Unfortunately, your browser is too old to work on this website. Please upgrade your browser
Skip to main content

Key points

  • Adult social care in England needs fixing – and has done for decades. Increasing numbers of people are unable to access social care and care providers are at risk of collapse. Yet successive governments continue to duck reform, and people and their families continue to suffer unnecessarily.
  • Based on our assessment of the evidence and analysis of the costs of reform, five priorities for government are:
  1. stabilising and sustaining the current system
  2. improving access to care
  3. providing social protection against care costs
  4. seeing the capped cost model as a flexible approach to reform
  5. exploring a range of options for raising revenue.
  • The immediate priority for government should be to stabilise and improve the current system, including by boosting staff pay and improving access to publicly funded care. Stabilising the current system by addressing demand pressures and increasing staff pay in line with the NHS would cost £1.0bn in 2020/21, £2.1bn in 2021/22 rising to £4.4bn in 2023/24, compared to the current baseline budget. Restoring access to 2010/11 levels of service would require around £8.1bn extra investment by 2023/24 on top of this (£12.5bn in total). (See our latest analysis on the 2019 Spending Round).
  • But more fundamental reform is needed to make the funding system fairer and provide government protection against social care costs. Policymakers have choices on how to do this, based on their values, priorities, and public spending plans. A Dilnot-style capped cost model – which focuses extra state funding on those with the greatest lifetime care needs and protects them from bearing costs above a certain limit – could be used flexibly by any government, and already lies on the statute book.
  • Any credible reform option requires government investment. The cost of implementing a Scottish-style model, for example, could add around £4.4bn to spending in 2019/20, rising to £5bn by 2023/24. The cost of a Dilnot-style model depends on where you set the cap to protect people against steep costs of care – ranging from around £1.7bn in 2019/20 rising to £2.1bn by 2023/24 under the least generous scenario (a £78,000 cap), to £6.8bn in 2019/20 rising to £7.8bn in 2023/24 under the most generous (a £0 cap – where the state covers all costs of a person's eligible needs).
  • Public spending on services in England was £525bn in 2017/18 (in today's prices). Of that, we spent around £18bn on adult social care. Across the whole UK, we spent £22.7bn on social care – 1.1% of GDP. This compared to wider health spending in 2017/18 of £153bn, equivalent to 7.1% of GDP – and projected to grow to 7.9% in 2023/24.
  • A range of options can be explored to raise revenue to fund social care reform. Increases in taxation would be an obvious route to paying for a more generous and fairer system.
  • Reform is often thought to be unaffordable. But if it chooses, the government can afford to provide more generous care, support and security for vulnerable people in society. If it doesn't, it will be choosing to prolong one of the biggest public policy and political failures of our generation.


In his first speech as prime minister, Boris Johnson promised to 'fix the crisis in social care once and for all', giving 'every older person the dignity and security they deserve'. This promise is welcome: the social care system in England is inadequate, unfair, and unsustainable.

Publicly funded social care – help, care and support that adults of all ages may need as a result of disability, illness, or other life circumstances – is only available to those with the highest needs and lowest means. And cuts in funding mean the number of older people receiving publicly funded care fell by 400,000 between 2009/10 and 2015/16. Many more go without the care they need. Around one in 10 people aged 65 face future lifetime care costs of over £100,000.

But the promise to fix social care isn't new. As prime minister back in 1997, Tony Blair told us that he didn't want our children to grow up 'in a country where the only way pensioners can get long-term care is by selling their home'. Since then, there have been a series of government reports on social care reform – two green papers, four white papers, and various consultations – as well as five independent commissions. Yet there has been no solution to the growing crisis.

Fixing social care isn't simple. Most people are not aware of what social care is, let alone how it works (many think it's 'free' like the NHS) making it hard to sell reform as better than what we've already got. Policy ideas are quickly politicised – labelled a 'death tax' or 'dementia tax'. And the right 'fix' depends on your interpretation of the system's problems. Are you worried about quality of services in the existing care system? The lack of fairness in who gets what? The absence of social protection against catastrophic care costs? The wellbeing of millions of unpaid carers – usually women? All the above? More?

Ultimately, social care reform is political: decisions will be driven by government priorities and values – for example, about the balance of responsibility between individuals and the state. But they should also be underpinned by evidence and analysis.

In this article, we outline five messages for government about priorities and options for social care reform in England, based on our assessment of the evidence and analysis of the costs of reform. In considering these options, we focus primarily on changes to the funding system. Other policy questions, such as the relationship between the NHS and social care, are not covered.

So: if the government wants to fix social care, what should it do?


1. Stabilise and sustain the current system

There's no doubt that the system needs fundamental reform. But the immediate priority for government should be to keep existing adult social care services running – providing a basic safety net for both older and working-age adults with high care needs and low financial means.

This safety net is failing. Government spending on adult social care in England fell from an average of £346 per person in 2010/11 to £324 in 2017/18. This is far less than spending on publicly funded social care in Scotland (£446) and Wales (£424).

Spending reductions have affected access, with fewer people eligible for care. But they have also affected the ability of social care providers to deliver high quality services. The amount local authorities are able to pay towards somebody's care in care homes is less than it costs to provide it. Many social care providers are handing back their contracts. And some are going bankrupt. Four in five social care services are judged 'good' or 'outstanding' by the Care Quality Commission (CQC), but there are variations in quality between regions and type of services. The quality of care in nursing homes is the biggest concern for the CQC.

Meanwhile, social care faces a staffing crisis. There are an estimated 110,000 vacancies in adult social care – around one in ten social workers and one in eleven care workers. Although increases to the national minimum wage have raised the wages of the lowest paid, the proportion of staff being paid the minimum has increased from 10% in 2016 to around 30% now. Many staff are on zero-hours contracts, and turnover is high. Workforce issues affect both care quality and the people delivering it. The vast majority of staff in adult social care are women, which means that the impact of poor working conditions exacerbates gender inequalities. Nurses working in the NHS earn 7% more than those working in adult social care – a gap set to grow under the new NHS pay deal. This can create problems for recruitment in social care and is unlikely to be sustainable long-term.

And then there's the need to meet increasing demand for care. By 2023/24, social care spending would need to grow at around 3.6% a year just to keep pace with the needs of an ageing population and growing numbers of younger disabled adults.

Simply meeting these demand pressures, under current eligibility, would require £2.7bn additional government investment by 2023/24 (compared to the current estimated adult social care budget). Boosting staff pay (to grow at the same rate as the NHS) to improve recruitment and retention, and hopefully quality, means this figure would grow to £4.4bn in 2023/24, with £1.0bn needed in 2020/21 and £2.1bn in 2021/22. These amounts are compared to the current estimated adult social care baseline budget, which does not include winter pressures monies and the social care grant, neither of which are committed beyond 2020/21.

The 2019 Spending Round provided roughly the £1.0bn needed for 2020/21, but this includes £500m that councils would need to raise through council tax increases.

Increasing fees paid to social care providers for their services (in addition to any price increases to boost staff pay) would be another route to stabilising the sector – and may be needed in some areas to cope with any increases in access. Extra funding may also be needed in the short-term to support social care providers at risk of failure.

Stabilising the current system also depends on wider public policy decisions. Future immigration policy, for example, must be designed to support the recruitment of international workers that the social care system relies on. More than 90% of care workers earn below the proposed £30,000 salary threshold that could be required to obtain a visa after Brexit, and 8% of staff are of EU nationality and 10% of non-EU nationality.


2. Improve access to care

The policy measures mentioned  in the previous section would help maintain a system in which only people with the highest needs and lowest means receive publicly funded services. Many more people need care and support. Some pay for it themselves or get help from friends and family. But large numbers go without. Unmet need for social care is likely to be high.

Between 2009/10 and 2014/15, the number of older people receiving publicly funded care fell by around 400,000, with likely year-on-year reductions in subsequent years too. At the same time, the number of people needing social care has been increasing. More broadly, Age UK estimates that 1.4 million older people have difficulties with activities of daily living – like eating, washing, dressing, and going to the toilet – but do not get all the support they need. We know very little about levels of unmet need for social care among working age adults.

The government must act to improve access to care. One way to do this would be restoring levels of eligibility for publicly funded care to those that existed in 2010, before cuts to services. This could be done through changes to eligibility criteria – for example, by reducing the national minimum threshold and providing additional funding for local areas to meet it. We estimate that doing so would require an increase of £6.0bn to 2019/20 budgets growing to an extra £12.5bn by 2023/24. (This is £8.1bn on top of the £4.4bn required just to stabilise the system.) Both estimates include meeting demand pressures, and the projection to 2023/24 includes the cost of increasing staff pay in line with the NHS.

Improving access and stabilising the provider market need to go hand in hand. One reason for this is that self-funders typically pay more for their care than those receiving publicly funded services, cross-subsidising the lower fees paid by local authorities. This means that any major increase in the proportion of publicly funded clients, without increases in the fees paid for them, could result in loss of income for providers. This could further destabilise the market, increase costs for self-funders, or both.


3. Provide social protection against care costs

Improving and expanding services is important. But this will not address the fundamental shortcomings of the system that most previous reform attempts have grappled with. People face great uncertainty about the future care costs they may incur, often having to sell their homes to pay for care or spend all of their savings. This uncertainty means that people are unable to plan ahead, creating fear and worry about the future. Without addressing this issue, Boris Johnson's promise made on the steps of Downing Street will not be delivered.

'My job is to protect you and your parents or grandparents from the fear of having to sell your home to pay for the costs of care.'
(Boris Johnson, 24 July 2019)

What creates this fear and worry? Under today's system, people with assets above £23,250 must pay for their own social care. But these costs are highly uncertain, and not spread evenly across the population (see the chart below). Some people may need no care at all, while others may need support in a care home for a decade, costing hundreds of thousands of pounds. The means test is more likely to affect older people in the care system, who have had longer to build up their assets.

In other parts of the economy, people are protected against these kinds of risks by forms of insurance – some provided by the private sector, some provided by the state. For example, the NHS provides insurance against the costs of health care, paid for primarily through taxes.

But for social care, there is no way for individuals to insure themselves. The private sector does not provide comprehensive insurance for social care costs. The same is true in other countries. This is for several reasons, including the challenges associated with predicting future care costs. Even if they did, these products wouldn't work without making them compulsory, given the market failures associated with voluntary insurance, such as 'adverse selection' (where sicker than average people sign up, increasing premiums). A compulsory insurance scheme would require major government involvement, including in underwriting costs, and may not help the current generation of older people. A voluntary scheme would also be unlikely to protect younger adults, who account for around 50% of the cost of the current care system.

This tells us that the government must play a fundamental role in providing people with protection against social care costs, pooling risks across the population in a way the current system doesn't. But it leaves choices for policymakers about the kind of protection offered.

Find out more by clicking below.

What kind of protection could be offered?

One approach would be to fund social care in the same way as the NHS, so that when individuals need care, the costs are covered by the state. We'll call this universal and comprehensive social care.

Under a universal and comprehensive model, the government must decide what kinds of care costs are included in the state's 'offer' – for example, the level of social care needs covered and the types of support that can be funded.

If the state was to cover spending on adult social care for all people currently receiving services in England (assuming all current self-funders would be eligible for publicly funded care under the new model and that policy was fully operational in 2019/20), this would equate to somewhere between £6bn and just over £11bn total public spending in 2019/20. (This figure doesn't include the costs of meeting additional unmet need that would likely present as a result of the policy change.) The lower estimate involves applying average state spending to those currently in the private system (so is therefore almost certainly an underestimate, given that self-funders pay higher rates to cross-subsidise low fees paid by local authorities), while the upper estimate includes the gross costs of those currently in the private system.

The principal objection to funding social care like the NHS, however, tends to be its cost to the state. Reducing these costs means reducing the state's 'offer' and individuals paying for some portion of their care.

For example, in Scotland, people over 65 (recently extended to all working-age adults) receive free personal care – including personal hygiene, continence, diet, mobility, counselling, simple treatments, and personal assistance services – in their own homes, if their needs are deemed eligible after a local authority assessment. The government contributes up to £177 a week to personal care costs for people living in care homes, with a further payment of £80 to people who need nursing care. But individuals cover additional costs, including contributing towards living costs if they need residential care (subject to a means test) – just as they would if they were living in their own home. Free personal care allowances in Scotland only meet around 25% of the weekly costs of a residential care home.

Introducing the Scottish system in England would require around £4.4bn additional government investment by 2019/20, growing to £5bn by 2023/24. This assumes a small increase in demand associated with introducing the policy, as observed in Scotland.

Under this kind of approach, the government is choosing to provide some level of social care support for all. For many, this would be an improvement on the current system. It would mean a more equal system for the care needs covered and clarity on the state's 'offer' to the population. But individuals with persistent and severe care needs – for example, a person with dementia, needing high intensity care for a decade – would still face high costs. And all individuals would still face uncertainty about what the future holds.

Illustrative chart depicting the future life time care costs by individual contribution against a basic state contribution for everyone

An alternative approach is to start by protecting individuals against catastrophic care costs – insuring against risks at the other end of the spectrum to the Scottish model (see the chart below).

This approach is based on economic theory suggesting that – unless the full costs of care are to be insured – it is preferable to cover all costs of an uncertain risk (ie the risk of needing social care) above a certain limit, rather than meeting some portion of costs with no limit.

The capped cost model, proposed by the Dilnot Commission in 2011, is designed to target additional state funding on individuals with the greatest and most expensive care needs over their lifetime.

Under this model, people with sufficient means pay their own care costs up to a cap. After that, the state pays (though, again, some costs – like accommodation in a care home – are not all included, and the cap only applies to eligible needs). Dilnot's recommendation was for a £35,000 lifetime cap in 2010/12, which would have risen to around £46,000 today. The government's proposed cap, before they ditched the idea, was £72,000 in 2016. Either option would eliminate catastrophic costs associated with care needs and target additional resources towards those with most expensive lifetime needs. But both options would still leave many people with eligible needs responsible for their own care costs until they hit the cap. These options also fall short of providing equal care for equal needs (as a universal model would).

Illustrative chart depicting how an individual might contribute towards their future lifetime care costs but with catastrophic costs

Dilnot's proposals also included a sizeable individual contribution to general living costs for people in residential care, at £10,000 a year (equivalent to around £13,000 today were they uprated as Dilnot proposed). This means that, in practice, individuals with long-term care needs may still need to draw on capital after hitting the cap. Reducing this contribution – for example, bringing it closer to the new full state pension (at just over £8,700 a year) – could make this model more generous and shift the balance of costs further towards the state.


4. See the capped cost model as a flexible approach to reform

The options discussed on the previous section offer policymakers choices about the kind of social protection that could be offered – from all care needs to just a portion of them. An important advantage of the capped cost model, however, is that it already lies on the statute book.

The principle of a cap on care costs was included in the 2014 Care Act. This means that social care leaders have a head start in implementing the policy – a part of the policymaking process that too often gets forgotten – and would be able to do so without new legislation.

The capped cost model could also be used flexibly by governments of different ideological hues, based on choices about the balance of responsibility between individuals and the state – as well as more pragmatic decisions about how much cash they are willing to spend.

One approach would be to implement the cap (combined with a more generous means test) at the middle of the range proposed by Dilnot in 2011 (equivalent to £46,000 in today's prices), or at the level proposed by the government in response (equivalent to £78,000 in today's prices). Improvements to the existing social care system and increases in access, would increase the number of people helped by these models and make them more expensive.

An alternative would be to set the cap at £0 – providing everybody with publicly funded care if their needs met nationally agreed eligibility criteria (equivalent to a version of universal and comprehensive social care). This would shift the balance of responsibility from individuals to the state. And it would remove the need for people with eligible needs (and enough means) to pay for their care until they've spent some of their own money. But it would also be more expensive. Another option would be to include a smaller cap than Dilnot suggested – say £10,000 or £25,000.

The table below estimates the cost of implementing a Dilnot-style means test and capped cost model under four scenarios by 2019/20 and 2023/24. These estimates are based on increasing the upper capital threshold from £23,250 to £125,000 (so, a more generous means test) and an individual contribution of up to £13,000 to annual living costs. They set out additional costs compared to the current system if the models were fully implemented in the current year – so do not include extra costs needed to improve quality and access. In practice, there would be a lag in implementing a capped cost or any new funding model, and under a capped cost model, the costs would be low until significant numbers of individuals reach the cap.

As we outlined above, the government could make a capped cost model more generous by reducing individuals' contribution to annual living costs to the level of the full state pension. This would, inevitably, increase costs to the state – for example, adding around £0.7bn in 2019/20 under a £46,000 cap. None of the estimates include any demand increase that could be stimulated by introducing the cap (which are probably more likely the lower the cap).

Additional costs of a Dilnot-style means test and capped cost model

2019/20 and 2023/24


Cap level                     




















These estimates illustrate how a capped cost model could be used flexibly by government depending on their values, priorities, and public spending plans. Political values, of course, should hopefully be shaped by public values. In our deliberative research with the public, most people favoured the idea of the state having most responsibility for funding social care. Survey data provide a more mixed picture, with most people (55%) in the 2017 British Social Attitudes survey favouring some sort of shared responsibility for social care costs, and 41% favouring a government funded system paid for by taxes – though survey data on social care should be treated with caution, given many people are unaware of how the system works.

Capping annual or lifetime costs?

Other reform proposals – for example, a Scottish-style model – could also be combined with a cap on care costs to protect people against unaffordable bills. For example, Policy Exchange recently proposed a variant on the Scottish social care model, with a £5,000 annual cap on individual costs (in the form of a co-payment).

If a capped cost model is to be explored by government, a lifetime – not annual – cap would likely do a better job of targeting state funding on those with highest needs. Consider person A with care needs of £4,500 a year for 10 years. Under an annual cap of £5,000, they would receive no additional state support over the decade, footing the entire bill of £45,000. But under a lifetime cap of – say – £25,000, additional state funding would kick in within just over half the time. Compare that to person B, with care costs of £6,000 a year for two years costing a total of £2,000. Under the £5,000 annual cap, the state would contribute to person B's costs in both years, while under a lifetime cap of £25,000, funding would be prioritised for person A, who has much greater lifetime needs, instead.


5. Explore a range of options for raising revenue

Whichever reform option the government chooses, additional public spending will be needed to fund it.

After a decade of austerity in public finances, cutting other public services to pay for social care is not feasible or desirable – particularly given the potential funds needed. Additional borrowing could play a part. But increases in tax revenue are likely to be needed.

Policy options often explored to raise additional revenue to fund social care include:

  • increasing general taxation (including from the wealthy)
  • taxing or redirecting spend on older people (eg by extending National Insurance (NI) contributions beyond retirement age)
  • taxing wealth (eg changes to council tax or new forms of tax on people's estates)
  • a hypothecated tax for social care.

Each option has advantages and drawbacks. While the new prime minister has promised tax cuts for higher earners and the current Chancellor has described himself as a 'low tax guy', tax increases would be an obvious route to fund a more generous system. Overall tax revenue in the UK as a share of national income is lower than most other OECD and EU15 countries. And people at the median and top of the UK income distribution pay less income tax than they would if they lived in many other European countries. Taxing wealth would be among the most progressive options, but is not easy politically, as Andy Burnham, a previous health secretary, learnt the hard way when he was accused of proposing a 'death tax'. And while taxing or redirecting spending on older people could help promote intergenerational fairness, this alone would be unlikely to cover the full costs of reform.

Some form of new hypothecated tax – a tax for social care – might generate support from the public. But the idea is a dud in policy terms, as the tax take would rise and fall with the economy, not changes in social care need or demand. A 'soft' kind of hypothecation, where the money raised is not actually ring-fenced in practice – like Labour's 2002 promise to increase NI to fund the NHS – could be used to sell tax increases via other options instead. Though politicians should also consider the risk that an unpopular tax linked directly to social care reform could – as we've seen in the past – help sink the policy proposal it aimed to fund.

Ultimately, the right approach depends on policy choices and trade-offs – for example, on the relative contribution of working age and older people towards paying for care. It also depends on the type of reform pursued. Given that any reform option is likely to be more regressive than the current system (where limited resources are targeted towards people with the lowest means), more progressive options could be explored for raising revenue to pay for it.


Where next?

The need to fix the crisis in social care has been well articulated for decades. Yet politicians continue to duck reform, and people continue to suffer unnecessarily.

The immediate priority for government should be to stabilise the current system and increase the number of people receiving publicly funded care. This will require additional government investment.

But more fundamental reform is needed to make the system fairer and provide people with protection against social care costs. Policymakers have choices on how to do this – and a Dilnot-style capped cost model could be used flexibly by government, depending on political values and priorities. Increases in taxation are likely to be needed to fund the costs of reform.

Other policy changes, like the closer integration of NHS and social care services, are an important part of making the health and care system work better for patients, but they are no replacement for the changes needed in the way social care funding and entitlements work.

Worries about increased government spending to pay for reform have been a barrier to change in the past – and things are likely to be no different this time around. Reform certainly requires investment: the cost of implementing a Scottish-style model, for example, could add around £4.4bn to spending in 2019/20, rising to £5bn by 2023/24. The cost of a Dilnot-style model depends on where you set the cap – ranging from around £1.7bn in 2019/20 rising to £2.1bn by 2023/24 under the least generous scenario (a £78,000 cap), to £6.8bn in 2019/20 rising to £7.8bn in 2023/24 under the most generous (a £0 cap). These are not small sums.

In the context of England's wealth and broader public spending, however, social care reform is not so unaffordable. Public spending on services in England was £525bn in 2017/18 (in today's prices). Of that, we spent around £18bn on adult social care. Across the whole UK, we spent £22.7bn on social care – 1.1% of GDP. This compared to wider health spending in 2017/18 of £153bn, equivalent to 7.1% of GDP – and projected to grow to 7.9% in 2023/24.

Simply meeting the government's basic promises – first the prime minister's pledge to provide security for people in their old age, and second the government's 2018 commitment that social care will not impose additional pressure on the NHS – requires investment and reform. The bare minimum for even the most unambitious government, therefore, would be to keep existing services running (£4.4bn by 2023/24) and implement some kind of credible reform to the funding system to provide security for people against care costs – for example, a Dilnot-style model with a £46,000 cap. This would add another £3.1bn investment by 2023/24 (so £7.5bn overall). But – as we've outlined – government must also act to improve access to publicly funded care.

In the end, the appropriate level of spending on adult social care comes down to political choices – for example, about the right amount and balance of spending between public services, and how to raise the revenue to pay for them. If it chooses to, the government can afford to provide fairer and more generous care and support for vulnerable people in society. If it doesn't, the government will be choosing to prolong one of the biggest public policy failures of our generation.

About the authors


  • All figures listed in this article are in real terms at 2019/20 prices.

Downloads and related content

Further reading

Kjell-bubble-diagramArtboard 101 copy

Get social

Follow us on Twitter
Kjell-bubble-diagramArtboard 101

Work with us

We look for talented and passionate individuals as everyone at the Health Foundation has an important role to play.

View current vacancies
Artboard 101 copy 2

The Q community

Q is an initiative connecting people with improvement expertise across the UK.

Find out more