A care service with no members: what Britain is actually proposing

On 29 September 2026, in Liverpool, the Prime Minister told the Labour conference that England would get a National Care Service "in the best traditions of this party, that everyone contributes and everyone is covered, free at the point of use and no care charges paid out of your basic state pension". It is the twenty-third attempt to settle adult social care in under thirty years, and the first to name the money on the same day.

Social care is the help adults need when age, illness or disability makes daily life hard: getting washed and dressed, eating, taking medication, moving around the house, and for some a place in a care home. In England it is arranged by councils, and unlike the NHS it has never been free. Some 889,000 people received long-term care from their council in 2024-25, and many more pay for it themselves or rely on family. It is the part of the post-war settlement that was left to a means test in 1948 and has stayed there. This briefing sets out what has been promised, what is being reviewed, what the system looks like now, and what the design so far leaves out.

What was promised

Three things were said, and they are not the same thing.

The first is a benefit. Personal care, meaning help with washing, dressing, eating and medication, would be free at the point of use for older people, starting in the home and extended in phases. Bed and board in a care home would still be paid for by the resident. This is the Scottish model, in place north of the border since 2002 for over-65s and since 2019 for everyone, not the NHS model: the NHS does not charge for the hospital bed. On 22 October Downing Street confirmed, according to Disability News Service, that the offer will apply only to people over 65; working-age disabled adults, about a third of those receiving long-term care (313,000 of the 889,000 in 2024-25), will wait for the commission described below.

The second is a protection. Nobody would pay care charges "out of your basic state pension". In practice that is a floor under the means test, not the abolition of it.

The third is the money. From April 2030 the state pension would rise each year by prices or 2.5 per cent, whichever is higher, instead of the current triple lock that also tracks earnings; officials say it would still be topped up "in some years, to ensure it holds its value relative to earnings". Labour's own figures put the saving at about £15 billion a year by the end of the 2030s and £50 billion a year by 2050. The Prime Minister said the service would be "fully funded and not through borrowing", that he would "stick to the manifesto" on income tax, employee National Insurance and VAT, and that he "may pay a political price". The change goes to the electorate in the next manifesto; nothing in it is legislated.

What is being reviewed, and when

The Prime Minister has said the design will come from review. There is, in fact, one commission with a two-part public consultation inside it, and a workforce stream running alongside.

Baroness Casey's Independent Commission on Adult Social Care was set up in January 2025 to report twice: a first report in 2026 on medium-term improvements and a second in 2028 on long-term transformation, including funding. On 29 July 2026, in his first major speech on care as Prime Minister, he collapsed the two phases into one and asked for final recommendations by summer 2027, a year early. Alongside it Casey launched the "Big Conversation on Care", running from July 2026 to April 2027 and aiming to reach a million people in England. Its first phase, to November, is about values, principles and who should be eligible; its second phase, over the winter, is about "how the country should pay for care fairly and sustainably". That second phase is the only formal window in this parliament in which the contribution side of the bargain is open for argument.

The workforce stream is a ministerial group under the Health Secretary with two other cabinet ministers, to turn the care-sector fair pay agreement into a broader settlement that moves care pay towards NHS scales with progression routes into the NHS, from 2028-29. Cross-party talks with the Conservatives and Liberal Democrats were opened the same day; the Prime Minister said "we're going to have to find common ground here if this is to be a lasting solution." The sequence, then, is: consultation to April 2027, commission by summer 2027, manifesto by 2029, pension change from April 2030, service built through the 2030s.

Scotland is the cautionary tale. Its National Care Service Bill was introduced in June 2022 and abandoned in January 2025 after £2.2 billion of projected cost over ten years and the collapse of the governing coalition; what survived was a non-statutory advisory board.

The system as it stands

The numbers describe a service already being run, and already being paid for, by the people who use it.

In 2024-25 English councils received 2.02 million new requests for support, 1.30 million from people over 65. During the year 889,000 people received long-term care, 672,000 of them on 31 March 2025, two thirds at home. Councils' gross spending on adult social care was £29.4 billion, up 9 per cent in a year; £23.6 billion of that was long-term care. Against it councils collected £9.7 billion of income, and the largest single line is now the people being cared for: client contributions reached £4.68 billion, up 14 per cent, overtaking the NHS as councils' biggest source of care income. In 2024-25 a week in a residential or nursing home cost councils £1,185.55 and an hour of home care £23.56, both up 7 per cent.

The means test that produces those contributions has an upper capital limit of £23,250, unchanged since 2010-11. Had it risen with inflation it would have been £33,654 by 2024-25. Anyone above it pays the full cost until they fall below it; the lower limit, under which capital is ignored, is £14,250. Casey's own launch material counted 1.46 million paid care workers and 4.7 million unpaid carers in England. On a day in June 2026, 13,600 people medically fit to leave hospital were still in a bed because there was nowhere to go.

One line in the accounts matters for the argument that follows. Councils spent £1.75 billion of long-term care money as direct payments, cash paid to the person to arrange their own care. That is 7.4 per cent of long-term spending: 24 per cent of community care spending on working-age adults, but only 16 per cent for people over 65. Individual funding exists in England. It is small, it is shrinking as a share for older people, and each payment is held by one person alone.

The money, read against the promises

The Health Foundation's estimates, published in July 2026 ahead of the speech, are the reference figures. Scottish-style free personal and nursing care for over-65s would cost about £6.5 billion in 2026-27, rising to around £7.5 billion by 2035-36. The cancelled £86,000 cap with a more generous means test would cost about £0.5 billion now, rising to about £4 billion. A universal, comprehensive, NHS-style service would need an additional £18.5 billion a year by 2035-36. The Prime Minister called that last figure "not that high".

The pension change produces its £15 billion only by the end of the 2030s, and the Institute for Fiscal Studies has pointed out that a formula which "holds its value relative to earnings" over time saves nothing relative to the earnings link in the long run; its director called the new formula "an improvement but still imperfect". A market commentator put the timing problem plainly: if the spending arrives years before the savings, Britain borrows the difference. The Nuffield Trust warned the Prime Minister to "tread carefully"; the King's Fund said "many genuine questions remain about how these reforms will work in practice and how much it will cost". Unite called the pension change "electoral suicide". None of this is new. Every previous attempt died at the moment the contribution was named: the 2010 white paper's estate levy, christened a "death tax" within days; Dilnot's £35,000 cap in 2011; the £72,000 cap legislated in the Care Act 2014 and never switched on; the £86,000 cap legislated in 2021 and cancelled in 2024.

What the design leaves out

Contributory systems that have lasted share three parts: a contribution, a defined benefit, and a member who can read both sides of the book. Beveridge's design had all three for sickness because it took over a working mutual tier; the approved societies had run sickness benefit for decades. Care was left out in 1948 because no such tier existed for it, and it was handed to families, parishes and a means test. Seventy-eight years later the state is proposing to build the benefit and the tax, and the member is still missing. There are three places in the current debate where that absence is a live design question.

The contribution side is open until April 2027, and it is being designed as a tax. The triple-lock saving is general revenue; it has no members, no fund, no account a contributor can read. The countries Britain is compared with did it differently. Germany's long-term care insurance of 1995 is a contributory fund with a defined benefit and a visible shortfall, which is why its members know to the euro that their co-payments rise by half in January and that the pool runs short in October. Japan's 2000 scheme collects from everyone over forty into a ring-fenced fund. France has pushed €1.4 billion of health cost onto its mutuelles, badly, but onto bodies that have members and annual meetings. The Netherlands has bread funds that its own legislature declines to recognise. Each of those systems has a tier between the Treasury and the household. Britain's Big Conversation asks in its second phase "how the country should pay for care fairly and sustainably", and the honest answer from the comparators is that a tax alone has never been enough to hold a care promise through three parliaments, and that a contributory tier with members has held in Germany for thirty years through four benefit cuts. The question worth putting to the Commission is simple: will the National Care Service have a fund, and will contributors be able to see it?

Individual funding already exists, and it is individual to a fault. A direct payment is a personal budget with nobody to pool it with. £1.75 billion is paid out each year to people who then each hire, each insure, each cover a sick week alone, and each lose the arrangement when they go into hospital. Australia's National Disability Insurance Scheme is the same instrument scaled to a nation, and it is being trimmed because the entitlement was written first and the pooling never was. The Care Act's statutory guidance already allows recipients to pool their direct payments; take-up is so small that nobody counts it, because the whole administrative burden lands on the household. The opportunity here is specific. A street or a condition group holding its direct payments in common, with a book every member reads, can buy respite, share personal assistants, cover the gap weeks and keep the arrangement alive when one member is in hospital. That is a friendly society formed from money the state has already agreed to pay.

What stays uncovered is the insurable part. The promise excludes bed and board, protects only the basic state pension, and leaves the £23,250 test in place for everything else. The cost left with the household is the accommodation bill, £1,185 a week at council rates and more at private ones, falling on a minority of households for a period nobody can predict. That is a textbook pooled risk, and Britain has no instrument for it: Dilnot found in 2011 that there was no private long-term care insurance market and that one would not form without a cap. Fifteen years on there is still no cap and still no market. A mutual tier, members paying small regular contributions into a pool nobody holds, with the claims rule written by the members and the ledger open, is the missing instrument between the tax and the means test. It is what the friendly societies did for sickness before 1911 and what the German fund does for care today. Nothing in the Prime Minister's speech rules it out, and nothing in it builds it.

There is a reason to think the door is ajar. Before Westminster, the Prime Minister spent a decade in Greater Manchester on neighbourhood working and the "Wigan Deal", a settlement in which the council stopped being a top-down provider and made a bargain with residents about what each side would do. The language of his July speech, "citizens' needs centred, removal of budget silos", is that tradition. A National Care Service built in that tradition would have a place for pools that communities hold. One built as a Treasury line, funded by a pension formula, would not.

Sources: Prime Minister's speech to the Labour Party conference, Liverpool, 29 Sep 2026 (as reported by Al Jazeera, LabourList, LocalGov, Politics.co.uk, City AM, Yorkshire Post); Disability News Service, 22 Oct 2026; Prime Minister's speech on adult social care, 29 Jul 2026 (HFMA, Care Home Magazine, CMS Law briefing); Independent Commission into Adult Social Care, "Big Conversation on Care" launch material, Jul 2026; House of Commons Library, "Adult social care reform in England", CBP-12218 (Health Foundation estimates, Jul 2026); NHS England, Adult Social Care Activity Report 2024-25 and Adult Social Care Finance Report 2024-25; King's Fund, Social Care 360 (2025); IFS commentary on the triple lock change (Sep 2026); Commission on Funding of Care and Support (Dilnot), 2011; Care Act 2014; National Care Service (Scotland) Bill record.