Winner of the New Statesman SPERI Prize in Political Economy 2016

Monday 5 November 2018

Health spending over time

There has been some comment on the fact that, with recent increases in spending on the NHS, the health budget is taking a growing proportion of UK state spending. I am missing Flip Chart Fairy Tales, so here is a chart heavy post to make one or two obvious points that regrettably are often missing from political reporting.

The first is that health has been taking up a growing slice of our total expenditure (i.e.GDP: expenditure on everything including investment) for a very long time. Here is a chart from a recent IFS publication which is a good source for more in depth analysis.

Note that real spending numbers can be misleading: although real spending has increased since 2010, as a share of GDP it has not, which is a reversal of previous trends. That alone does not inevitably explain recent problems in the NHS, but it certainly could do.

So why is it only recently that the growing share of public spending has been so obvious? Again the IFS have a handy chart that goes a long way to providing the answer.

In 1955/6, defence spending was over 20% of total spending, while by 2015/6 it had fallen to just 5%. This peace dividend (actually two: first a retreat from empire and then the end of the cold war) masked a steady rise in heath, which was only 7.5% of total spending in 1955/6 but was approaching 20% by 2015/6.

Many economists would simply describe this as reflecting that health was a luxury good, which means that spending as a share of income rises when income rises. Not all the evidence confirms this, e.g. the spending patterns of lottery winners. In reality I think there are various things going on. One may be that medical science has got better at prolonging life faster than it has held back the aging process. Another is that medical innovation is increasing the scope of what medicine can do. For example cancer is now increasingly survivable, but only with expensive care. While there is productivity growth in the NHS, it is below the national average and therefore fails to match increases in wages. In the document all the figures so far come from, the IFS expect these factors will require real health spending to increase by 3.3% each year over the next fifteen years.

Politicians, particularly those adverse to taxation, love to think that some kind of reorganisation will somehow change the inevitability of an increasing share of government spending and GDP. But this chart, taken from this source, suggests these trends are not some peculiarity of the way we organise things in the UK

In 1970 health spending was between 4-6% of GDP in these 5 countries, but by 2016 it was between 9-16% of GDP. (There is a definitional break in the UK series in 2013: there was no leap of spending in 2013 as earlier graphs show.) If there is any organisational lesson here, it is not to run a health service in the way they do in the US. It is indicative of the mess the world is currently in that politicians are busy trying to dismantle the positive recent reforms in the US and key politicians in the UK have once talked about making the UK health system more US like.

If the IFS is right, this inevitably means that taxes of some kind will have to rise significantly. Yet the Conservatives have repeatedly pledged not to raise any of the headline taxes, and Labour have felt compelled to match these pledges at least in part. That the budget included increases in the tax thresholds, and Labour’s internal spat over whether to vote for them, illustrates nothing has changed in this respect. This year this tax/spend dilemma was avoided by a tax windfall no one had forecast. But at some point in the near future something will have to give, and I really hope it is not once again the quality of our health services.


  1. It would be worth posting a table of health outcomes against health spending, too.
    The message I get from this is that we need to encourage rich individuals to spend more on therapies, thereby breaking the zero-sum system of funding via taxation. Do you have private healthcare? You should: it is one of the more public-spirited things you can do.

  2. The rise in health spending across the whole OECD looks inexorable. The one country that can easily reverse the trend for awhile is the USA, which could slash costs by copying the system of any other OECD country picked at random.

    If you want to be a techno-optimist, there are two grains of comfort. One is the analysis submitted to LBJ by the great medical writer Lewis Thomas. He distinguished three levels of medical technology: primitive (Victorian), cheap and largely ineffective; intermediate (iron lungs and sanatoria for TB), partially effective and therefore expensive; and the high technology of truly effective preventive measures and cures, like vaccinations and penicillin, which are cheap because they work. He proposed giving spending priority to the third. His hopes have not been borne out in the half-century since, and we are faced in say cancer with grinding incremental improvements of enormous cost, as each one will only have a short shelf life. But the story of bacterial ulcers gives hope: if an ulcer is bacterial in origin, as Barry Marshall showed, it can be cured - not treated, cured - with £100 of antibiotics.

    The second grain is even more speculative. European royalty are a pretty random genetic sample of the population since they started marrying commoners a century ago. They get the best medical advice and care (Queen Elizabeth made Prince Philip give up smoking to marry her in 1948, two years before Doll's first epidemiological publication). But they don't get magical super-doctors, just the best practice of the day. Their health and longevity is remarkable. Elizabeth is still working at 92 - it is work, even adapted to her age and organised by the best assistants in the business. It's a similar story in the Netherlands, Spain and the three Nordic monarchies. I expect that their lifetime medical expenditures (priced at the national norm. not the private rates they pay) are also low. The general population will, I suggest, follow them in due course. This is a long shot, but worth proper investigation.

  3. «key politicians in the UK have once talked about making the UK health system more US like.»

    Because "Thatcher dixit" of course, and the goal is a Pinochet style economy:
    «His first challenge at health was heading off Thatcher, who "wanted to go to the American system", he reveals. "I had ferocious rows with her about it. She wanted compulsory insurance, with the state paying the premiums for the less well-off. I thought that was a disaster. The American system is hopeless … dreadful." He prevailed on her to take a different route by introducing more competition into the NHS. It became known – in a phrase he didn't like – as "the internal market". Ever since then, successive governments have pushed in broadly the same direction.»
    «I was aware of the remarkable success of the Chilean economy in reducing the share of Government expenditure substantially over the decade of the 70s. The progression from Allende's Socialism to the free enterprise capitalist economy of the 1980s is a striking example of economic reform from which we can learn many lessons. [ ... ] Our reform must be in line with our traditions and our Constitution. At times the process may seem painfully slow. But I am certain we shall achieve our reforms in our own way and in our own time.»

  4. Those who wish the UK to have health outcomes similar to the USA have several options. For example, the lifespan in the UK is a few years longer than in the USA, for similar economic situations. So, if the UK wishes, the Parliament could move to encourage all elderly in the UK to voluntarily terminate their own lives. This Modest Proposal would save expenses in care, provide an inheritance economic stimulus, and help the UK to achieve the moral plane of the USA. Is that what the complainers desire?

  5. Economies are zero sum. Costs equal wages, especially for health care.

    What would individuals buy to employ as many workers in equal pay jobs if health costs were lower, ie paying 25% fewer workers?

    More food consumption to make the population fatter?

    Unless UK employment is maxed out, pretty much anything that the health care spending paying workers could switch to can be done by paying idle workers to work. Doctors and nurses, etc wont be better at building housing than currently idle workers.

  6. The trend does not necessarily reflect that health care is a luxury good. Health expenditures as a share of GDP would rise even if it is a necessity if health care prices grow more quickly than the aggregate price level (they do) without clear substitutes. The best explanation likely comes from Baumol: productivity growth for most medical treatments is extremely low and so a greater share of the economy's resources must be devoted to health care over time (and of course, taxes would have to steadily rise across time).


Unfortunately because of spam with embedded links (which then flag up warnings about the whole site on some browsers), I have to personally moderate all comments. As a result, your comment may not appear for some time. In addition, I cannot publish comments with links to websites because it takes too much time to check whether these sites are legitimate.