Intended for healthcare professionals


70 years of NHS funding: how do we know how much is enough?

BMJ 2018; 361 doi: (Published 14 June 2018) Cite this as: BMJ 2018;361:k2373
  1. Anita Charlesworth, director of research and economics1,
  2. Karen Bloor, professor2
  1. 1The Health Foundation, 90 Long Acre, London WC2E 9RA, UK
  2. 2Department of Health Sciences, University of York, York YO10 5DD
  1. Correspondence to: K Bloor karen.bloor{at}

Although taxation is arguably the most efficient funding mechanism, NHS expenditure is a political choice, say Anita Charlesworth and Karen Bloor

In his landmark 1942 report, William Beveridge said that a comprehensive health service would “diminish disease by prevention and cure”1 and that costs would consequently be relatively stable. This optimism was unfounded, and vast unmet need took doctors and policy makers by surprise in the early days of the NHS. The first winter crisis was in 1948, and substantial overspends in the NHS’s first two years alarmed the Treasury. Responses included introducing user charges for optometry and dentistry, which prompted Aneurin Bevan’s resignation, and commissioning the Guillebaud committee to investigate NHS costs. “Rolling crises”2 have characterised debate about NHS funding ever since, often accompanied by lobbying not just for increased spending but for increased user charges and diversity of funding streams.3 There is no right answer to the level of NHS funding, but we can approach the debate in several ways.

How does the NHS compare?

The UK spends around 10% of its gross domestic product (GDP) on health.4 The proportion of GDP spent by government on health has doubled since 1948.5 Most other industrialised countries have also seen rapid and sustained growth in health spending, exceeding inflation and economic growth, irrespective of their funding systems (tax, social insurance, or private insurance) (fig 1).6 The UK’s increase in expenditure has exceeded what would be expected from increasing overall national wealth.

Fig 1
Fig 1

Growth of real health spending and GDP per capita in OECD countries (2000-15)

UK health expenditure has increased by almost 4% a year in real terms over the lifetime of the NHS.5 Real per capita spending was £268 in 1949-50, increasing to £2273 by 2016-17. But spending has been volatile, driven by both political ideology and the need to fund substantial reforms (fig 2).5 Periods of relative famine, creating political crises fuelled by concerns about quality of care, have been followed by periods of relative feast, often stimulating price and wage inflation.7 Particularly at crisis points, governments and other organisations have tried to identify how much needs to be spent and how to convince the public and service that it is spending the right amount.

Fig 2
Fig 2

Average yearly change in UK health spending over parliaments, 1951-52 to 2014-15

In most sectors of the economy, the amount we spend is simply the aggregation of each individual’s decision about where to spend their money. Health is different. In common with most countries, and for sound economic and social reasons, the UK funds most of healthcare publicly (around 80%),4 so decisions are made through a political process. The scale, importance, and public funding of healthcare puts the focus on spending and whether it should grow. Fig 3 shows that government spending on health accounts for almost £1 in every £5 of the taxes paid in the UK, double the proportion of spending when the NHS was founded.8

Fig 3
Fig 3

Annual UK public spending on health as a percentage of total public spending. Source: OBR Public Finances Database.8

If we choose to spend more on health and care this means either giving less priority to other public services or raising overall tax revenues. These preferences are based on individual judgments of desirability—but there are ways to benchmark spending decisions for the NHS and social care that can help guide public debate and policy choices.9

What metrics can we use to determine the “right” amount of funding?

Broadly, existing data can be categorised as focusing on inputs or resources for healthcare, on processes (which might indicate mismatch between demand and supply), and on outcomes.

In 2000 Tony Blair compared international input data to benchmark NHS spending, famously committing on national television that the UK would spend in line with the EU15 average share of GDP. In 2000 the UK spent just 6.3% of GDP, well below the EU15 average of 8.5%. Blair’s commitment was substantially met, spending increased substantially, and, even with the period of austerity after 2010, the latest data show that the UK’s share of GDP devoted to health matches the average of the EU15 and exceeds that of EU28 and the Organisation for Economic Cooperation and Development (OECD).456

The UK currently gives roughly the same priority to healthcare as other European countries do, and the mix of public and private spending is also in line. But devoting a similar share of GDP to health does not mean that the same resources are available—this depends on GDP per capita and input costs (clinical staff, facilities, drugs, and equipment). The UK’s GDP per capita is below the EU15 average, which implies that less money per person is spent on health. In terms of tangible resources and measures of process the UK compares poorly. The UK has below average numbers of doctors, nurses, hospital beds, MRI machines, and CT scanners per head.10

After Blair’s announcement in 2000, the government commissioned Derek Wanless to undertake a systematic investigation into the health funding needs of the NHS in England.11 The Wanless review was a broad examination of the effect of population and demographic changes; the health status of the population, including risk factor trends (smoking, obesity, alcohol); expectations of service quality (waiting times, wider patient experience); the scope to improve health outcomes in key areas such as coronary heart disease and cancer; and the role of new technology in extending the range of treatments and the productivity of the service. A similar approach was taken for the Five Year Forward View programme in 2014.12

The core components of these analyses are reflected in international modelling work by the OECD and the European Commission to produce projections of trends in future spending.1314 These studies focus on the contribution of five main factors: population size and age structure, technology, productivity, pay and other cost pressures, and the relative prioritisation countries attach to healthcare as they get richer (what economists call income elasticity). The UK’s official economic forecasting body, the Office for Budget Responsibility, uses a similar method to look at long term trends in health spending.15 These macro studies all find that, for the UK and other developed countries, health spending is projected to continue to outpace inflation and economic growth. The budget office’s analysis found that long term future pressures grow at around 4% a year—although this is in line with the rate of increase in spending in the 70 years since the NHS was founded, it is much higher than rates of health spending increase seen since 2010. Detailed modelling of demographic change, population health, and cost data for the UK also finds that, to secure modest improvements in NHS services, funding increases of around 4% a year would be required over the next 15 years.16

Is spending on the NHS a “good buy”?

Ideally, decisions about health spending would be based on likely improvements in health outcomes. Health economists consider whether the benefit of additional services justifies their cost, which is difficult for an individual clinical intervention and virtually impossible for the whole system. Evidence shows that care quality and health outcomes did improve in the UK in the period of rapid investment in the NHS that followed the Wanless review. Most obviously, waiting times fell, more staff were employed in the NHS, buildings and equipment were modernised, access to mental health services improved (albeit from a very low base), public satisfaction increased, and mortality amenable to healthcare fell.17 But input costs also rose, some of the extra funding went into higher relative pay, and there were concerns about the productivity of the service.17

Most data indicate that healthcare is not the prime predictor of overall population health.18 Considering changing patterns of disease, the National Institute for Health and Care Excellence found that around 30% of recent health improvements might be attributed to healthcare.19 OECD analysis of 35 countries found that between 1995 and 2015 health spending doubled in real terms, and this was associated with an average increase in life expectancy of almost three years. Improvements in income and education were also major contributors to improved life expectancy, with reductions in smoking and alcohol consumption making an important but smaller contribution.20

Healthcare costs might increase at a faster rate than other areas of the economy for many reasons. Healthcare is labour intensive, which means that there are genuine constraints to productivity improvements, and the share of the overall economy taken up by healthcare has long been predicted to increase over time.21 In other industries, technology has lowered costs, whereas the opposite has generally been true in healthcare. Patents are necessary to encourage medical research, but they create monopolies and a need for price and reimbursement regulation using mechanisms like NICE. Over optimistic predictions that technology will reduce healthcare costs—most recently characterising debate around digital health—often fail to account for the need for human interaction in the care process, the increasing complexity of patients with multiple morbidities, and the potential for widening stubborn health inequalities.

By any measure, over 70 years the NHS has succeeded in the core objectives at its creation: financial protection and equity of access to care. Healthcare remains largely free at the point of use, and the NHS provided services that carry charges (dentistry, optometry, and prescriptions) have not changed since the 1950s. Rationing decisions have been made, particularly since the introduction of NICE, which mean that some interventions are not funded by the NHS at all, but so far these remain relatively few.

Some of the current pressures on the NHS and other systems around the world are a product of the success of healthcare: making previously fatal illness survivable has resulted in increases in chronic disease, comorbidities, and public expectations. In recent years these pressures have been exacerbated by funding constraints not just on the NHS but more severely on social care, which is inextricably linked to increasing demands on the healthcare system. Process measures that may indicate excess demand, such as breaches in mixed sex wards and waiting times (in emergency departments and for planned surgery) have recently worsened, particularly since 2013-14.22 Some outcome indicators, including life expectancy and infant mortality, are showing signs of worsening, although many factors contribute to these trends.23 Brexit is likely to cause further resource constraints, particularly in terms of staff.

NHS expenditure, like other areas of public finance, is essentially a political choice, hopefully reflecting public opinion. Left leaning governments tend to nurture public trust in their management of the NHS, and right leaning governments are ideologically drawn to constraining or reducing the size of the state, although other factors clearly contribute. Regardless of political trends or economic circumstances, public support for the founding principles of the NHS remains high.24 Recurring debate about the need for a hypothecated (earmarked) “NHS tax,” although gaining in popularity at present, has substantial practical limitations and does not overcome the basic issue of how much spending is enough.25

General taxation is arguably the most efficient and equitable way of generating NHS finance; it has the widest tax base, it is progressive, and it has low administrative costs of collection—characteristics that cannot be as clearly shown by many of the alternatives. But resistance remains; a former permanent secretary to the Treasury recently tweeted referring to the NHS as a “bottomless pit,”26 reminiscent of the Treasury’s panic almost 70 years ago.

So, how do we know how much spending is enough? As outlined, we can examine inputs (by comparing spending with other countries), processes (by examining waiting times, which might indicate demand exceeding provision), and outcomes (by analysing life expectancy over time, although this is complicated by wider determinants of health). Interestingly, after the creation of the NHS, Bevan said in a speech to the Royal College of Nursing that “we shall never have all we need. Expectations will always exceed capacity.”27 Technical indicators of expectations, capacity, need, and demand for the NHS can inform expenditure decisions, but amount of spending remains largely a matter of politics.

Key messages

  • Debate about the “right” amount to spend on the NHS has recurred regularly over the past 70 years

  • Metrics to decide this can focus on inputs to healthcare (eg, analysing resources over time, comparing spending levels between countries), processes (eg, waiting times), or outcomes (eg, mortality rates)

  • Even with the spending constraints since 2010, the UK’s share of GDP devoted to health matches the average of the EU15 and exceeds that of EU28 and OECD countries

  • General taxation is arguably the most efficient and equitable way of generating NHS funds—alternative funding mechanisms are likely to be inflationary and do not overcome the basic issue of how much is enough

  • Technical indicators can inform decisions, but NHS expenditure is essentially a political choice


We thank the reviewers for their useful comments, and also Michael Hill, University of Newcastle, and Jonathan Bradshaw, University of York.


  • Contributors: AC is director of research and economics at the Health Foundation. She led a team working with Derek Wanless on his 2002 reform of NHS funding and was specialist adviser to the House of Lords’ select committee on the long term sustainability of the NHS. KB is professor of health economics and policy at the University of York, and co-lead of a fast response analytical facility funded by the Department of Health Policy Research Programme. Both authors in collaboration drafted this paper and both have agreed the final version. AC is guarantor.

  • Competing interests: We have read and understood BMJ policy on conflicts of interest and declare the following: AC has provided advice on future funding of the NHS to the House of Lords select committee and is a member of the Office of Health Economic research and policy committee. KB coleads a fast response research facility funded by the Department of Health's Policy Research Programme, (grant number PR-X06-1014-22005).


View Abstract