They're angels of course, best to use their time efficiently - Copyright NHS Public Domain

The Institute for Fiscal Studies are pretty good at the bean counting and they’re – well they say they are – impartial as to political wings. This doesn’t though make them wholly economically literate. This being something apparent in their report on the funding of the National Health Service, something they’ve done along with the Health Foundation.

Sorry, really, they’re not right here:

The NHS will need an extra £2,000 a year from every household in order to function properly, experts have said.

A joint report by the Institute for Fiscal Studies (IFS) and the Health Foundation found there was “no more room” to increase health spending by taking from other Government budgets and concludes that “taxes will have to rise”.

The new analysis of what the NHS needs to cope with future demands predicts that UK spending on healthcare will have to rise by an average 3.3 per cent a year over the next 15 years just to maintain NHS provision at current levels.

This is of course very much the conventional wisdom. Even Polly Toynbee tells us that the NHS always has had, always will have, a different inflation rate from the rest of the economy. I would always hope that people like the IFS would grasp the deeper point here but apparently not. Their report is here:

As new treatments are introduced, the cost of drugs used in hospitals is also rising.
Assuming new drug costs rise in line with recent experience, for each person treated in
hospital, the cost of their drugs would increase by 5.5% a year going forward.

That’s simply ridiculous. We only use new drugs if doing so is cheaper than not using them. This is obvious in comparison to other treatments of course and drugs are indeed cheaper. They’re – in a squinty sort of way – the automation of treatment. Automation is cheaper. For things we couldn’t treat before but do now we also apply pretty strict cost benefit analysis to their use. That’s what NICE does. More drugs, newer drugs, they’ll not increase the NHS bill, they’ll decrease it for any given level of curing stuff.

But it’s a deeper economic point where they stray:

Pay will also need to rise at least in line with public sector average earnings if the
NHS and the social care system are to recruit and retain the staff they need. The challenge
for all healthcare systems is that, as a service sector, healthcare productivity over the
longer term has traditionally lagged economy-wide productivity (the so-called Baumol
effect). It is true of all healthcare systems, however they are funded (tax or social
insurance) and however they are delivered (public, private or not-for-profit). The gap
between earnings growth and productivity is a key driver of spending pressures.

Entirely correct, Baumol and all that.

At which point we need to dive a little deeper. As any passing economist will agree we can increase output by increasing inputs. Like, say, pouring more money in to the NHS. We can also increase output by becoming more efficient – by, say, raising NHS productivity. We’ve also got some pretty convincing evidence about what works here:

How, then, have today’s advanced nations been able to achieve sustained growth in per capita income over the past 150 years? The answer is that technological advances have lead to a continual increase in total factor productivity–a continual rise in national income for each unit of input. In a famous estimate, MIT Professor Robert Solow concluded that technological progress has accounted for 80 percent of the long-term rise in U.S. per capita income, with increased investment in capital explaining only the remaining 20 percent.

When economists began to study the growth of the Soviet economy, they did so using the tools of growth accounting. Of course, Soviet data posed some problems. Not only was it hard to piece together usable estimates of output and input (Raymond Powell, a Yale professor, wrote that the job “in may ways resembled an archaeological dig”), but there were philosophical difficulties as well. In a socialist economy one could hardly measure capital input using market returns, so researchers were forced to impute returns based on those in market economies at similar levels of development. Still, when efforts began, researchers were pretty sure about what they would find. Just as capitalist growth had been based on growth in both inputs and efficiency, with efficiency the main source of rising per capita income, they expected to find that rapid Soviet growth reflected both rapid input growth and rapid growth in efficiency.

But what they actually found was that Soviet growth was based on rapid growth inputs–end of story. The rate of efficiency growth was not only unspectacular, it was well below the rates achieved in Western economies. Indeed, by some estimates, it was virtually nonexistent.

Planned, Stalinist, structures don’t increase productivity. Market based ones do. Hmm.

Note what Baumol has been saying here. Not that services will inevitably become more expensive compared to manufactures. Rather, wages are set by economy wide productivity. This is easier to increase in manufactures than services. Thus manufactures will become cheaper compared to services given this difference in relative productivity, while wages are driven by general productivity. But it is “more difficult,” not impossible. If we could increase services productivity as fast as general productivity across the economy then the Cost Disease disappears.

From the IFS report:

One of the great successes of the NHS in England since 2010 is that, despite very tight
spending settlements, activity has risen substantially. In other words, productivity has
grown and, unusually, since 2010 measured productivity in the health service has
been growing faster than productivity across the economy as a whole. Whether this
could be sustained over a longer period is unclear.

What have we been doing since 2010? Adding more market processes to a previously near Stalinist planned organisation. Productivity has been rising as a result. And it’s been rising fast enough to have completely reversed Baumol over this period of time.

Which is pretty good really. And it also tells us something about the future of the NHS and its funding. That is, we don’t – not inevitably at least – need to be pouring more inputs in. We can instead make the machine itself work more efficiently, raise the productivity of the use of those inputs. That is, continue to add market processes, the only way we know of to generally increase productivity over time.

We even have evidence that as we’ve been doing this, adding markets, it has been working, it has been increasing productivity. So, why not do more of what works? Rather than what this IFS report suggests, ignore our own recent experience of what works and retreat back to the decades long practice of throwing more money at it, the answer which doesn’t work.

Subscribe to The CT Mailer!

Leave a Reply

Please Login to comment
7 Comment threads
6 Thread replies
Most reacted comment
Hottest comment thread
7 Comment authors
SouthernerBB01bloke in spainTim Worstallmole125 Recent comment authors

This site uses Akismet to reduce spam. Learn how your comment data is processed.

newest oldest most voted
Notify of
Rhoda Klapp
Rhoda Klapp

New drugs that cure people allow them to live longer. And THAT is what costs the real money.


But that’s backwards; people want to live longer, enough to pay extra to get the new drugs, where the old drugs only let them live as long as they used to live. Instead of looking at how to give the people what they want, we are looking at how to save the NHS, and it will lead us to legislate that people shall not live longer, because that gets in the way.

And to think that single-payer health care is sold to a nation on the basis that insurance companies continually renege on their promises of coverage!


Yup, unless there are market incentives, it won’t matter how much more per-household the NHS gets. And there is no market incentive when “free health care” is everyone’s right, as many medical services as you say you want, from life extensions to tummy-tucks to curing the disease of having the sex organs that you have, and these entitlements are independent of reckless personal behavior. And there will still be waiting lines and people will still die of thirst on their gurneys in the hallway. “taxes will have to rise” — This ought to be chiseled on the gravestone of the… Read more »


New drugs for conditions that couldn’t previously be treated effectively cost more than a couple of asprin, bed rest and a funeral. So as more treatments are available the average price of the treatments will rise. Particularly as drug companies have the incentive to develop expensive drugs for new conditions rather than cheaper low profit drugs for existing conditions. In fact for rarer conditions, even if they have the drugs already developed (to treat something else like cancer) they may not necessarily fund the research to prove that it is effective as the costs are too high, even when they… Read more »

bloke in spain
bloke in spain

The problem with talking in terms of productivity is that there’s an assumption it will be productivity of something for which there is a demand. For instance, a shoe factory could vastly increase its productivity by only producing a shoe in a single style & size. And further, by only making shoes for left feet. Likewise the NHS can increase its productivity by sending out twice as many nagging health warnings. It can increase productivity by maximising the efficiency of its A&E staff & facilities by making the patient wait 4 hours to receive treatment. It can increase productivity by… Read more »


Baumol is one of those things where the principle seems obvious but it’s hard to find valid examples in practice. A symphony orchestra can’t play The Emperor more productively? Microphones and amplifiers allow them to play to fifty thousand in a football stadium instead of just five hundred in a concert hall. A CD would do the same, and television expands the audience to maybe five hundred thousand, five million? A nurse can only nurse one patient at a time? Hook the patients up to monitors and the limit per response team is determined by queueing theory. It must be… Read more »

bloke in spain
bloke in spain

And where did you get that photo of a line-up of nurses? If it was anything like the nurses one actually sees there’d only be room for four in the frame. And two & a half of those would be one nurse.


Archive picture from 1974.


State monopoly that cannot be contested, no price system, no profit motive, unknown costs, staff reward by length of service and grade not merit, patients/treatments cost without revenue, controlled by budgets, must meet political aims not practical aims… how can it possibly fail?