I have a brilliant idea for a new tax. The rules are simple:
- Pick a card. On your 65th birthday you shuffle a deck and draw one card. An ace or a king, and congratulations: one day you will pay my new tax. Any other card, and you will never pay a penny of it.
- You are not told when. The tax might start at 68. It might start at 95. It might start the week after your wife dies. I know, but you don’t. There’s no way to find out in advance, and no way to insure against it.
- The wildcard. An official decides whether to give you a wildcard that will exempt you from the tax. All the officials are different – some only hand out wildcards 3% of the time; others 60%. The decision is based on an artificial distinction that you are unlikely to understand, and is nothing to do with your financial position. If the official gives you the wildcard, you leave the game – you’re exempt from the tax and pay nothing at all, however rich you are. If you don’t get a wildcard, you are still in. Time to pay.
- Monkey throws darts. A monkey throws a dart (with your name on it) at a board full of numbers. It lands somewhere between £58,000 and £66,000 – and that’s your tax bill for that year. The monkey hates London and the South of England, so adds about 40% extra for Londoners, and 25% extra if you live elsewhere in southern England.
- Monkey keeps throwing. Half of all players are finished inside fifteen months. But more than a quarter keep spinning for over three years, and some more than twenty. Each year, the monkey throws a new dart, and you get a new bill.
- Pay from your income. Then from your assets. If your income won’t cover the tax – and it probably won’t – you have to sell stuff to fund it.
- Under £14,250 of assets: your council pays the tax instead of you. Which of course goes onto everybody’s council tax.
- Between £14,250 and £23,250: you pay £1 a week for every £250 you have over £14,250. Your council pays the rest.
- Over £23,250: you pay the whole of the monkey’s number yourself, every year, until you die.
- Rich losers pay poor losers. I thought it would be fun to inject some socialism into the randomness. So about 30% of the tax paid by rich losers (who are paying in full) goes to people who also lost but don’t have enough income/assets to pay themselves.
- Your house is in play. It doesn’t count as an asset while you, your partner, or an elderly relative still lives in it. Once you have all moved out, you get twelve weeks, and then it very much counts.
- Pocket money. You may lose almost everything you have ever owned. But don’t worry: you get to keep £31.80 a week – don’t spend it all at once.
This is, obviously, insane.
Except it’s real
The above description is, near enough, how catastrophic adult social care costs work in England today.1
The deck of cards is the DHSC’s own estimate that “one in seven individuals over 65 will face care costs above £100,000 and roughly one in ten individuals will face care costs above £120,000 over their lifetime.”2
The monkey throwing the darts is the going rate for a care home if you are paying for it yourself. Self-funder data published by carehome.co.uk puts the average weekly fee in North East England – the cheapest region in the country – at £1,112 for residential care and £1,264 for nursing care. In other words, £58,000 to £66,000 a year. Everywhere else is more expensive.3
The spins come from the largest study of its kind: 11,565 people who died in Bupa care homes over an eighteen-month period. Mean length of stay was 801 days; median 462 days; longest, 8,725 days.4
The wildcard is NHS Continuing Healthcare. If you are assessed as having a “primary health need”, the NHS pays for everything, including your accommodation, and there is no means test at all. Same care needs, same care home, same bill – but if an assessor decides your needs are medical rather than social, you pay nothing, and if they decide the opposite, you can lose your house. Age UK found that the proportion of assessments resulting in an award “varies from 3.4% to 57.9%, depending on where you live”.5
The means test is just quoting the rules. £14,250 and £23,250 are the statutory capital limits, and they have not changed since 2010 – it’s a fiscal drag that nobody talks about. £1 a week for every £250 in between is the statutory tariff. £31.80 a week is the personal expenses allowance for 2026/27.6
Socialism for the losers. Councils use their buying power to push care home fees down to something close to cost, so homes recover the difference from the people paying their own way. The Competition and Markets Authority found self-funders “typically paid some 40% more than local authority residents for the same care”.7
For all the pedants out there, of course I understand that England’s social care system isn’t legally a tax – it’s a means-tested charging system. But economically it works exactly like the mad tax described above: a random, catastrophic liability lands on the people unlucky enough to need care. If they can’t pay, the bill lands on their council, and so onto local council taxpayers.
We are all losers
People who win the game or haven’t started playing it also lose, because adult social care now dominates many council budgets.
In 2025-26, English local authorities budgeted £26.7 billion for adult social care: almost one pound in every five of all service expenditure.8
The left-hand chart shows all English local authority service spending. The right-hand chart strips out education, police, fire and public health. On that narrower measure, adult social care is almost 40% of the remaining budget. Children’s social care is another 23%. No wonder we have potholes.
This is, however, just the start. As demographic pressure builds, the Office for Budget Responsibility expects adult social care spending to rise “from 1.2 per cent of GDP in 2030-31 to 1.8 per cent of GDP by 2075-76” — half as much again, measured against everything the country produces.9
This was fixed twelve years ago. And then it wasn’t.
Parliament legislated for a cap on lifetime care costs in the Care Act 2014. Like most legislation, it just needed a minister to make a commencement order to bring the cap into force. Unlike most legislation, no commencement order was ever made.
So there is no cap.10
What’s the solution?
Unfortunately I can’t pretend to have an answer, not least because it’s well outside my expertise (the underlying analysis behind this article was contributed by a local authority care specialist).
But the menu is short:
- Keep the status quo, and accept that we all play the game, and one in seven of us loses (and potentially loses everything they have).
- Reduce the cost for the unlucky one in seven and raise it for everybody else, through direct taxation – as in Denmark or Sweden.
- Create a compulsory social insurance scheme, with subsidies for people on low incomes, on the model of Germany or Japan.
All three cost money. Only the first one also makes us collectively poorer for no reason at all.11
Theresa May deserves far more credit than she gets for having tried to do the right thing. I hope Andy Burnham will be equally brave, and rather more successful.
Many thanks to S for their help on the social care rules, and to B for help with the analogy.
Photo by National Cancer Institute🔒 on Unsplash🔒.
New data suggests Scotland’s 48p tax rate may be losing money
John Healey should cut National Insurance, not raise the personal allowance
What would a land value tax actually do?
What if Andy Burnham lowered the mansion tax threshold to £1.5m?
How Andy Burnham could raise £15bn – without a tax rise.
The Palace’s fake tax transparency — and the Prince of Wales’ real tax problem
Footnotes
The game covers catastrophic care costs only — the ones that wipe people out. It ignores the much larger number of people who face smaller care costs: DHSC’s charging-reform impact assessment says “around three out of four adults over the age of 65 will face care costs in their lifetime”, but for most of them that means a few hours of help at home, at a fraction of the cost – and the value of your home is always disregarded when you are being cared for in it. This is also England only. Scotland, Wales and Northern Ireland each do it differently. ↩︎
Drawing an ace or a king is 8 cards in 52, or 15.4% – as close to one in seven (14.3%) as I could manage without making the rules too complicated for me. You could replace it with an annual lottery, but then I can’t use a pack of cards, and I like cards. ↩︎
Weekly self-funder fees from carehome.co.uk, dated 9 September 2025. Against the North East, the average of the South East, East and South West is 24% higher for residential care and 29% higher for nursing. London is 39% higher for both (£1,548 residential, £1,759 nursing — about £80,000 and £91,000 a year). ↩︎
Julien Forder and José-Luis Fernández, Length of stay in care homes, commissioned by Bupa. Around 27% of residents stayed more than three years. At North East prices a mean stay costs roughly £127,000 for residential care and £145,000 for nursing; in London, comfortably north of £175,000. Which is how you get to the DHSC’s one in seven above £100,000. Note that the study was published in 2011, and things may have changed since then (although it is not clear why they would). The study was, of course, limited to Bupa care homes, which probably biases the data in a variety of complex ways. The study only covers those who died in care homes, not those who left, although I’d expect this to be small for permanent admissions. ↩︎
Age UK, Parliamentary Briefing: NHS Continuing Healthcare, December 2024. Age UK calls it an “extreme postcode lottery” which “brings into question the idea that CHC is a rules-based system which is not influenced by the state of local NHS finances”. The number of people newly found eligible on the standard pathway fell by 43% between 2017/18 and 2023/24. ↩︎
All four figures are from DHSC’s local authority circular on charging for care and support, 2026 to 2027: “the PEA will increase in line with inflation from its current level of £30.65 per week to £31.80 per week”, and “£23,250 for the upper capital limit and £14,250 for the lower capital limit”. 2026-27 is the sixteenth consecutive year the capital limits have been frozen; Community Care calculates that “had the thresholds been increased in line with prices since 2010, the upper threshold would now be worth about £36,400 and the lower threshold £22,300”. You can avoid losing the house in your lifetime by asking for a deferred payment agreement under the Care Act 2014, under which the council pays the fees and registers a legal charge over your property, recovering the debt with interest from your estate. It stops a forced sale in your lifetime. It doesn’t stop you (or your children) losing the house. ↩︎
House of Commons Library briefing CBP-8003, Social care: care home market – structure, issues, and cross-subsidisation (England), summarising the CMA’s November 2017 care homes market study. The CMA declined to ban the practice, calling instead for more money for councils so they could pay realistic fees. ↩︎
The charts use England 2025-26 budgeted net current service expenditure from MHCLG’s local authority revenue expenditure and financing release. The full denominator is £138.668 billion. Adult social care is £26.701 billion, or 19.3%. Children’s social care is separate: £15.549 billion, or 11.2%. The narrowed denominator excludes education, police, fire and public health. That leaves £68.005 billion. Adult social care is then 39.3%. Children’s social care is 22.9%. Note that this is a conservative picture of the pressure on councils. MHCLG’s adult social care category excludes some NHS funding. DHSC’s 2024-25 adult social care finance report puts gross current expenditure at £29.4 billion and estimated net public spend at £27.8 billion. ↩︎
Office for Budget Responsibility, Fiscal risks and sustainability, July 2026, baseline scenario. The OBR notes that “social care spending is more affected by population ageing than health spending given its concentration at the end of lives”. ↩︎
The full history. In July 2011 the Dilnot Commission’s report Fairer Care Funding recommended a £35,000 cap on what anyone over 65 should have to spend on their own care. The Care Act 2014 created the framework, but the relevant sections were never commenced. In September 2021 the then government announced an £86,000 cap, to start in October 2023, alongside a much more generous means test: the upper capital limit rising from £23,250 to £100,000 and the lower limit from £14,250 to £20,000. That £86,000 covered personal care only, so “daily living costs” – food, rent, heating – sat outside it and carried on forever. That is not unreasonable, given that we all cover our own daily living costs. The government then amended the Care Act so that only money an individual paid themselves counted towards the cap, and not their council’s contributions. The House of Commons Library records drily that “this change proved controversial”: it meant poorer people took very much longer to reach the cap than richer ones. A separate reform would have brought section 18(3) of the Care Act fully into force, letting self-funders buy at council rates and ending the cross-subsidy described above, backed by £1.36 billion to help councils pay a “fair cost of care”. In November 2022 the whole package was delayed by two years to October 2025, and most of the fair cost of care money was reallocated. On 29 July 2024 the Chancellor, Rachel Reeves, cancelled it outright as part of her response to a projected £22 billion overspend. It is now with the Casey Commission, which reports in two phases: the first in 2026, the second by 2028. Source: House of Commons Library briefing CBP-9315, Proposed adult social care charging reforms (including cap on care costs). ↩︎
i.e. because whilst on the surface this looks like a zero-sum game, it isn’t. The present arrangement has costs that aren’t transfers to/from anybody. People hoard assets they would otherwise have spent or given away, in case they draw the ace or king. People turn down care they need because of what it will do to the house. Self-funders pay a 40% premium for the same room as the person next door. Councils spend a fortune administering a means test on people who are dying. None of that money ends up in anybody’s pocket. It is pure waste, and any kind of risk-pooling system would prevent it. As DHSC’s own impact assessment put it, it’s a market failure. ↩︎
Leave a Reply