November, Mrs Mahmood set out her earned settlement model in the Commons alongside the Government’s response to the consultation: a new default qualifying period of 10 years, 5 years for those earning above £50,271 and specified health and education professionals, 3 years for the highest earners, all gated by a spotless criminal record, English at B2 standard, no outstanding Government debt and a record of National Insurance contributions.
Then came the retreat, and the Home Office’s own response document sets out its true scale. The existing care worker cohort comprises an estimated 196,000 workers and 186,000 associated dependants. Of those, the Home Office indicatively expects around 146,000 care workers and 136,000 dependants, together in the region of 282,000 people, to qualify for the transitional five-year route, provided they entered before the May 2025 White Paper, earn £25,000 a year, hold genuine sponsored employment and meet the strengthened requirements. Around 50,000 care workers and 50,000 dependants are indicatively expected to follow the longer pathway, with access to a further five-year Health and Care visa carrying no recourse to public funds. These are indicative estimates of eligibility, not confirmed settlement grants: every applicant must still qualify at the point of decision.
Two things must be stated plainly. Eligibility to apply is not an automatic grant, and the concession is a grandfathering of an existing cohort rather than an open invitation. But the direction is undeniable. A Government elected on the promise of “control” has moved its own settlement crackdown in one direction only: outwards, for a cohort of approximately 282,000 care workers and dependants, under sustained pressure from its own backbenches.
WHAT SETTLEMENT ACTUALLY UNLOCKS
Settled status matters because ILR generally removes immigration restrictions on access to public funds. It does not, in itself, entitle anyone to universal credit: claimants must still satisfy the benefit rules on income, capital and household circumstances. Access to the welfare system and receipt of welfare payments are different things, and this article draws that line deliberately.
The line is precisely what makes the missing assessment so serious. The relevant question is not whether each settlement grant converts into a claim, but whether the Government has published a credible estimate of the additional benefit eligibility and likely expenditure created by its revised arrangements. On the evidence available, it has not.
THE ASSESSMENT THAT HAS NOT ARRIVED
Credit where it is due, and precision demands it: the Home Office has published the consultation response, illustrative cohort estimates and the methodology behind them, and it has committed to a full impact assessment “to accompany the implementation of Earned Settlement”. But the completed assessment setting out the projected fiscal consequences of the care worker concession has not yet been published.
But a promise of future transparency is not scrutiny of the evidence available today. The political decision was announced on 2 November with the completed impact assessment not yet published. The implementation clock is running: the Government intends to begin amending the Immigration Rules from December 2026, with implementation progressing through 2027, and the B2 English requirement takes effect in March 2027. Settlement policy is made through statements of changes to the Immigration Rules which, under section 3(2) of the Immigration Act 1971, are laid before Parliament and take effect without an affirmative vote, with either House entitled to pass a disapproving resolution within 40 days of the statement being laid, as the Commons Library’s briefing on the immigration rules explains. They are not voted on line by line. The separate Immigration and Asylum Bill, which cleared its Commons third reading on 17 November by 290 votes to 94 and now proceeds to the Lords, covers asylum and returns, not this concession. So the honest position is this: the concession has been announced in detail, the rules to enact it begin arriving next month, and Parliament and the public have not been told when they will see the completed fiscal assessment, nor whether they will have any meaningful opportunity to scrutinise it before the relevant changes take effect.
LABOUR’S DEFENCE DOES NOT ADD UP
A government spokesman offered this response to the record figures: “The proportion of universal credit claimants who are foreign nationals has fallen since June 2024.”
The claim deserves to be tested against the right statistics, because the Government’s statement concerns claimants, not payments, and the two are different measures. What the published releases substantiate is narrower than the claim. The DWP’s August 2026 release shows the proportion of claimants in the “CTA: UK, Ireland, Right of Abode” category, which covers British and Irish citizens and right of abode holders, rose from 84 per cent in August 2025 to 84.6 per cent in August 2026, meaning the foreign national share of claimants slipped slightly over that year. Full Fact separately records the share of universal credit payments going to foreign nationals at 15.5 per cent in January 2026, which is an expenditure measure, not a claimant count, and cannot be read across to the Government’s statement about claimants. Whether the claimant proportion has fallen continuously since June 2024 is not directly demonstrated by the published releases.
We report the Government’s defence accurately, and then observe what no percentage can do. A declining proportion of claimants can sit comfortably alongside a record absolute total of claimants and record absolute expenditure, and it is the absolute figures that taxpayers fund: 1,266,000 foreign national claimants and £11.9 billion in gross payments to households containing at least one foreign national, up a quarter in a single year. The Government’s percentages do not contradict these facts. They simply do not address them.
The spokesman added that the reforms mean over 500,000 people will wait longer for settlement, “saving billions from the welfare bill”. Those savings are projections against a baseline the public has not seen; the claimant total and the £11.9 billion are recorded facts. Labour is entitled to blame the Conservatives for the Boriswave, the 4.2 million arrivals between 2021 and 2024 that drove net migration to a record 944,000 in the year ending March 2023. It is not entitled to claim the credit of control while its own concession moves hundreds of thousands of people onto a faster track to public funds eligibility without showing its completed workings.
For British households the consequences are practical. Welfare budgeting is zero-sum at the margin: every unexpected billion must be found elsewhere or borrowed, and an £11.9 billion annual outflow growing by a quarter in a year now sits at the scale of a major departmental budget. Councils commissioning social care and Treasury spending teams alike are being asked to plan without the Government’s own modelling of its most consequential settlement change in a generation. The case for protecting care workers who staffed the pandemic is genuine, and so is the response to exploitation: 1,800 rogue sponsors have been stripped of their licences, and from February 2027 workers who lose their jobs through sponsor licence revocation will have up to 180 days to find genuine employment. But if the exemption is justified by contribution, the contribution should be costed, published and debated. Sentiment is not an impact assessment. And behind all of it stands a wider fiscal picture that puts the scale of the welfare state’s claims on the taxpayer in perspective.
GREAT BRITISH PAC REACTION
Richard Thomson, National Director of Great British PAC, said: “The welfare bill now exceeds what the Government collects in income tax. That is completely unsustainable. Labour is building a dependency class and using ever-greater handouts to secure political loyalty, regardless of who must pay the bill.
“At the same time, billions are going to households containing foreign nationals while British citizens – pensioners, working families and small-business owners – struggle with punishing taxes, soaring bills and the cost of putting food on the table.
“A government that rewards dependency while squeezing those who work, invest and contribute is driving the country towards disaster.”
The official figures give his opening claim its factual footing. Total welfare spending, which includes the state pension alongside universal credit and other benefits, exceeded income tax receipts in 2025/26: the OBR estimated income tax raised £329 billion that year, against welfare expenditure of around £333 billion in OBR figures as checked by Full Fact with the Institute for Fiscal Studies. Much of that bill goes to pensioners and working households, and the comparison is forecast to reverse in 2026/27 before welfare spending climbs again, from £353 billion to £413 billion by 2030/31, according to the Commons Library’s summary of the OBR forecasts. His wider argument about Labour’s intentions is political commentary, reported as his view. What is documented fact is this investigation’s core finding: a record claimant total, rising gross expenditure, and a settlement concession announced before its completed impact assessment was published. Great British PAC will not stand by while ministers make decisions with potentially significant long-term financial consequences without first placing the full evidence before Parliament. Great British PAC demands the immediate publication of the full fiscal assessment and proper parliamentary scrutiny of the Government’s settlement concessions before the relevant rules take effect.
GREAT BRITISH PAC DEMANDS
- Publish the completed economic impact assessment of the earned settlement reforms, including the projected benefit eligibility and public service costs of the care worker concession, before the relevant Immigration Rules come into force.
- Require the Government to provide both Houses with a dedicated opportunity to debate and vote on the care worker and dependants concession before the relevant Immigration Rules take effect, with the completed impact assessment published in advance. Ministers should introduce primary legislation if they wish Parliament to amend the substance of the proposed settlement arrangements.
- Convert the DWP’s Immigration and Nationality breakdowns from “statistics in development” into a standard quarterly official statistic, and add routine reporting of universal credit expenditure by immigration status, ending reliance on freedom of information requests for the spending picture.
- Publish a breakdown of the roughly 50,000 care workers and 50,000 dependants indicatively expected to follow the longer pathway, separating those affected by sponsor exploitation or licence revocation from those in genuine breach, with quarterly reporting on enforcement and removal outcomes for each group.
- Ask the Commons Public Accounts Committee to commission a National Audit Office examination of the net lifetime fiscal cost of settlement at five, ten and fifteen years for the care worker cohort, and lay the findings before Parliament.
- Commit that no further exemptions to the earned settlement model will be announced without the relevant impact assessment being published on the same day, before implementation.
- Publish updated estimates of the number of dependants covered by the care worker exemption and forecast their projected benefit and public service costs, reported to Parliament annually.
THE U-TURN BEFORE THE INVOICE
There is a word for a government that sets a policy first and publishes the completed bill alongside implementation, and that word is not “control”. Mrs Mahmood asked the country to trust that her concession still saves “billions”, while the assessment that would prove it has still not been published, as the first rule changes arrive in December and implementation advances through 2027. What those figures demonstrate is the scale of the existing welfare expenditure that demands proper scrutiny. Households containing at least one foreign national received £11.9 billion in Universal Credit payments in 2025. That is not the projected cost of Mahmood’s concession. And that is precisely the problem: Labour is asking taxpayers to trust its promises of future savings without publishing the completed assessment needed to judge them. Until ministers publish that assessment and give Parliament the dedicated vote on the concession that Great British PAC is demanding, with the evidence in front of it, Labour’s earned settlement promise stands exposed as a crackdown that cracked. Great British PAC will not stop asking for the invoice.


