Although politicians may wish it wasn’t so, the state of the NHS and its need for more resources will be an unavoidable issue during this year’s general election campaign and beyond.
Politicians will claim that the NHS has “more money than ever before” – a mathematical fact that, given the almost relentless upward march of inflation and population growth, would be more remarkable if it were untrue than true. And yet across the country, patients, their friends, neighbours and relatives report record-level dissatisfaction with NHS services, driven largely by painfully long waits to receive care from NHS organisations that do not appear to have sufficient resources to cope with demand.
We know that overall, between 2013/14 and 2023/24, NHS current expenditure in England increased on average by 3.1% a year in real terms – a rate that is around half a percentage point slower than the long-term average. But how were these annual funding increases shared between different “front-line” NHS services – the hospitals, GP practices, mental health organisations, ambulance trusts and community health care services that deliver patient care? And how does the level of resources received by those different types of service compare to the level of need we find for them in the population as a whole?
Answers to those questions are not readily available.
In response to parliamentary questions last year, the Department of Health and Social Care (DHSC) released a breakdown of spending from NHS England’s commissioning budget, drawn from commissioner in-year performance reporting. While this purported to show how around 95% of the NHS budget for England had been spent in recent years, closer inspection begged more questions than answers. In particular, the 47% cash increase in funding reported in the figures for community health care between 2019/20 and 2022/23 did not marry up with experiences of people running and using such services, leading some to suspect a classification error.
A further problem with the DHSC data is the very significant increase over the last five years in the proportion of funding for NHS services that comes from outside the formal NHS England commissioning budget, and instead stems from central top ups, in the form of funding for Covid, elective recovery, mid-year emergency budget supplements and employer pension contribution rate increases – a reflection, in part, of the unstable and short-termist nature of recent NHS funding settlements. It is not clear how this funding has been factored into the figures reported to parliament. Nor is it clear what the 15% of commissioner expenditure described as “other” in that data consists of, nor what proportion of spending in each category has been in the form of funding for NHS-run services, and what has been spent on the private or voluntary sector.
Such definitional issues will always dog analyses of health care spending. In addition to health care itself being a practice that changes and evolves over time – with care that was once confined to acute settings gradually shifting to others, for example – accounting for income and expenditure in any field is an art not a science. Consistency is particularly a problem with data that has not been subject to audit.
New analysis
With those important caveats acknowledged, we here present our own estimates of NHS funding by broad health care type between 2016/17 and 2022/23 – the earliest and latest dates that published accounts allow. Our analysis here is drawn from the detailed annual accounts of all 212 NHS provider trusts, supplemented with information from NHS England’s and the DHSC’s annual accounts. For acute, mental health, community and ambulance service care, the data tells us what the NHS trusts providing this care record in their audited accounts as the financial resources going into these services, and we have deliberately excluded from the analysis NHS provider spending on outsourcing to the private or voluntary sectors. This means that the figures presented here give our best estimate of the level of resources that have gone into local, NHS-run services across England, by broad service type, and as such help add to our understanding as to why these services appear to be under so much pressure.
In this analysis we also combine data on service-level funding levels with separate data on the level of need for each service type in the population in England as a whole. These needs assessments are produced by NHS England and are based on detailed analyses of how health care need changes with age. While the needs assessments are used to help NHS England determine how total funding should be distributed on a geographic level – for example to integrated care boards – they can also be used to assess how well national funding on different types of NHS services reflects estimated levels of need for those specific forms of services, particularly as this changes as the population ages.
Combining the analyses in this way gives us a unique indication of the relative priority that different forms of NHS-run services have been given over most of the last decade, and an insight into the particular challenges experienced by community health care services, which have seen very significant real-terms spending cuts relative to need. For more information on our methods, see the box at the end of this explainer.
All of our charts show funding for current expenditure, that is: funding for everyday consumables such as staff costs and medicines, rather than for capital investment on equipment, buildings and IT systems, which tends to fluctuate from one year to the next (as large projects are procured and completed) and is not available at a patient care or service level. Our analysis covers around 90% of health care current funding over the seven years from 2016/17, including the local authority public health grant. When adjusted for economy-wide inflation, our analysis shows that while total funding on patient care has increased in real terms by 20% over the seven years – an average of 3.1% a year – how that growth has been shared between different types of service varies quite radically.
Funding for NHS acute, ambulance and NHS mental health care services has grown much faster than overall funding, with average annual increases of 4.4%, 5.2% and 5.3% in real terms respectively. As a result, these services have increased their share of total funding, with the largest share increase seen in mental health, which increased its share of the total from 8% in 2016/17 to just over 9% in 2022/23. By contrast, funding for NHS community health care services has grown at a much slower rate, increasing by only 3.2% over the entire seven-year period – the equivalent to average annual increases of just 0.5% a year, leading to a reduction in its share of total funding from 8% in 2016/17 to 7% in 2022/23.
The largest decrease in funding was in the local authority public health grant, which fell by a real-terms 21% between 2016/17 and 2022/23 (despite a brief increase for two years during the pandemic), decreasing its share of the total from 3.6% to 2.3%. Funding for GP primary care has grown at a similar rate to overall funding growth, at 3.3% average real terms a year, entailing that its share of the total has stayed constant at around 8% over the period. Funding for community pharmacy, optometry, dentistry and on prescriptions has also fallen in real terms over the period, resulting in the share of the total going to those services also shrinking.
This analysis shows that despite the often repeated ambition to shift more care and health care resources into services “closer to home” – to keep patients out of hospital, to better manage long-term conditions and to boost primary and secondary prevention – funding for the NHS services that primarily provide those forms of care (NHS community care services, primary care and local authority public health services) grew in real terms by less than half a percent on average a year between 2016/17 and 2022/23. Perhaps even more strikingly, funding for NHS community health care services was cut in real terms in three out of the six years between 2016/17 and 2022/23.
And although community service funding did enjoy a sudden uptick of just over 5% in 2022/23, that modest recovery was almost entirely driven by the centrally funded, one-off, non-consolidated pay rise (or “Covid bonus”) awarded to staff on the Agenda for Change contract that year – making it even less certain that the 2022/23 turnaround in community service funding will have been sustained into the financial year that followed, and for which accounts are not yet available. By contrast, the mental health investment standard commitment to increase funding for mental health services at a faster pace than other forms of care does appear to have been met.
When assessing changes in health care funding between different services, it is important to also consider changes in the population such services are intended to serve. A crude way to do this would be to examine funding levels against the population count. While that would factor in growth in the total population, it would ignore the fact that the population is rapidly ageing, with the numbers of people aged 70 and above growing at over nine times the rate of people aged five to 29 over the period here examined.
That demographic change matters to health care funding needs, because health care cost tends to increase dramatically with rising age, with an 85-year-old male expected to consume on average around 16 times more in acute care cost terms than a 10-year-old male – although at an individual level, health care need and the pace of biological ageing, as opposed to chronological ageing, varies very dramatically by deprivation level, with poverty and other forms of disadvantage closely linked to higher levels of poor health at an earlier age.
Adjusting real-terms funding levels for age and sex-related estimates of need reveals interesting differences in the pattern of funding per need-adjusted head across the different types of care. Most strikingly, the analysis shows that real-terms funding for NHS community services per head of need ended 2022/23 around 4.2% below the level of 2016/17, with even sharper falls for dental, community pharmacy and ophthalmic service funding. While the local authority public health grant per raw (unadjusted) head of the population rose sharply in 2020/21, it fell back almost as rapidly, and by 2022/23 was 24% below the level of 2016/17.
For the four forms of health care service that did experience consistent real-terms increases in funding over the full period – acute hospital care, ambulance services, mental health and GP primary care – adjusting for need moderates the rate of that growth, reducing the overall average level of real-terms funding growth per head to 2% a year over the period from 2016/17 to 2022/23. And with the exception of mental health services and the sudden increase in community health care funding mentioned above, our analysis shows that 2022/23 saw a real-terms decrease in funding per needs-adjusted head for all other services, as the NHS was asked to rein back spending increases permitted during the pandemic.
This decrease averaged at 2.2% across all services, but ranged from a 0.6% drop for acute hospital care, to a 7% cut per head of need for GP primary care, to a 16% cut for community ophthalmic and pharmacy services, and a huge 30% cut per head in the local authority public health grant, measured against total unadjusted population.
It is important to note that our figures for acute, mental health, community health care and ambulance care are drawn from the accounts of NHS providers of those services, and therefore exclude direct commissioner expenditure with alternative providers in the private and voluntary sectors (the details of which are not available by service type). If that data were available, funding going into private and voluntary sector health care services could either increase or decrease the rate at which overall funding for particular forms of health care have grown, but such analysis would tell us less about the level of resources going into NHS-run services, which deliver the vast bulk of NHS care.
As noted above, NHS provider spending in the private and voluntary sector – for instance, through outsourcing – has also been excluded from our figures. This is particularly important when assessing the level of resources going into NHS mental health services, as some mental health providers since 2020/21 have taken on some commissioning responsibilities, which means that in addition to providing services directly themselves, they also receive funding to purchase care from others, both within the NHS and outside it. That move has seen the headline incomes of those organisations grow quite significantly, but a large portion of that growth has in fact been passed on to other providers.
For example, in 2022/23 NHS provider trusts (predominantly mental health trusts) received £1.2 billion income specifically for commissioning mental health services from other organisations – up from £149 million in 2020/21. Of the £1.2 billion in 2022/23, roughly half was spent on mental health care services from other NHS trusts, and the other half on services from the private or voluntary sector. In addition to this, NHS provider trusts outsourced a further £2.5 billion of health care to private and voluntary sector providers in 2022-23, up from a total £1.1 billion in 2016/17. As this spending does not represent expenditure on NHS service capacity, we have excluded it from the analysis above.
It is interesting to note that over the period examined, however, NHS trust spending on outsourcing has grown at a faster rate than their funding for services they have provided directly. For acute care, this means that spending by NHS trusts with non-NHS health care providers grew from the equivalent of 1.2% of their funding level in 2016/17 to 1.9% in 2022/23. That steady rate of growth is something we expect to have increased further in 2023/24, as the combined pressures of industrial action and the need to reduce waiting times for elective care saw quite a significant increase in the number of NHS patients treated in the private sector.
The most significant growth in outsourcing has been in mental health services, where even without the recent move for provider trusts themselves to hold part of the mental health commissioning budgets (which may mean that some outsourcing spend previously recorded as having been spent by NHS commissioners is now recorded as spending by NHS providers themselves), NHS mental health trust outsourcing increased from the equivalent of 2.9% of funding for NHS-run services in 2016/17 to 5.1% in 2022/23.
The need for a sustainable long-term vision
Politicians from the two largest political parties in England have been keen to highlight a potential expanded role for the private sector in tackling NHS waiting times for acute elective care. But with the NHS budget this financial year (2024/25) rising by only 1.4% in real terms (less than half the rate at which it has increased on average each year since 2016/17), there are questions as to the extent to which spending a higher proportion of an already very tight budget on providers outside the NHS will lead to a further deterioration in the quality and sustainability of NHS-run services.
This is particularly the case for those services such as community health care, primary care and public health, which will need to sit at the heart of a sustainable, long-term vision for health and social care, where ill health prevention is prioritised and better long-term condition management both improves healthy life expectancy and reduces the need for emergency acute hospital care, which is one of the driving factors behind today’s high waiting lists.
The Nuffield Foundation is an independent charitable trust with a mission to advance social well-being. It funds research that informs social policy, primarily in Education, Welfare, and Justice. The Nuffield Foundation is the founder and co-funder of the Nuffield Council on Bioethics, the Ada Lovelace Institute and the Nuffield Family Justice Observatory. The Foundation has funded this project, but the views expressed are those of the authors and not necessarily the Foundation. www.nuffieldfoundation.org / @NuffieldFound
Methodology
Our analysis relies on three sources for funding levels:
For GP primary care, optometry, dentistry and prescribing we use spending figures reported in NHS England’s annual accounts for the years from 2016/17 to 2022/23.
For public health, we use figures published in the Department of Health and Social Care’s annual accounts for the local authority public health grant, again for the years from 2016/17 to 2022/23. We do not adjust these figures, with the exception of our adjustment for inflation and population need, which we describe further below.
For acute care, mental health care, community health care and ambulance services, we use the Trust Account Consolidation (TAC) accounts data, published by NHS England. This data is drawn from the audited accounts of NHS provider trusts and has been published on a consistent basis since 2017/18. For 2016/17 figures, we use the “prior year” figures in the 2016/17 accounts. The latest available files are for 2022/23.
To estimate the level of funding (or income) going into the four broad categories of NHS-ran services reported in the annual accounts, we extract the following from the TAC files:
Provider trust patient income by broad care type (acute care, mental health, community services, ambulance services). We net off from this mental health income that was received from 2020/21 by lead providers in “mental health collaboratives” and was intended for the commissioning of other providers (both within the NHS and outside it). As the TAC files do not contain consolidation adjustments between NHS providers, including this income line would lead to double counting – for example, with one trust reporting the income first as a commissioner, and then another reporting it as income for providing the service. Instead, our method only counts the income when it is received as mental health funding for a service provided.
We then adjust those patient care funding levels to reflect the following:
- “Other patient care income” reported in the provider accounts but that is not designated to a specific form of patient care. This includes funding for elective recovery and for centrally funded pay rises. This section in the accounts also includes central funding for the 2019/20 increase in employer pension contributions and income received from treating private patients. We exclude those two specific “other income” lines as they do not represent funding for additional levels of NHS care (and in the case of the 2019/20 pension increase would distort the time series).
- Income and expenditure surpluses and deficits. As a deficit represents a higher level of expenditure on a service than is reflected in income, we convert deficits into positive figures and add them to service funding levels for that year. As a financial surplus indicates that funding received was not entirely spent on a service, we do the opposite and deduct surpluses from funding levels.
- Central funding to support financial sustainability and/or additional costs of dealing with the pandemic. This funding is not reported as “patient care” income yet directly supports patient care, and so we have included it.
- Outsourcing to both other NHS organisations and to organisations outside the NHS, excluding where that is part of mental health collaboratives, which has already been taken into consideration at the patient care line. As with the adjustment to exclude mental health collaborative commissioning income, this adjustment is to avoid double counting patient care income when one NHS provider receives funding for care that it subsequently outsources to another NHS provider. This adjustment also excludes funding thafive is subsequently spent in the private and voluntary sectors and so does not represent increased spending in NHS-ran services.
Figures for the above sources of income and expenditure are not available at patient care service level. Instead they are reported at individual provider trust level and can be aggregated into 5 broad provider trust types reported in the accounts: acute, ambulance, mental health, community and specialised. These designations are broad. For example, a community health care trust may in fact undertake a significant portion of mental health patient care (and receive funding for such) and a specialised trust may undertake both community and acute care. To cater for this, our analysis calculates the percentage share of patient care category income received by each trust type (net of mental health commissioning income) in each year, and then allocates the adjustments listed above to patient care category according to those shares.
Without these adjustments, real-terms funding increases for all four patient care sectors reported in our analysis would be lower over the 2016/17 to 2022/23 period, most significantly for community health care, which would have recorded just 1.3% real-terms growth in funding over the entire period, instead of 3.2% after we have adjusted for additional funding streams. Overall, our adjustments increase total real-terms funding from 26.9% over the full period to 27.6%.
To adjust funding levels for inflation we use the official GDP deflator, published on April 2 2024. This is the government’s preferred measure of inflation and measures inflation in the domestic economy and so excludes the impact of prices on imported goods and services. In practice, the NHS at times experiences significantly higher levels of inflation, particularly on drug costs and other imported goods, and also on staff costs, which can be sensitive to broader, consumer measures of inflation.
To adjust funding levels by population need, we have used the age cost curves set out by NHS England in the 2023/24 and 2024/25 allocation formula for distributing NHS funding to different areas based on the age structure of their populations. These figures set out how health care expenditure for different types of service vary by age band. By applying those figures to five-year age bands for the population in England between 2016/17 and 2022/23, we have estimated the “needs adjusted” population appropriate to each service. For dental services, ophthalmic services and community pharmacy, we have used the primary care needs adjusted population. For public health, we have not calculated a needs adjustment and instead used the actual raw population.
Suggested citation
Gainsbury S and Julian S (2024) “Where does the NHS money go?”, Nuffield Trust explainer