A risky business

Since 2011, when Sir Andrew Dilnot and his commission published their seminal report on the funding of social care, successive UK Prime Ministers have promised reform but ducked the final decisions. 

That’s perhaps not surprising, because reforming social care is a politically risky business - either confirming the financial burden that individuals have to meet, requiring an increase in taxes, or a reallocation of planned spending. Or, probably, all three. 

Now the latest inhabitant of No. 10 has announced he will make it “free at the point of use”, although the Chair of the latest review, Baroness Casey, is not due to produce her final report until next summer. So what can other countries teach the UK about how to overcome the risks?

Fifteen years of funk

Social care has been a hot potato passed from government to government in the UK for decades.  Since the Dilnot report, numerous further inquiries have been launched.  Reforms along the lines of the Dilnot proposals finally reached the legislative starting-gate in 2024, only for that to be closed at the last minute.

So there’s still no certainty that Prime Minister Andy Burnham, coincidentally confronted with the need for a steep increase in defence spending and nervous bond markets, will finally grasp the nettle.  But there are lessons to be learnt from other countries’ reforms that may help to lessen its sting.

The old problem

The fundamental and universal problem all these countries have faced comes from ageing populations, with longer but not necessarily healthier life expectancy, and falling birth rates. These changes set the pace at which demand for care from the elderly rose over the past quarter-century, and the extent to which it can be expected to rise further over the next half-century, both increasing the total bill and putting upward pressure on individual costs.

Japan has led the ageing trend with the proportion of the population aged 65 and over now nudging 30%. But a number of European countries aren’t far behind. Italy has the highest proportion of elderly people, but in plenty of its neighbours this is well over 20%.  Percentages are lower in the US, UK and Australia, but still rising fast.

Japan’s old age dependency ratio, at over 50%, is the highest in the world. But again, other countries are following in its path.  And - critically for social care - in most it is the numbers of the very oldest which are increasing most rapidly.  In the UK, for example, the number of people aged 90 or over has increased by about 50% in the past 20 years, while the number of centenarians has doubled.

These same countries have also seen falling birth rates.  That’s played havoc with the overall dependency ratio, calculated by including those 15 and under as well as those 65 and over.  Fewer births first lowered the overall ratio, but then began to increase it, as there are fewer youngsters joining the workforce while more of their parents live on into a long retirement. The total dependency ratio is over 70% in Japan - and, for comparison, not far short of 60% in the UK.

Worst for women

What is less often observed is the extent to which women suffer disproportionately from a failure to fix social care. Although life expectancy has been rising faster for men, women still live longer.  According to the UN, worldwide the number of men aged 80 and older rose from only 25m in 2000 to 65m in 2025, though it is forecast to rise to 183m by 2050.  But meanwhile, the number of women aged 80 and older rose from 48m in 2000 to 104m in 2025 and is forecast to reach 263m by 2050. 

Moreover, longer female lives do not seem to have led to better health in old age, since the UN reports that women face a higher incidence of Alzheimer’s and other neurological conditions.  And to cap it all, once past the age of 60, more women live alone - twice as many women as men in the UK.  Living on your own is, obviously, a situation much more likely to lead towards a need for care.

The final gender inequality is that most unpaid care is delivered by women, who compound their disadvantage by dropping out of the labour force to care for the elderly, reducing their pension entitlement and hence increasing their risk of poverty in old age.

Crisis in common

No advanced country is immune from these pressures, while they are emerging rapidly in the rest of the world.  Every advanced country has therefore been struggling with the twin problems of funding and providing care.

Despite funding reform and a sophisticated approach to caring for old people, Japan is still struggling with costs and labour shortages.  In August, the IFO Institute published a bleak analysis of the rise in the German budgetary burden, as there too reform has not prevented costs coming under acute pressure.  However, the UK displays a particularly troubling combination of pressure points.

• A shortage of readily available care places, and complex assessment processes stemming from the disconnection between the NHS and social care systems, are trapping many old people in hospital beds and causing A&E departments to overload.  This has resulted in a notorious increase in the number of patients receiving distressing, inadequate and even life-threatening care in hospital corridors.

• An increase in labour costs for care homes, accelerated by higher national insurance and the minimum wage, and compounded by a rapid drop in legal immigration, is causing homes to raise fees sharply or close.  This puts pressure on council budgets and those paying privately for residential care (now over half the total), often charged significantly more than councils and resenting the implicit cross-subsidisation. 

• A cap of £23,250 on assets is all (in England, at least) that is disregarded before requiring individuals to pay for social care.  According to Carehome, the UK online directory and review platform, fees for such self-funders now average £1,298 per week for residential homes and £1,535 a week for nursing homes.

• This assets cap has the further knock-on effect of making more families, faced with high social care costs for anyone with assets over the limit, reluctant to see their parents leave (free) hospital care.  Discharging them becomes even more difficult for an overstretched NHS.

• A rapid increase in demand for residential care for younger people, at a higher cost per resident than for the elderly, has put further upward pressure on the social care bill.  Costs for this rapidly-increasing group now amount to almost half the total.

Life’s a risky business

The social care problem is a compound of old age, lifestyle and luck.  Some people live into their nineties able to care for themselves; others become unfit to do so long before that; still others die earlier from accidents or critical illness.  However, the risks of needing urgent medical care in a hospital, and needing “social” care support, are treated rather differently almost everywhere. 

So far as the urgent health care is concerned, most countries provide medical care that is not means-tested but available to all, either free at the point of use or covered by social or private insurance.  For ongoing social care of the frail elderly, systems vary much more, and there is much more “self-funding” out of savings.  This may originally have reflected the principle that people should continue to meet the day-to-day living expenses they would have had to meet anyway, although most principles have been lost in the maelstrom of accelerating demand and fast-rising costs. 

The table below sums up the broad approaches used by advanced economies. A key element is the extent to which the risks to the individual are “pooled”. 

In principle, pooling the risk of long-term care, so that no one has to pay for more than the average risk, looks a good idea.  However, the extent of the advantage to the individual depends on how pooling is achieved.  If risk is pooled through taxation, the better-off will pay much more than the average cost; if it is pooled through private insurance, the cost will depend only on the probability of need, and risk will be shared only between these who can afford the premiums.  Flat-rate levies and fixed payroll percentages (like the UK’s national insurance system) fall somewhere between the two.

Most European systems are based on the social insurance model, following the “Bismarck” approach to social security (as opposed to the “Beveridge” state-based model developed in the UK).  However, in most of them taxpayers have had to top up insurance funds, as well as paying for those who cannot contribute, and many of them also levy direct user charges.

Germany provides perhaps the most interesting example.  The reforms introduced in 2017 combine social insurance, a state safety-net and a cap on user charges. Universal modest user charges blunt the difference between state and private funding, while social insurance pays a proportion of fees.  A key feature is that the remaining fees are on a sliding scale, declining with your length of residence, and so blunting the risk of needing to pay for very long-term care.

This model is complex enough to provide a number of levers to pull when funding goes wrong: higher user charges, higher contributions, a bigger state subsidy.  But complexity has its disadvantages too - and there have been tensions over different elements since the scheme’s introduction.  Moreover it has not saved the German government from having had to provide hefty top-ups over the past few years. 

Japan implemented major reforms to social care in 2000, after a sustained campaign to achieve social acceptance.  The construction of a system to provide for a uniquely high proportion of the population was exceptionally challenging, and the introduction of technological solutions, including the use of humanoid “caring” robots, as well as the emphasis on lifestyle programmes for the elderly, are leading-edge.

Japan too follows a mixed approach of social insurance, state funding and user charges. Contributions to care costs are means-tested. However, the test at present only relates to income and ignores assets, a limitation that is coming under pressure as costs mount. And the shift from informal to formal care, intended to reduce the burden on women family members, has inevitability added to the upward pressure on costs.

The US follows the private approach, with social care covered by private savings or insurance and a Medicaid safety net. However, some states are beginning to introduce mandatory social insurance schemes, as these safety nets begin to be overwhelmed.  The US probably has the most developed private insurance market in the world, but it is still patchy because the larger health insurance market largely steers clear of social care risks.

The present English system is tax-funded and means-tested.  Since over half of care and nursing home residents have assets above the £23,250 cap and therefore now have to self-fund , the system has come to look less “Beveridge” and more “American”.  But the multiplicity of risks relating not only to health but to policy uncertainty has meant that only a very limited private insurance market has developed, mostly in the form of deferred payment schemes based on the high level of residential property ownership in the UK.  There is no risk-sharing across society for anyone above the asset ceiling for the safety net. 

The Welsh and Scottish systems are more generous, provoking English irritation with the “Barnett formula” that provides for higher public spending in these territories.  In Scotland, notably, the asset cap before self-funding is required is £50,000, and some care home costs are met for everyone – a system to which Prime Minister Andy Burnham seems to be turning his attention.

A further challenge faced by many systems is that medical care is still, at least in theory, provided free at the point of use irrespective of means, or funded by health insurance (as in the United States). Defining and maintaining this boundary between different needs and funding sources adds complexity and administrative cost, compounded in some countries by a distinction between “hotel costs” and “care”.

Finally, policymakers struggle to balance care packages for people in their own homes against subsidies for residential care.  Care at home is usually much preferred by the individual, less expensive, and more likely to be state-funded, but in Japan, as noted above, one of the drivers of reform was resentment that the emphasis on home care offloaded the problem on to (unpaid) women family members.

Changing caps

In the UK, the erosion by inflation of the protected assets cap has made it more and more of a sore point. Failure to raise the asset cap in England since 2010 was recently described by the King’s Fund as a “stealth tax”. The Dilnot model and subsequent variants not only proposed increasing this, but also introduced the idea of a totally new kind of cap.  

As well as a cap on assets that individuals didn’t have to realise to pay fees, Dilnot proposed the introduction of a cap on the total amount that anyone might be asked to pay, irrespective of their length of time in care. The report proposed a limit on payments of £35,000. In the latest version - the plan dropped in 2024 - the maximum an individual might be asked to pay for social care had been raised to £86,000. 

Setting this kind of cap would, it was argued, not only reduce fear among the elderly of “catastrophic” long-term care costs, but also reduce some of the risk to financial firms of offering social care insurance.  Fixing a ceiling on the amount to be insured against would, it was believed, encourage the development of insurance policies, and the Dilnot commission gave examples. Such a reform was intended to enable (private-sector) risk-sharing up to that defined limit on individual charges, combined with (public-sector) risk-sharing above it.

The cost falling on both individual and state would of course depend critically on the level of the cap.  The higher the cap, the smaller the bill to taxpayers, and the larger the sum to be insured against by individuals. At £86,000, it was officially estimated that the cost to the state of introducing this reform in England would plateau at about £4bn a year. By contrast, the Health Foundation has estimated that funding all social care in England would cost £18.5bn and rising, while introducing the Scottish model would cost £7.5bn.

Of course, as well as receiving much support, the Dilnot approach has had its critics, particularly since the main beneficiaries of a costs cap would be families with at least some assets, notably homeowners. And some were sceptical about the extent to which financial markets would develop good-value insurance products, partly because of “adverse selection” (i.e., those most likely to need long-term care would be the most willing to insure against it).  Still others (now seemingly including the Prime Minister) favoured the Scottish system, although since the “hotel costs” of care homes account for about 80% of the total, this still leaves the personal risk of “catastrophic costs”.

As a result, some have continued to argue for compulsory social insurance. Still, others have stood out for leaving the present regime more or less untouched, arguing that it was wrong to ask the taxpayer to take on any costs that could be met from private savings.

In all the subsequent reviews, no one has so far come up with a more comprehensive set of policies than Dilnot’s. But the fact remains that no government has so far been prepared to put them in place.  While Baroness Casey makes another attempt to tackle the dilemma, a summary of what has meanwhile been learnt elsewhere may provide some useful prompts.

Mix or match?

From any survey of the main features of different national approaches to social care, a number of lessons stand out:

• All systems in developed economies are facing cost pressures, as labour shortages (and a growing hostility to immigration) put upward pressure on wages.  In the UK, the Casey Commission’s emphasis on improved pay and careers for social workers may help attract recruits, but will also put upward pressure on fees.  The funding analysis will need to recognise this.

• Almost every social care system has therefore required emergency injection of state funds over the past three years.  And these have not gone to benefit care home residents meeting their own fees. No reform will be long-lasting without addressing the inequity between state and privately-funded places.

• The costs of reform are almost always underestimated because improvements in the system release suppressed demand, reveal unmet demand or simply encourage more. Germany perfectly illustrates the Catch-22 of reform: a recent report by the International Longevity Centre (ILC) on Germany, Japan and Australia[1] highlights the surge in demand that followed reform in 2017. Modelling work must make due allowance for the behavioural effects of reform.

• Even in a sector that self-evidently relies on human interaction, you can’t afford to ignore any potential for technology to ease costs.  Even if you don’t like the thought of being cared for by Paro (Japan’s therapeutic seal robot), the use of humanoid robots as a substitute for care workers is spreading to other countries (mostly in Asia). A full understanding of the potential and early testing of applications is essential to the design of a system fit for the future.

• The easiest technology wins are in the provision of equipment (some of it actually quite low-tech) to help people stay in their own homes. Meanwhile the use of AI could self-evidently reduce the administrative costs of all the necessary assessments of health and means and the interface between different health and social services. Again, an enhanced understanding of the potential, backed by thorough testing, is vital to system design.

• Many countries struggle with the interface between health and social care, often with one provided nationally and another at the local level. Germany, again, offers a case study in the difficulties.  As an increasing proportion of the population comes to need both, boundaries need to be lowered and cross-working made easier.  Proposals for devolution and local authority reorganisation in the UK needed to be rigorously tested against this requirement for seamless provision.

• State-imposed compulsory elements (from taxes to payroll levies) achieve greater acceptance if the connection to social care costs is clear, and payment is calibrated to perceived need (in Japan, for example, such levies are only paid by the over-40s.)

• A multiplicity of funding sources makes for flexibility in budget management - when costs over-run there are more levers to pull.  However, this militates against simplicity and an ease of understanding, found to critical in reforms elsewhere (Australia provides a useful case-study of the need to test understanding and acceptability.) 

• A key insight from the ILC report mentioned above is that reforms work better when they go with the grain of the pre-existing systems in the countries they concern, or represent easy-to-understand changes to them. Wholesale reforms based on unfamiliar concepts are much less likely to succeed.  Moreover, the durability of reforms depends on wide political acceptability, requiring a balance between tax-based funding and private needs-based payment.

Back to the future 

The authors of the Dilnot report understood that the system for social care had to address the fears of individuals, both of being unable to pay at all, and of facing the catastrophic costs that would come with a long period in care.  Its key insights were the advantage to be had to the individual from risk-pooling, together with the disadvantages of seeking to provide all that risk-pooling through taxation. Its disadvantage was that successive governments, and Chancellors, have been unwilling to put them in place.  

Self-evidently, any scheme to fund some or all of social care through tax-funded public spending needs to be acceptable to taxpayers.  The cost of introducing a cap on total payments could be reduced further by setting it higher than £86,000, so that it would only cover (say) those in care for more than two years.  The problem of the existing £23,250 cap on protected assets is more difficult to solve cheaply.  And a sustainable reform needs to look both at the current social care bill and the way in which a younger generation is required to prepare for its impact on their own lives. Prime Minister Andy Burnham has indicated that reform could be introduced slowly, on a timetable that would allow it to be at least part-funded by ending the pensions “triple lock”.  But that, of course, would not address the demand to end “catastrophic costs” now.

A quick look at other countries shows how they seek to share the burden between tax-funded subsidies and private contributions - the main difference being whether these contributions take the form of pre-payment through insurance or user charges at the time, or a mixture of the two.  Many try to set a maximum cost, or a sliding scale.  But further differences lie between “insurance” which is in reality taxation by another name, or a genuine funded system for meeting costs with contributions; and between “contributions” which are based purely on the probability of need or are income-based.   

The challenge for the UK is to devise proposals that are easily understandable, based on familiar concepts, acceptable to taxpayers and therefore likely to prove politically durable.  They will need to be tested not just against the insights in the Dilnot model, and the lessons from the failure to get agreement for its implementation, but against the experiences of other countries with their similar attempts at reform.