Section 2 – Options for reform, their costs and considerations

All our policy, economic and public engagement work on this project has linked to exploring five approaches to funding social care for older people in England, which are discussed more fully in our interim report, Approaches to social care funding. They were chosen to reflect solutions commonly raised in the debate around social care funding:

  • improving the current system
  • introducing free personal care
  • introducing a cap on costs and a revised ‘floor’ to the means test
  • introducing a hypothecated tax for social care
  • introducing a single budget for health and social care.

We are clear that these options are neither directly equivalent nor mutually exclusive, and that they tackle very different aspects of the social care funding challenge. They are also not an exhaustive list of the possible models that policymakers could consider.

So in this overview report, we review them by way of developing the discussion of some of the wider issues each raises, such as:

  • the costs of improvement (or inaction) in the current system
  • the nature of the ‘offer’ itself and issues of expanding access more widely
  • social protection and fairness
  • how funding could be raised for specific options.

While a joint health and social care budget might support progress towards more integrated care, it will not alone generate additional revenue for either health or social care, nor change eligibility for care. Therefore, we do not explore this option in detail in the costing analysis but do consider the issue of further integration between health and social care in the context of reform and public attitudes (see Approaches to social care funding for more on this option).

These options focus on changes to public provision of social care for older people. This reflects, among other factors, the reality that in general, working-age people who use social care services have very different financial characteristics to users aged 65 and older. For instance, there is far greater scope for funding older people’s social care by drawing on personal and property wealth than there is for funding working-age adults’ social care, where a fully tax-funded solution is likely to be the only appropriate approach.

This does not mean that services for working-age adults do not require attention, and indeed additional funding. In fact, the opposite is clearly true. However, as described in our interim report, we see no alternative to continuing a largely tax-funded option for providing care for younger adults, so have not explored alternatives. It remains vital that the social care system works for all adults with social care needs, and so our projection of the associated demand and costs for the future include spending on all adults, which changes when alternative approaches are applied to services for those aged 65 and older.

The pros and cons of the models are adapted from Approaches to social care funding.

All the financial figures in this section are drawn from the paper, Social care funding options: How much will they cost? and are based on data from the Personal Social Services Research Unit (PSSRU). The PSSRU projects that on current trends, social care spending pressures will rise at an average of 3.7% a year until 2030/31, compared with our projections that local authority spending would rise by just 2.1%. Unless stated otherwise, all figures in this section are expressed in 2018/19 prices, relate to England only and are rounded to whole billions of pounds where uncertainty makes a higher degree of accuracy spurious.

‘Improving’ the current system – and maintaining or restoring access

This option, as defined in Approaches to social care funding, would mean retaining the existing social care system but seeking incremental improvements over time.

There are several areas in which the government could make incremental improvements to the current system without changing legislation or large-scale reform. These could include financial support short of a new tax, perhaps directly increasing the amount of funding available through the improved Better Care Fund or the adult social care support grant, by allowing councils the freedom to raise additional income, or incentivising private funding. Other improvements might include expanding the eligibility for publicly funded care, improving integration with other services through sustainability and transformation partnerships and integrated care systems, and spreading innovation and best practice. The nature of the offer and who receives support could be addressed by revising either the means or needs tests. Options for improving the current system are discussed further in the interim report Approaches to social care funding.

We have not attempted to model a version of incremental change, as described in the interim report. Instead we have modelled two options at different ends of a scale to set out the funding that would be needed, from no change to the current system and keeping pace with pressures, to more widespread improvements to the system by returning to levels observed in 2009/10. By doing so we provide a range of costs that may occur for different levels of improvement.


This approach would recognise the great difficulty successive governments have faced in achieving major reform. Compared with implementing a new model, it would also involve minimal disruption and minimal transition costs – though the existing funding gap and growing pressures will still need to be met. Making small improvements would not address many of the fundamental problems with the current system, including the downward trend in the number of people receiving care. Even substantial levels of new money would not fix the issues of the system’s complexity. Nor would it protect people against catastrophic care costs.

  • Through its eligibility criteria, the current system aims to focus public resources for social care on those with the greatest needs and the lowest means. Yet the number of people receiving publicly funded care is declining, and there is evidence that the number of people with unmet care needs is growing. In addition, due to financial pressures on care providers, the rate paid by those who self-fund their care is likely to be higher than it is for those funded by their local council.
  • Arguably, the current system has proved resilient to both growing and shrinking budgets due to the range of ‘levers’ (such as the means test thresholds), which enable local or national government to adjust the level of demand for funding.
  • However, without reform, and even with significant investment, issues will remain over the system’s complexity, equity and failure to protect individuals against catastrophic costs without reform, even with significant investment. There is also a question as to whether recent (often short-term) efforts to address the shortfall in funding have been sufficient. In the case of the social care precept, for example, some councils’ reluctance to apply this, and the variation in the level raised between local authorities, means that many consider it unsustainable as a funding source. Councils’ other options for raising revenue locally are limited and many of the possible mechanisms, such as council tax increases, run the risk of exacerbating local variation.
  • Issues around sustainability of funding also affect the stability of the social care provider market: this has been evident in councils reporting the handing back of contracts by both home and residential care providers.

To explore some of the costs inherent in any attempts to improve access to the system while not seeking to wholeheartedly reform it, two scenarios were modelled in the paper Social care funding options: How much will they cost?

  • Maintaining the system at 2015/16 levels and keeping pace with pressures would require additional investment of £4bn more by 2020/21 than was spent in 2015/16. This is £1.5bn higher than our projection of additional spending by local authorities based on current trends. By 2030/31, an extra £12bn would need to be spent, £6bn higher than projected spending plans.
  • Restoring the system to 2009/10 levels and restoring the level of eligibility that existed at that time (perhaps through changes in the eligibility criteria) would require an additional £8bn in 2020/21 above estimated plans. Projecting forward to 2030/31, the funding gap grows to £15bn.

These two modelled levels of funding, without change to the current system, provide estimates for a potential funding gap of between £1.5bn and £8bn in 2020/21 and between £6bn and £15bn in 2030/31, depending on various factors, but crucially, the level of quality and access desired.

The lower cost option, maintaining funding at 2015/16 levels, is simply in line with PSSRU projections of current provision. As such, it is not enough to lead to real improvement. This just stops the system declining any further, but does not address issues such as fewer people receiving care or market instability (although some level of improvement is possible through efficiency savings).

On the other hand, restoring funding to 2009/10 levels represents an upper estimate of the investment needed to restore services to the levels of access previously observed. This level of investment could indeed improve access and quality without primary legislation, but the major concerns over the design of the system even at that time, would go unresolved if this option was taken.

It is true that genuine efficiency gains will have been made since 2009/10, so the actual cost could be lower if these have been sustained.

Free personal care – broadening the offer

Free personal care has been offered in Scotland since 2002. Under this model in Scotland, personal care is provided to anyone aged over 65 based solely on need and not their ability to pay. Personal care includes personal hygiene, continence, diet, mobility, counselling, simple treatments and personal assistance. Those at critical or substantial risk to their independence or health and wellbeing are expected to be provided with social care services within a short period of time.

The approach is different depending on whether the individual receives personal care in their own home (domiciliary care) or in a care home:

  • People receiving domiciliary care are not charged for any personal care services. The package offered varies on a case-by-case basis. According to free personal care data, net spending on domiciliary care services for older people in 2015/16 was £196 per week per user in 2018/19 terms: of this £161 was spent on free personal care.
  • For people who receive care in a residential home, the local authority contributes to the cost of their personal care (at a flat rate) directly to the care provider. As of April 2015, this contribution is £171 for personal care, plus an additional £78 per week for nursing care services if needed. This payment does not cover accommodation costs, which are subject to a means test. This is currently mandated by the Scottish government, so cannot vary across different local authorities.

This system relates to the nature and scope of the government’s offer on entitlement for social care, but is not a means of generating funding.


Free personal care would mean extending the government’s ‘offer’ on social care to more people. By aligning eligibility with health, free personal care would also remove one of the biggest national obstacles to integration of health and social care. However, given that this would require an increase in public spending, there is a question – as with all models – as to whether this would be the best use of additional funding for social care.

At current levels of eligibility, free personal care could require around an extra £6bn in 2020/21 and £8bn by 2030/31, compared with continuing levels of access and quality under the current system. This would increase the estimated funding gap to £7bn in 2020/21 and £14bn in 2030/31.

However, this funding would be used only to expand the number of people who could access publicly funded personal care by removing the means test from these services. It would not allow for any improvement in the needs threshold to expand access beyond those with severe need to include more people with high or moderate need, as would be possible in the scenario of returning to 2009/10 levels. If improving the needs threshold were also to be included, this would add considerably more to the total cost.

It is likely that, initially at least, the costs could be higher due to behavioural effects. In Scotland, introducing free personal care created unexpected levels of increased demand for domiciliary care which we might also expect to occur in England.

Table 2: Estimated number of people aged 65 and older receiving fully publicly funded care in 2015/16

Current system

Free personal care

Domiciliary care



Residential care






Note that the table uses ‘snapshot’ (on-a-given-day) data and numbers are accordingly lower than some ‘in-year’ figures reported elsewhere.

Indeed, local authorities in Scotland experienced a loss of income and an increase in costs when free personal care was introduced, and the system has become increasingly expensive over time. However, by supporting older people to live at home, helping to prevent costly hospital admissions, and delaying the need for residential care, the system may have resulted in lower total government expenditure as compared with no policy being in place.

The model has proved popular and durable in Scotland and is now being expanded to adults of working age. The number of people in receipt of personal care grew significantly once the policy was introduced, suggesting that the system is providing care for people whose needs were previously unmet.

The system also supports the longer-term vision for social care (and health) more broadly, by supporting older people to stay in their own homes. However, introducing free personal care also appears to have reduced the provision of care services which do not meet the narrow definition of personal care in fixed budgets, and charges have increased.

The ‘cap and floor’ option – protecting people from catastrophic costs

In the run-up to the 2017 general election, the Conservative Party proposed the following (in its election manifesto and through subsequent clarifications):

  • Two changes to the means test for social care:
    • replacing the existing upper and lower thresholds with a single threshold, set at £100,000, much higher than the current upper threshold (of £23,250), thereby abolishing tariff income and ensuring that users always retained at least £100,000 in assets
    • including property assets in the means test for both residential and domiciliary care, rather than for residential care only, as is the case currently.
  • A cap on the lifetime costs of care, meaning that no one would need to pay above a certain level. The level of the cap was not specified.


The inclusion of a cap on care costs and the proposed changes to the means test are likely to mean a more generous system for some, offering protection against catastrophic care costs. This option constitutes a policy that has previously received some support from across the political spectrum. However, there is a question as to whether this alone is the best use of increased spending on social care, given the complex pattern of ‘winners’ and ‘losers’ (some of whom will make big gains).

In principle, a cap on care costs would protect people from very high costs of care. However, the extent of this protection (and naturally, the cost to the Treasury) would depend entirely on where the cap was set. Even with the introduction of a cap and a floor, many people would still be liable for relatively high costs – including all care which falls outside of needs eligibility.

There is a risk that including property in the means test for domiciliary care would reduce the incentive for people to remain in their homes (although it is difficult to predict how behaviours would change in practice). This may be seen by many as unhelpful, given that current health and care policy is aimed at supporting people to live independently, and avoiding the need for long-term care as far as possible.

Implementation of this system would be aided by the fact that some of the principles set out by the Dilnot Commission, such as a cap on costs, are already provided for through the 2014 Care Act. However, communicating this system – which has added complexity – to the public is likely to be difficult, given the limited understanding of the current system. It will, in practice, also be a very different system, with vast numbers of deferred payments perhaps becoming the norm.

Our analysis in Social care funding options: How much will they cost? shows that the ‘cap and floor’ model, with an assumed cap of £75,000, would cost an additional £4bn in 2020/21 and an extra £6bn in 2030/31 above the costs of maintaining the current model. Introducing this model could therefore increase the total projected funding gap against our estimated budget from £1.5bn to £5bn in 2020/21, and from £6bn to £12bn in 2030/31.

Our modelling shows that such a new system would make a saving on the public funding of domiciliary care (ie be less generous to users) because it includes individuals’ housing assets in the means test. However, it would cost considerably more (ie be more generous to users) for people needing residential care, as can be seen in Table 3. 

Domiciliary care under this proposal will cost around £270m less than it does currently but costs for publicly funding residential care would double, from £3.4bn to £6.8bn. Note that, in our modelling, the costs of assessment are held constant, although system change will likely incur some implementation costs.

Table 3: Estimated number of people aged 65 and older receiving fully publicly funded care in 2015/16

Current system

Cap and floor

Domiciliary care



Residential care






Note that the table uses ‘snapshot’ (on-a-given-day) data and numbers are accordingly lower than some ‘in-year’ figures reported elsewhere.

As this offer is more generous to potential residential care users, there is a risk that it could create additional demand for residential care versus domiciliary care, running contrary to the long-term strategic direction of most local authorities.

Summary of costs

The costs of all options are summarised in Table 4.

Table 4: Summary of costs for the cap and floor model and free personal care alongside additional costs for maintaining and improving access and/or quality of current system

Current system


Maintaining at 2015/16 levels

Restoring to 2009/10 levels

Cap and Floor

Free personal care


Projected cost pressures





Increase from 2015/16 spend of £17.1bn





Additional cost above maintaining 2015/16





Projected funding available


Extra funding required






Projected cost pressures





Increase from 2015/16 spend of £17.1bn





Additional cost above maintaining 2015/16





Projected funding available


Extra funding required





Figure 1: Costs for the cap and floor model and free personal care alongside additional costs for maintaining and improving access to and/or quality of current system

How to raise the funding needed

In the previous section, we showed that any option that avoids further decline in access and quality will require additional investment, which in turn will require additional taxation or change to current benefits. By 2030/31, even maintaining the current system would require an additional £6bn above estimated spending. Improving access to similar levels achieved previously under the current system (ie 2009/10 levels) would be the most expensive option, requiring an extra £15bn by 2030/31. Improving access by introducing free personal care would need an extra £14bn, while protecting against catastrophic costs would need an extra £12bn.

How these sums are funded depends on several trade-offs regarding the amount required and how it is to be raised, including:

  • the balance between protecting those with the lowest personal resources and protecting people from catastrophic costs
  • the relative contributions of working-age and older people towards paying for care
  • whether funding is raised against people’s income or their wealth.

The analysis in the paper Social care funding options: How much and where from? sets these out in detail (summarised here), including the amount that can be raised through current taxation (Table 4).

At least some of the additional funding required could theoretically be found from other areas of public spending or additional borrowing. But with other services already facing further cuts, and the government's commitment to reducing the national budget deficit, the scope for additional social care spending without the government raising additional tax revenue is very limited.

Increasing general taxation (including taxing the wealthy)

Adding 1p to the main rate, higher rate and employers’ contribution to NI by 2030/31 could raise enough to fund the introduction of the cap and floor model. If, at the same time, means testing was introduced for winter fuel payments, this could be enough to introduce free personal care.

Alternatively, adding 2p to the basic, higher and top levels of income tax, or 3p to VAT, would provide more than enough to fund either the cap and floor or free personal care, and would be sufficient to improve access and quality close to the levels observed in 2009/10.

A different approach favoured by some is to limit tax increases to those earning the highest incomes, by increasing only the top and higher rate of income tax. This approach would be progressive in nature but would require substantial increases in the rate. For example, just funding the current projected pressures would require a 3p increase in the higher rate of income tax, or 6p in the top rate in 2030/31.

Improving access back to levels observed in 2009/10 would require a 9p increase in the higher rate of income tax, or 17p in the top rate in 2030/31. Introducing the cap and floor model would require an extra 8p or 16p in the pound respectively, and for free personal care an extra 7p or 16p. It should also be noted that there is a significant risk that a rise of this magnitude would not in fact generate the expected level of income as the fnanical behaviour of the highest earners might well change in response to the planned tax increases.

This would also increase the burden placed on a very small portion of the population. Currently the top 1% of earners pay around 28% of total income tax, and households earning the top 10% of incomes pay around 27% of total taxes.

Taxing or redirecting spending on older people

These approaches raise concerns about intergenerational fairness – taxing (via NI) lower income families to provide free services to wealthy older people looks regressive. A common argument for how to raise funding for social care is that as older people are the greater benefactors of an improved care system, they should contribute towards funding any improvement to the model.

For example, linking the winter fuel payment to Pension Credit could make an additional £1.8bn available for social care services but is potentially a regressive option, given the high rate of eligible poorer pensioners who do not claim Pension Credit.

Table 5: Estimated revenue raised from a 1 percentage point increase to key tax rates (England's share of additional tax revenue)


2020/21 (£bn) England

2030/31 (£bn) England

Tax type



(2018/19 terms)

(2018/19 terms)

Income tax


IPPR model




IPPR model



Top rate

IPPR model



Fiscal drag (no uprating)

IPPR model



NI contributions

All rates (employee and employers)

IPPR model



Extend beyond retirement age

IPPR model




Main rate




Note: IPPR: Institute for Public Policy Research.

Another option is extending NI contributions beyond retirement age, given the large growth in the number of people working beyond retirement age. At current levels this could raise a further £1bn. However, care should be taken with this option; the Mirrlees Review of the tax system argued that we should be looking to strengthen work incentives at older ages, not weaken them, as older people are particularly responsive to incentives. Therefore, one option would be to introduce a lower rate of NI contributions for this group, for example the 6% rate recommended by the Barker Commission.

A move to a ‘double-lock’ pension was initially proposed in the Conservative Party manifesto but the amount released would also depend on the outlook for wages growth. In reality, it is rare for both average earnings and inflation to be below 2.5%, so this might do little to change the projected long-run generosity of the state pension.

This approach might close the funding gap under the current system but on its own would not provide enough additional funding for either of the alternative models of reform explored. It would therefore need to be part of a wider solution.

Taxing wealth

There is mounting interest in moving the balance of the UK tax burden away from income towards accumulated personal household wealth, the estimated scale of which, at £12tn, makes it an option which is impossible to ignore. Inheritance tax receipts have been rising in recent years (and were identified in the 2015 Spending Review as a potential source of funds for Dilnot reforms).

A reform along these lines was proposed by the then Health Secretary Andy Burnham in 2010: this ‘care duty’ could raise something in the order of £6.5bn. This would make a significant contribution towards funding improved care in the short term, but implications for long-term funding will depend on factors such as changes in the property market.

Other proposals include changes to council tax beyond the social care precept such as an additional council tax band. The Resolution Foundation proposed that a 1% tax on the value of properties above £100,000 could potentially raise £9bn in revenue. This would therefore cover the cost of each of the options modelled in 2020/21.

Wealth taxes such as these are generally seen as progressive options (or at worst proportional). Clearly, options exist that could fund at least part of an improved social care system. However, they can prove controversial, as many people’s homes have risen dramatically in value while their incomes have not. Equally, introducing the value of property into the means test for domiciliary care in the cap and floor proposal could also be argued to be a form of tax on property.

Introducing a hypothecated social care tax

A hypothecated tax has been often suggested in the context of health and social care. However, this would only raise funding if it were set at a level that would increase the total national tax revenue in the manner illustrated above for general tax rises. As such, it is a method for raising revenue rather than a new source of funding in itself. Ensuring that tax raised is used for a specific purpose helps clarify the link between taxation and government spending, and can increase a sense of accountability – although in practice, this would depend on the form of hypothecation (full, partial or ‘soft’):

  • full hypothecation, where spending is linked directly to revenue raised – all funding for social care would come from a clearly identified source and could be used only for social care (countries such as Germany fund social care in a similar way through social insurance schemes, but in England, which has no history of social insurance models, this is more likely to be a specific tax)
  • partial or incremental hypothecation, where revenue from tax increases is used to raise spending, but most funding for social care comes from other sources and is not hypothecated
  • soft hypothecation: a rise in tax is symbolically linked to a purpose, but in practice the take is not ring-fenced for that purpose (for example, the Labour government’s 2002 increase in NI to ‘pay’ for extra NHS spending).


The potential revenue raised would depend on the form of tax increase. In any case, it would be likely to require very substantial tax increases to bring about improvements (and to be worth the upheaval).

Implementation would be a major issue. A hypothecated tax for social care would be a major change from the current system of public finance in the UK. Many of the countries that have introduced social insurance have been able to build on established insurance schemes, often in health.

Public acceptability is often considered the primary strength of hypothecation, as raising public awareness of debates around available funding and trade-offs in service delivery may offer the best chance of building public support for raising additional funding. This is discussed further in Section 3 on public attitudes.

A key weakness of a hypothecated tax is that any ‘take’ will rise and fall with the economy, rather than being aligned to changes in need or demand. In practice, social insurance systems have often required a ‘stabilisation fund’, which is paid into when the economy is stronger, to provide additional support in a downturn. While this can smooth over the impact of economic boom and bust, it also weakens the short-term link between taxation and spending.

Introducing a hypothecated tax for social care alone would risk exacerbating the separation between the health and social care systems. Conversely, a social care and health hypothecated tax would be a huge undertaking which risks leaving social care as the poor neighbour.

This is based on current spending plans for adult social care to 2017/18, expected local authority spending power to 2019/20, and then assuming spending would rise in line with GDP beyond 2019/20 in line with Office for Budget Responsibility methodology. In reality it will depend heavily on the shift among local authorities towards greater retention of business rates, which we do not attempt to model here.

§ These updated projections for pressures are lower than previously reported (see:
and those used in our previous reports (see: This reflects a number of changes, including fewer people currently receiving services, lower estimates of population growth, and lower projected growth in national wealth (measured using GDP). See accompanying report Social care funding options: How much will they cost? for more information.

The Scottish government calls this ‘free personal and nursing care’.

** Excluding client contribution included in gross spending.

†† Under the current system, people contribute a percentage of the costs of their care if they have assets between the upper and lower thresholds.

‡‡ Figures do not sum due to rounding, specifically when comparing free personal care and cap and floor models. Figures are rounded for ease of reading and to reflect uncertainty in the estimate.

§§ The estimates provided here do not include any behavioural effects due to changes in taxes, so represent the upper bounds for the revenue that would be raised. For example, increases in income tax may discourage some people from working, reducing the overall revenue raised.

Previous Next