The NHS is, at bottom, a system for rationing healthcare. Give it more money, and the line determining how much it is prepared to spend to try and save a life, or make a patient more comfortable will be drawn a little more generously. Give it less, and those euphemistic "difficult decisions" will need to be made.
Pick a number, then, and the NHS will almost certainly find a way to spend it.
In the real world of course, that is not how decisions about funding the NHS are made. What we spend on the NHS is instead always a trade-off between how much healthcare we want for ourselves and our fellow citizens, and how much tax we are willing to pay for it.
And as the chief executive of the NHS in England made clear recently, the tight spending envelope set by the government up until 2020-21 entailed a "trade off" in the form of longer waiting times for routine operations and a crack-down on medicines deemed of "low clinical value".
So far this general election, none of the three main political parties in England have made NHS spending proposals that will negate the need for those trade offs and others. Indeed, as analysis at the Nuffield Trust shows, under the proposals of all three, we would be spending a smaller share of our national wealth on the NHS in 2022/23 than we will spend this year.
How then, might the line be drawn differently?
Constant share of national wealth
One simple solution might be to at least keep the share of spending on the NHS as a proportion of our national wealth (measured in terms of Gross Domestic Product) constant, at the 7.3% share it is today. Under that scenario, if the economy grew at its projected 2% real terms a year, so too would the NHS budget. In a way, that potential solution would pay for itself, as the tax base needed to fund the health service would grow more or less in line with the size of the economy.
That would require the NHS budget to be £13bn higher in real terms 2022/23 than it is today - the equivalent to around £2.6bn extra each year.
But while the size of the economy is a good indicator of how much NHS spending we can afford (without substantially increasing taxes or reprioritising other forms of government spending) it is a very arbitrary measure of how much might be needed. For one thing, GDP could shrink, if there was another recession, yet that certainly wouldn't entail people would get less sick.
A more immediate problem is that actual costs in the NHS are at present rising faster than the rate at which the economy as a whole is growing. So keeping up with GDP would still require potentially unpopular trade-offs, such as a continued freeze in NHS staff pay awards (despite significant shortages for clinical and medical staff) or the waiting time increases mentioned above.
Meeting current rate of increase in NHS costs
So what would a spending scenario based on forecast increases in the cost of things the NHS actually buys (medicines, doctors and nurses) look like? Such a forecast would need to take account of three factors: first, NHS specific inflation. If the cap on staff pay is removed, cost inflation in the NHS is set to be in the region of 2.9% from next year onwards.
Some of that cost inflation - around half of it - could be off set through NHS hospitals and other services being more efficient. Those two factors combine to determine the cost of each unit of care the NHS provides - in essence, the cost of treating each patient. However demographic change and increased demand for healthcare means we want more of it, and at a higher quality, with every year that passes - treating more patients, with more complex needs, through more advanced care. So although, in real terms, the trend is for the cost of treating each patient to fall, the increased number and complexity of patients cared for each year means that overall NHS costs are set, on recent trends, to rise around 0.7% faster than inflation in the economy as a whole.
Setting an NHS budget on the basis of these recent trends in costs, efficiencies and activity rates would require spending in 2022/23 to reach £141billion in today's money - £18billion higher than in this year. That would see NHS spending increase from 7.3% of GDP today to 7.5% in 2022/23 and would be the equivalent of spending an extra £338million a week on the NHS that year, compared to spending now.
Getting back to the long term average
But that level of spending increase would still be lower than the rate of growth the NHS has been given since its establishment, which averaged around 4% real terms growth a year until 2009. Getting back to that average would bring spending in 2022/23 to around the £150billion mark, and push spending as a proportion of GDP up to 8% - still only just more than half the level of GDP spent on public and private healthcare in the United States today.
Matching independent economists' projection
But the past is not necessarily a perfect guide as to what will happen in the future. The independent economists at the Office for Budget Responsibility have made their own projections about possible NHS spending into the long term future (up until 2066) with much riding on what assumptions are made about factors such as the future health of the population, and cost increases for drugs and new treatments.
Under one of their scenarios, they project NHS spending to rise by around 4.5% a year in real terms, which would bring spending in 2022/23 to £155bn, or 8.3% of GDP.
Under each of these scenarios, all three of the main political parties in England fall short. That leaves the question begging as to the cost of not increasing spending on the NHS. At the very least, the funding pledges set out by the parties would mean NHS activity not increasing at the rate we have become accustomed too - entailing longer waiting times but also potentially slower uptake on new drugs and treatments. That could see the NHS in England falling behind standards in other, similar countries. So far this election it seems those are prices all the main political parties are prepared to pay.