Examination of witnesses
John Appleby
Q49 The Chairman: Can I remind everybody that we are being broadcast now, so any conversations you have will be picked up and listened to by the world? I have to repeat this every time, which gets a bit boring, but if there are relevant interests please declare them at the time if they are not already stated in the Register of Members’ Interests.
Mr Appleby, we welcome you here to give crucial evidence. Can I thank you officially for allowing us to consider you as a specialist adviser? We had too many people apply, but we are grateful that you had considered that and you are here to give evidence. That is very important because we will get it on the record now and we could not do that if you were an adviser. I am happy for you to make an opening statement before we progress with the questions.
John Appleby: Briefly, given that this Committee’s objective is to look at sustainability, and, in a sense, jumping straight to my conclusion, there is an issue not so much of long-term sustainability in the period that you are looking at up to 2030, as is my understanding, but certainly of a real financial problem in the short and medium term. There is a danger of conflating what is going on now and over the next few years and what has happened over the last five years with a deep, systemic problem with the nature of the NHS and its funding.
The Chairman: Is that all you want to say?
John Appleby: That is all.
The Chairman: Of course, we have to emphasise that, whilst we recognise the short and medium-term problems that we read and hear about all the time, in this report we are trying not to get tangled up in the current debate.
John Appleby: I understand that.
Q50 The Chairman: We are trying to see if we can come up with ideas that would make the NHS sustainable in the longer term. I use the term in the broadest sense of prevention, social care and other healthcare. What are the projections up to 2030 for health and social care funding pressures from different models? What drives those projections? What is a realistic rate of efficiency improvement, because we hear a lot about efficiency improvement filling some of these financial gaps?
John Appleby: There are two types of model that are used to look at future spending. One model is so-called policy neutral. The Office for Budgetary Responsibility produces fiscal projections for public spending every year. This year has been a bit unusual. Given the Brexit vote, it has postponed its latest set of projections, which I understand will come out at some point this year although I am not sure. These are usually based on population projections essentially—what is driving demand from a population point of view—and not just the size but the demographic structure of the population. One way of doing that is simply to look at how much we spend by age group now and do a simple multiplication almost to see what happens into the future. It is not taking account of any policies regarding organisational change, which can be a pressure on costs, spending and so on.
There are other models, including the work that I know Anita was involved in with Derek Wanless, that are not policy neutral. Back in 2002, Wanless was asked to look at funding for the NHS over a 20-year period. It was not simply a mathematical calculation with population, although that was in there; there was also a desire to think about what sort of health service we want, how short the waiting times, what sort of quality and so on. That is a different sort of modelling. It is not so much a projection or prediction; it is more of a policy-driven type of model.
I should say that many countries do this sort of work—I have looked at this in the past—including Sweden, Australia and Germany, but not so much France. The US is very keen on looking at projections in health spending. The Congressional Budget Office does this work. It is a bit like the OBR but it also assesses the impacts of policy on spending. The Centers for Medicare & Medicaid Services—CMS—produces annually not just a projection of public spending in the US but private spending. It employs models to do with population. It looks at inflation, relative price effects, and so on. There is a variety of different ways of tackling this, but there are essentially two: one is policy neutral and one looks at what sort of service we want or how policies drive cost pressures, and so on.
I have looked at five models for the UK. The primary one is from the OBR. As I say, it produces this every year. I have figures from its 2015 report. I have also looked at Wanless, which is interesting, although that stopped in 2022, which is before the period that you are looking at. McKinsey has done some modelling work, as did the European Commission back in 2007, and the OECD. They have all come to slightly different conclusions about where spending may be going in the UK.
I can give you a flavour of the numbers. The OBR started with a base for UK NHS spending—and this is the number to remember—of 7.3%/7.4% of GDP in 2014. That is the baseline year. The OBR produces a central projection, and projections based on tweaking its model slightly, which is to do with different assumptions about NHS productivity. High productivity implies less need for more public money.
Lord Willis of Knaresborough: Can I clarify that you are talking about both NHS spending and social care spending when you give us this 7.4%?
John Appleby: No, that is just NHS spending. I also have the figures for social care and long-term care. The OBR looks at both NHS and long-term care spending. The baseline was 7.3%/7.4% in 2014. It does projections for 50 years, so I have just taken those up to 2030. It produces a range of numbers that go from about 6.8% to about 8.2% of GDP over that 16-year period. That is its range, and as I say it does about eight different tweaks to its model to get those different numbers, but they fall at around 7% as a ballpark figure for what the OBR projects.
McKinsey did some work in 2007 and produced a high and a low figure. Its low figure was 10% of GDP and its high figure was around 12% of GDP. The European Commission produced two figures, again dependent on two different scenarios. One scenario is where the UK health system would contain costs in some way, and the other is where this system would not be able to contain costs. The high figure was 11.3% and the lower figure was 8.4%.
Lastly, the OECD produced some slightly lower figures: by 2030, 7% and 8%, depending on the models it used and the assumptions it made.
There were a lot of numbers there and I apologise. I have offered to put these in a written note to the Committee, as I think they would be easier to consume that way. What I take from those sorts of figures is that they are not huge. The latest projection to 2030 for the US is something like 21% of GDP; over $1 in $5 of the entire US economy spent on healthcare. We are looking at a range of figures. If we leave McKinsey on one side for a minute, because I am not sure of the robustness of that, we are looking at between 7% and 8% of GDP over a figure in 2014 of 7.4%. It is not a huge rise over that period. I think those are broadly the figures for health. I am just checking my notes.
I want to say one thing about the short term. I realise that the Committee is not looking at that, but it has a bearing on where the figures are ending up, because they are starting from a certain position. The year 2009-10 was the peak year for spending on the NHS in the UK. It was around 7.8% of GDP on the NHS across the UK. The OBR predict that by 2019-20 that will have reduced to 6.2%. We are in a slump, as it were, and a lot of these projections come out of that slump and rise again.
To make a final point about the short term affecting the long term, on the OBR’s central projection it will take until 2049 for the UK to reach the peak it was at in 2009. That dip has a bearing on the long-term trajectory. That is my point about the short term: that that is a long time to get back to where the system was in 2009.
Taking all that together, my broad view about sustainability is that, at least within the period that we are talking, about these figures are not outwith anything that you might think is unaffordable. In a sense, we are getting back to where the system was in 2009, possibly a bit higher but not tremendously so.
You can look at some other countries. In 2014, the EU14 average was about 10.3%, which includes private spending, so even then the average was higher than most of the projections to 2030 for the UK. That gives you another triangulation of it. I have to say that international comparisons are difficult to make. There have been some changes in the way the OECD accounts for health and social care spending. That is my take on the projections for health.
You mention long-term care, and the OBR also looks at that. Others, including Wanless and the OECD, have also looked at that. In 2014, the baseline figure for social care/long-term care spending in the UK was around 1% to 1.1% or so of GDP. All projections show that going up, whether they are based almost purely on population change or not, as you would expect with an increasing proportion of older people in the population. The OBR’s central projection is from 1.1% to 1.6% in 2030. The various different tweaks to its model, as I mentioned before, do not have much of an impact on that. By 2026—so just short of your time period—Wanless was looking at something like 1.5% or 1.6%. The OECD had two figures of around 1.6% and 2.2%, again depending on the assumptions that it makes. The figures are in a ballpark of about 1.6% or 1.7% compared to 1.1%.
Lord Warner: Can I seek some clarification? Are we talking about publicly-funded services or publicly and privately?
John Appleby: Publicly funded.
Lord Warner: So all these figures that you have been talking about are publicly funded?
John Appleby: Yes. That is an important point, given the recent pressures on local authorities regarding public funds and the trade-off with people spending out of pocket and so on.
Lastly, I wanted to mention some work that was commissioned independently of the King’s Fund from Kate Barker and colleagues, who were asked to look at future funding for health and social care. I will not summarise the entire report, but in the end they went for some sort of combined figure of health and social care. They thought that it was important not to treat them separately. That also had implications for organisations regarding how services were delivered. They came up with some similar figures, to be honest, depending on whether you had free social care and how tight the eligibility criteria were for accessing that publicly- funded care. There are two figures from their report: 1.8% of GDP by 2030; or 2.3%, depending on how you tweak the eligibility and how tough or loose you made that. That is the range of figures for social care. It is hard to judge sustainability. They are relatively small percentages of GDP. As I say, the OBR predicts it rising to 1.6%. That is something like a 5% real increase each year on average between now and 2030.
The Chairman: Can you repeat that, to be clear?
Lord Willis of Knaresborough: Is it 5% in cash terms?
John Appleby: No, in real terms. The move from 1.1% to 1.6% of GDP—I will check my figures later—is something of the order of a 4% or 5% real increase each year. You have to remember that GDP is also growing, so although 1.1% to 1.6% is 0.5% of GDP, the pie is much bigger. Those are the sorts of projections that you have for social care. Barker and perhaps Wanless were looking at policy drivers, so they asked what sort of social care we would like to see. Kate Barker and her group certainly made the point that they would like to improve the quality of and access to social care.
The Chairman: What assumptions about productivity are made when you quote these figures?
John Appleby: Most of these projections have some assumptions about productivity. OBR assumes a high and low level of productivity. By the way, “high” is not tremendously high. For the economy as a whole, something of the order of 2% is generally thought of as labour productivity improvement on average across the whole economy long term. Over the last 20 or 30 years, by most estimates, the NHS has been trundling along at about 1% on average, or something like that. Recently it has had some better years. The OBR makes some assumptions, but they are not outwith broad assumptions about productivity and our experience of productivity in the economy as a whole.
I know that Wanless had quite a battle with the Treasury about what assumptions to make about productivity, because tweaking the productivity measure by half a per cent over 20 years accumulates quite quickly, and if you have a higher productivity assumption, that can be a big offset to the amount of public money going in.
The Chairman: I am going to move on. We will come back to some of this.
Lord Warner: Before we move on—and it is related to the question I want to ask—can we explore for a moment the point that you make? What struck me from what you have just been saying is that it will take a long time to get back to where we were in 2009-10. My question relates also to your point about assumptions about productivity improvements. In effect, you end up with a position where in the short term you could almost make it inevitable that you will never catch up. That seems to be the implication, or else you have suddenly to do what a previous Labour Government did—I am not making a political point; it is a factual point—which is to shovel a huge chunk of money into this system, and which is probably not a very smart thing to do.
John Appleby: And over a relatively short period, I think is also the point there.
Lord Warner: The inference from what you are saying is that if you do not build in some incremental increases and have realistic estimates of productivity, you are creating some very serious problems for yourself in the longer term. That is what I have taken away from this.
John Appleby: I suppose the point I was trying to make was on what the OBR is saying. You could decide to put money into the service, and that would be a policy decision to make, but the OBR’s figures are based more, as I say, on population projections, and that is what is driving its central projection. On that basis, it will take a long time to get back to where we were in 2009-10. I am staring at a graph here that I cannot hold up. There has been a large fall over a short period. The graph trundles along. There was a big increase in 2000 to 2009 and it has come down again—the OBR says that it will come down further—and then the projections sprout out of that trough, as it were. That is just the way the projections work out, but you could decide not to follow that path of course.
Q51 Lord Warner: Is the current health and care system fiscally sustainable? If not, what options should the Government be considering in the period, let us say, up to 2030?
John Appleby: The conclusion that I draw from these sorts of projections is that it is fiscally sustainable in the sense that we have been there before and the sky did not fall in. Decisions were taken to spend money. You can try to triangulate that with other things, such as international comparisons, although that can be a bit tricky. We have been in a period of austerity where budgets have been cut. The social care system has tackled that in a slightly different way from the NHS for a whole variety of reasons. On nearly all the OBR’s projections and those of others, the pressure is always up, but we are starting from a much lower base. My conclusion would be that by 2030 it is fiscally sustainable.
Lord Warner: From your evidence, are things not made worse if you do not regularly each year put a chunk in? If the demography is remorselessly rising, there have to be some guarantees about the annual increase, otherwise you keep falling behind. Is that what you are saying?
John Appleby: That is a political decision about the path you would take.
Lord Warner: Forget about whether it is policy and look at the arithmetic. You were saying that in health and care you would need to put something like 5% a year in real terms into the system to stop things getting worse, in effect. I am trying to get at this point about how we stop going backwards.
John Appleby: The 5% refers to long-term care, not to healthcare, but that figure is not far off. The long-term real increase the NHS has had since 1948-49 has been around 3% to 4% a year on average. It is very bumpy, of course.
Lord Turnberg: Is that in cash terms?
John Appleby: No, that is in real terms, allowing for general inflation that is the average increase that it has had.
Lord Turnberg: Is that as a percentage of GDP?
John Appleby: As a share of national wealth it went from about 3% of GDP in 1950 to that figure of 7.8% in 2009, so it has more than doubled as a proportion, but it has taken a long time to get there. I am not quite sure I understand your point, Lord Warner.
Lord Warner: My point is that I am trying to think about 2030, not 2016. I am struck by what you have said happens in relation to the demography and associated disease profiles if year on year you do not put realistic sums in real terms into the health and care systems, because you end up with a situation where you push yourself towards what Labour did, which is shovel in a big catch-up sum of money. I am trying to understand the flow.
John Appleby: I understand your point now. That is true. When you look at the historic path the UK NHS has taken as a proportion of GDP, you can see where the recessions are, where decisions were made, and so on. My point is that, yes, there has been a falling off recently, so we are potentially on a parallel path to where we could have been, with the potential that the system becomes so poor in its performance and quality that some decision has to be made, and suddenly decisions are made. I agree that it would be better to smooth that over a longer period.
Lord Willis of Knaresborough: I want to link that line of questioning to the previous one, because “fiscal sustainability” is quite a glib phrase. When this report comes out—I was going to swear then—it will mean very little to a lot of people. When you put the figures together for us, can you include the cash in real terms, because I think that really drives home that particular point?
Secondly, when you are linking sustainability to GDP, the reality is that GDP often goes down. It blips, as it certainly has from 2009. Counterintuitively you need to put in even more money in cash terms during those times to maintain a healthy line. Given that we have the Brexit issue, where we may—and I say only “may”—see a significant dip in GDP, albeit in the short term, how do we compensate for that in our projections, and indeed what do we say in our report to overcome that?
John Appleby: I would hate to use a figure of £350 million a week, but, anyway, I think that has gone.
Lord Willis of Knaresborough: Well done, Nigel.
John Appleby: Yes, and that links to my point about the OBR’s projections for this year. It has delayed them because of the Brexit vote. It is reassessing all its economic models. We do not know what will come out of it, but given what a lot of economists’ models have shown about the impact of Brexit on GDP in the medium to long term, it could be that GDP does not follow the path upwards that it would have done and it is slightly lower, in which case that could change some of these numbers because the denominator changes, of course.
Your point about fluctuations in GDP is a good point, and it is one of the issues that I would have with, for example, a hypothecated tax linked to GDP.
Lord Willis of Knaresborough: That is the point I am making.
John Appleby: I think there are tremendous problems with that. Just when you may want to put more money in, in a sense the thing driving the tax is going the wrong way and providing less money for the system. I certainly take your point about the jargon, the percentages of GDP and so on. I did the numbers work for Kate Barker’s committee, and that is one of the points she made. We did talk about percentages of GDP, but we also put in the cash figure to show the billions and billions that it actually means.
Q52 Bishop of Carlisle: You have referred several times to the Barker commission, including in your last answer, and the work that has been done on the integration of health and social care. Could we focus for a moment on social care specifically? My question has to do with the scope for efficiency savings in social care. I think you said a moment or two ago that social care tackled difficulties with funding in a slightly different way from the NHS. Looking to the future, what possibilities are there, particularly in light of the new living wage and the pressure that is putting on the whole system?
John Appleby: Some work was done on the living wage by the Resolution Foundation that I thought was quite interesting in that it showed that many care workers were getting an increase in their wages but that in fact the private sector was in a sense passing on the extra cost in part to local authorities. That is a good thing regarding people getting a better wage and so on.
The point I was trying to make, but did not quite make, about the way local authorities were tackling financial pressures compared to the NHS is, as I understand it, that local authorities have to stay within budget by law. The NHS does not. Last year, of course, the NHS in England at provider level overspent by between £3 billion and £3.5 billion. In a sense, it took its own decision to spend more.
The scope for productivity improvements in social care seems to me far less than in healthcare. One thing that local authorities have done to stay within budget is simply to close services. My first job was in the health service over 30 years ago, and that was what we did then too. We closed wards. We literally went round and chained doors, and consultants could not admit patients. The funding system was such that it was not payment by patient, so we saved some money, and then we opened the wards a month later. By the way, I think that was unacceptable then and it is certainly unacceptable now.
Lord Willis of Knaresborough: The good old days.
John Appleby: Local authorities have tackled their financial pressures in a slightly different way. They have tightened up tremendously on the eligibility criteria. The numbers of people receiving publicly funded social care packages dropped by between 25% to 28%/29% over five to six years. I am not sure where those people went. It is not as if they just disappeared. The presumption is that they started to self-fund, that they perhaps used the health service more—A&E, general practice and so on—and got more informal care and support. With the nature of social care, it is difficult to see where the productivity improvements could come. I have to say that there is not much data on productivity in social care. In fact, I have not come across any that looks at this. I just think that healthcare has a bit more scope over the longer term to improve its productivity.
Bishop of Carlisle: That is helpful. Do you feel that the integration of the two in the way that Barker has recommended would help overall regarding the finances?
John Appleby: I think so. We have a bit of a model in the UK. Northern Ireland, notionally at least, has an integrated health and social care system. I did some work there some time ago and remember talking to some of the social care people. I said, “This is great. You have them integrated, which is the sort of thing that everybody keeps talking about”, and they said, “The thing is when there is a bit of a financial squeeze, we are the ones who get it”. They were integrated but not necessarily in the sort of way that Kate Barker discussed, which was in a sense at a bottom-up, professional level. We all have elderly relatives and can see where the integration should happen. To talk about social care and healthcare is wrong, because it is all care. I think that is what Kate was talking about. It was not so much about saving money or being more efficient; it was more that it was the right thing to do regarding improving the quality of care that people received. At the margins in some areas it may be a bit more efficient, but not necessarily.
Bishop of Carlisle: That is very helpful, thank you.
Q53 Lord Warner: Can we go back to this issue of rates of change as between social care and healthcare? On the principle that the best predictor of future behaviour is past behaviour, is there some analysis from the data you have available to you that shows the extent to which the annual increases in social care and the health service, for the same demographics, are not running in kilter with each other? Your point about small annual mistakes in assumptions, particularly about productivity, has significant effects over time, and we are interested in what happens in the longer term. Anecdotally, if you talk to directors of adult social services, they will say, “We have had a poor deal compared with the NHS over a long period and we are going to go on getting a poor deal”. Is that true? Does the data support their claim? What are the implications when you project that behaviour up to 2030?
John Appleby: They have had a very poor deal. The NHS alone among publicly-funded services has had a relatively—and I emphasise relatively—good deal. I cannot remember the exact figures now, but local authorities have had something like a 30% plus real-terms cut in funding over five or six years. They have done their best to protect their services, especially adult social care services. I would remind you that the NHS is also putting money into local authorities, and has been for the last few years. It has now turned into the better care fund. Approaching £1 billion via the NHS budget is going into social care. Even with all that, when you look at the performance and activity of social care, as I mentioned before, you see that there has been about a 25% reduction in the numbers of care packages. Yes, they have had a tough time and there will be a crisis. In fact, there is a crisis now. Local authorities buy a lot of care from the private sector. They have been very tough negotiators, as far as I can see, in getting a good deal from the private sector, but there comes a point where you cannot squeeze down any more without cutting into the quality of care that is delivered to people.
Lord Warner: Could you go back 15 years and look at those comparative annual increases between health and social care and see, if you carried on down that path to 2030, what the result would be? This is quite a critical issue for us.
The Chairman: Are you able to send us that?
John Appleby: For social care I have figures back to 1994-95 for England. It is not the whole of the UK, but they can be scaled up. There are some trends there. For the NHS we have figures back to 1950, so yes.
The Chairman: Thank you. We look forward to that.
Q54 Baroness Blackstone: Can I come back to productivity savings in the NHS? You implied just now that you thought there was scope for more. The level of productivity savings is around 2% per annum at the moment. If you think there is scope for continuing at this level, or indeed increasing productivity savings, could you say what the components of these savings would be? In other words, where is the scope?
John Appleby: The 2% figure I quoted was a broad figure for the economy as a whole and all industrial sectors. The NHS fluctuates, but it is about 1% on the ONS figures, and from the work of the Health Foundation and the Centre for Health Economics in York. In some sense that is not bad, but it is not brilliant.
When I was at the King’s Fund I did some work looking at three areas that have driven productivity in the NHS historically. One was reductions in the length of stay. This is not unique to the NHS; it is across medicine and the world in health systems. People stay less time in hospital. That has allowed health systems to get rid of some beds and, more specifically, to treat many more people and improve the throughput of patients. That has been a big driver of productivity in the past. There is still some scope for that. I should say that it has taken a long time. It was not part of a five-year forward view. Over 20 to 30 years you can see consistent reductions in the length of stay. As far as we can tell, that was driven largely by changes in medical technology and anaesthetics and partly by changes in culture and recovery from operations: why spend your time in hospital when you could be at home? A combination of things drove that.
There are two other examples. The switch to generic drugs has been amazing in this country and in a lot of other countries. Without the increase in generic prescribing since the mid-1990s, the drugs budget for the NHS would be double what it is now. It has had a big impact. It has allowed more drugs to be prescribed per pound, so we have a bigger bang for our buck. Clearly, there is a limit. When you get into 80% to 90% of drugs prescribed and dispensed generically, there is not much further to go. Lastly, there have been changes in surgery and a big trend towards day casework. It has been cheaper and, I think, largely better for patients, and allowed hospitals to treat more people. These are all great productivity improvements.
Baroness Blackstone: But this is about the past.
John Appleby: There is a lesson to be learned from the past, which is that these things take time. You do not see any dramatic jumps. Also, they were often driven by medical technology and breakthroughs in surgical techniques, anaesthetic drugs, and so on. It was not a memo from the Department of Health imploring the system to be more productive. The lesson for the future is how you encourage that sort of medical technology in its broadest sense.
The Chairman: Coming back to Baroness Blackstone’s question, it may be that in the future drugs will be more personalised and more expensive and the technology for improving healthcare may become more expensive, so that whilst the use of generic drugs might be applicable now, it may not be then.
John Appleby: It is true that they may be more expensive at an individual drug level, but they may be more cost-effective, which is a different thing. We may be spending a bit more, but the benefits might offset that. There will still be interventions that require people to stay in hospital a long time. In mental health care, people often stay for a long time in hospital, so there may be more scope there. I am trying to draw a lesson from the past and see where things may go. It is hard to see some huge breakthrough that suddenly dramatically changes the productivity of the health system.
There is also Lord Warner’s point about incremental change on the productivity side. The health system has to try to get to grips with what helps to drive and encourage that. We have tried various things, including a payment system of payment by results. In part, that was designed to encourage hospitals to be more efficient. I am not quite sure what the results are on that. We have introduced various incentives. We have had a quasi-competition within the system. I have not seen any work that has suggested that that has bolstered efficiency and productivity that much. From my reading of the history on productivity, such that it is, I would emphasise that it takes time, and it is largely medically driven. It is how you encourage that sort of thing.
Lord Turnberg: I wanted to ask about productivity being measured solely by what the NHS does, when, if we cure someone with some very expensive drug and they become productive in other ways, that productivity never gets put in. If they go back to work and they are no longer on social benefits, that productivity never gets calculated in, and yet it is an important element.
John Appleby: You are right. In a sense you are talking about the benefits and the wider outcomes of a health intervention.
Lord Turnberg: And financial benefits.
John Appleby: Yes, and financial benefits. There are studies of that. The word “productivity” has a very precise meaning in economics. It is like an engineer’s use of the word. It is what you get out relative to what you put in. What you get out from the health service is outputs—activities, visits, drugs prescribed, operations performed, and so on.
Lord Turnberg: It is a broad measure.
John Appleby: It is a measure. The presumption is that being healthier is a good thing in its own right. You could extend the scope of how you measure the outputs or outcomes of the health system. It gets a bit difficult to know where to stop then.
Q55 Lord Kakkar: I want to return to the question of the way in which we fund health and social care systems and whether that has any impact on the spending required. We have received some evidence from Jennifer Dixon, referring to a report by Mark Pearson from the OECD, which suggests that if you are looking to improve the performance of your system, you should not look towards other systems but focus on driving improvement within your own system, because, quite frankly, there is very little variation of performance between the different systems and much more variation within systems.
John Appleby: I would accept that completely. There are two points here. The first is about the relationship between the source of funding and the total amount of funding. I had a quick look across the EU15 countries; you can do a scattergram of the percentage of GDP spent privately and publicly, and there is a clear relationship. Countries that spend more privately on healthcare tend to spend less publicly, so there does seem to be a trade-off there. Quite what you would take from that for policy I am not sure. I would be very cagey about then suggesting that one way of containing public spending is more private spending, for three reasons. The first is that we have to be careful about what we mean by private spending in other countries. It is not quite what private spending is here, ie out of pocket or through private medical insurance. It includes that and elements of social care spending that are bundled and called private because they come from individuals’ pockets. However, when you look at it in Holland or Germany, or wherever, most people consider this to be a tax because they have to spend this money.
The other point is how you encourage people to spend more privately. That produces a whole range of issues regarding problematic incentives, and so on. We have tried it in this country and it was abandoned. Of course there are distributional issues. If you want to switch the proportions of funding from different sources—from public to private, from collective to more individual—that raises a whole lot of distributional and equity issues. From the evidence and from looking at other countries, there is, in a sense, a trade-off between different sources of funding.
Lord Kakkar: Regarding the overall performance, is there evidence that one approach or another, or a combined approach, will deliver better performance?
John Appleby: Regarding the source of funding?
Lord Kakkar: Yes.
John Appleby: I have not seen any convincing evidence that that is the case. You can take some extreme cases such as the US; its total spending is around 17% or 18% of GDP. It spends more as a proportion of GDP publicly than the UK does, but there is a big chunk of private. They spend a lot and get some very good results in certain areas, but some very poor results in others. The source of funding is possibly part of the explanation if you were to try to describe variations in performance between health systems, but I would not attach that much importance to it. There are other more important things, such as how you organise your health service, the economic and professional incentives within the system, the regulatory model you may have, and so on. I think those are much more important in driving performance variations.
Lord Warner: Can I ask a question about targeted charging, because some systems have used that to reduce demand? I am not arguing for or against it, but what is the evidence about targeted charging, in France, for example, going to see a doctor? Is there any evidence that targeted charging for particular functions reduces demand? One of the issues that we are having to grapple with is this demand issue.
John Appleby: Yes, there is evidence that it does that. There was the famous RAND study in the 1970s or 1980s, I cannot remember the exact date, which looked at the introduction of charges and, yes, it had an effect on demand. There are other studies that show that the levels of charges did not have any effect on demand. There was a study on the abolition of prescription charges in Wales by economist David Cohen and his colleagues. One of their hypotheses was when the prescription charge was abolished there would be higher demand because the charge had been suppressing demand. As far as I can remember, nothing happened; demand did not change at all. There is that sort of study. It can be a bit of a sledgehammer to crack a nut, it seems to me, and it can have some adverse effects. The RAND study famously showed that there were people in need of care who were dissuaded from seeking care. You could say that is a price worth paying to get rid of the frivolous demand, or whatever it is, but that is a judgment and it seems to me that that is not without its potential costs.
Bishop of Carlisle: Can I ask one further thing about targeted funding and whether there is any evidence that the cost of administering these systems outweighs any potential benefit?
John Appleby: From memory, I think the prescription charge raises £300 million or so a year.
Lord Warner: It is more than that.
John Appleby: The administration of it is far less than that, so that is one example. I would add a point, though. If you take charging to see your GP—they do that in France and a number of other countries—it could change slightly the relationship between the patient and the general practitioner. It depends how you structure and organise these things. The point about France, of course, is that a lot of the money is just claimed back through insurance, so it does look like a bit of a bureaucratic chasing of money: having to pay, claiming it back, paying insurance, and so on. By and large, these things can work, but the charge has to be pretty high, and then you run into problems of interfering with people’s actual needs for healthcare.
Q56 Lord Turnberg: You may have answered this question earlier in talking about the evidence that private funding could fill a gap if public funding falls. At the moment, as a proportion of GDP, what percentage of health funding is privately funded care in the UK?
John Appleby: It was assessed to be about 1.5% of GDP, and it has been pretty flat for quite a long time now, compared to the 7.4% for publicly funded. I mentioned earlier that the OECD has new accounting methodologies for health to try to get consistent comparisons between countries. On that basis, a lot of private spending on social care is now counted as part of health. On the new figures from the OECD, just over 2% of GDP is out of pocket or through private medical insurance.
Lord Turnberg: Is that in the UK?
John Appleby: It is.
Lord Turnberg: Is there scope for more private?
John Appleby: Two per cent is about average for the EU15 countries now. It could be more. Other countries spend more. For one country—I cannot remember which—it is something like 3.5% to 4% of GDP. I think the question is how you do it. People are already free to spend privately if they want to and can afford it. You could have tax breaks or you could make it cheaper. I do not think that the experiments that we have had in this country have shown that it has been that beneficial in boosting the entire aggregate spend on healthcare. Also, as I say, it is rather skewed by who can afford to cough up the money, crudely.
The Chairman: You have mentioned previously very briefly other ways of funding—I think Lord Turnberg also referred to it—and an ex-Minister of Health recently wrote about a possible hypothecated ring-fenced tax. Do you have views about that model or any other model of funding?
John Appleby: I would suggest that hypothecation is not a good idea on balance in that it does not really solve the problems that people think it solves. First, you have to decide how you do it. One model could be to link a tax to GDP and what the country can afford. We know that there is a relationship between changes in the wealth of the country and how much countries decide to spend on healthcare. It is generally upwards, so as countries get wealthier the decision tends to be to devote a higher proportion of that increased wealth to healthcare and not to, say, potatoes or other choices in life. You could link a tax to GDP, but it could run into a number of problems. What happens when GDP goes down just when you might be wanting to spend more money? That is one problem. You still have to decide how much public money you want to put into healthcare. Somebody has to set the tax, as it were, and make these decisions. Hypothecation does not take away the broadly political public decision about how much we want to spend. I feel that sometimes people put forward hypothecation as a technical answer. You cannot escape the rather difficult decision about how much to spend and how much less to spend somewhere else regarding the opportunity costs. For those reasons, I would say that the way we decide how much to spend on healthcare is not perfect, but it is perhaps a more honest way of doing it.
Lord Willis of Knaresborough: A great deal of your work, and indeed a great deal of the analysis that we have, is about activity within the NHS. We do not have a link from that activity to successful outputs. I have looked at the States, and Magnet hospitals particularly, to see how much more effective they are in not having re-referrals into the system, which, regarding a patient’s journey, cuts down resources. Is there any evidence here that anyone is doing the work on whether that efficiency is not simply about the outcome at that point but about the sustained outcome from a particular procedure?
John Appleby: Yes. I suppose there are two answers. We have a process via NICE, which looks at the cost-effectiveness of new interventions and existing ones. It demands evidence of the new technologies that it assesses not just of whether the patient survives but of more complicated outcomes. It also tries to gather the evidence over time—if that is what you are getting at—about the benefits of the intervention. It is not just about health; there may be other organisational benefits, too. That happens once and it may make a recommendation that, “Drug X is cost effective. Okay, NHS, now you can start prescribing it”, but that is not in the sense of a continuous follow-up on the outcomes and longer-term aspects. I guess there will be some ad hoc examples in the health service, but it is not a systematic way of looking at the outcomes. This is partly Lord Turnberg’s point about the range of outcomes and over what period you get a return from this investment.
Lord Warner: Could we go back to this issue of private and public funding? Is there any evidence that the more you spend from the public purse, the more the spend on private funding reduces? The reason for my question is that when I was a Minister, the private sector used to moan about the improvements in the NHS driving down its trade. Is there any evidence that there is some correlation between what you spend publicly and what a nation ends up spending privately?
John Appleby: Yes, there is. I have looked across countries, and certainly countries that spend more publicly tend to spend less privately. That is quite a strong trend. The time trend for the UK on the proportion of public and private spend is difficult to discern, to be honest. I suspect that it is not just about money; it is about some of the achievements of the NHS. I am not sure how to put this, but one of the achievements between 2000 and now has been a reduction in waiting times. That is linked partly to extra money and partly to a focus on waiting times. Waiting times used to be a big factor in driving people to go private; they are not nearly so much now. As I say, that is linked partly to money but partly to other things, too.
The Chairman: Of the different models that you have referred to, is there any evidence to show which one delivers more efficiency, more productivity and makes it affordable?
John Appleby: Regarding sources of funding?
The Chairman: Yes.
John Appleby: I do not know that there is any evidence out there that links it in a direct sense.
The Chairman: You referred to a hypothecated tax, but does an insurance model, whether it is a single-payer insurance or multiple levels of insurance, such as the Dutch model, do better?
John Appleby: I am not aware of any evidence that directly links the source of funding with productivity or efficiency in that sense. As I say, what explains productivity and relative performance is complicated and we do not fully understand it. It is hard to get the evidence on this. Clearly there may be links between how you manage your health system and what incentives you have within the system itself, regardless of how you fund it, but they are fairly tenuous. There are probably five or 10 different factors that would explain relative performance between health systems, including their performance on productivity, but I would not lay much emphasis on the source of funding as driving that.
Q57 Lord Ribeiro: I have a question that will feed into our next session. You mentioned the French and the insurance side of it, and the fact that they can claim it back. Is there any evidence that that system has a significant impact on demand?
John Appleby: That is interesting. I do not know. I suppose it is economics 101: if you do not have a price for something, demand will be very high. I have to say, though, that I do not have much of a price for going to see the dentist and I am not clamouring to go and see a dentist just because it is cheap. The French system has some strange figures. I think the French are one of the biggest consumers of pharmaceuticals, for example, and I am not quite clear what has driven that. I suspect that culture is driving that and not so much the money aspect.
The Chairman: Is it true that the volume activity in French healthcare is no different from our volume activity?
John Appleby: Again, I do not know offhand. I would be surprised if it were radically different. I would emphasise another point regarding international comparisons. Some years ago I went to a joint conference of French and British health economists in Paris. It was in 2001, just after the WHO had come out with its ranking of health systems in the world, which put France at number one. I cannot tell you the number of French health economists who came up to us and said, “Please don’t believe that. We have real problems in our system. Look at mental health care and care of the elderly. Don’t just look at hip operations and cataracts, and so on. We have some very long waiting times, but we do not record them properly”. I was quite struck by some of the public messages and private experience in different health systems.
Q58 Baroness Blackstone: Do you have a key suggestion for change that the Committee might recommend that would support the long-term sustainability of the NHS? I know that is a difficult question, but it would be very helpful if you could answer it.
John Appleby: It does feel like the question that you get asked at the end of an interview: “If you ran the world, what would you do?” I certainly do not have an answer that the UK needs to spend 10.3% of GDP. It is clearly a choice, and what we spend is what we spend.
A broader point that I would like to see the Committee make is one that Derek Wanless made in his first report. When he came to the King’s Fund about five years later to work with me and some others to look again at his work, we made this point yet again, which is that we are relatively poor in this country at doing this sort of work, ie looking ahead and thinking not just about the next few years but the next 20 or 30 years. The only group that does it formally is the OBR, which is pretty constrained in its remit and resources, and so on. I would agree with Derek Wanless that his sort of work—not just the policy-neutral work but more expanded work, such as that done by the Congressional Budget Office and by the Centers for Medicare & Medicaid in the States—is to set out the numbers. They will change continually and medical technology will move forward, but every three to five years somebody or some organisation needs to do this sort of work and lay out the choices. It is not that there is some pre-prescribed path into the future that spending on health and social care should or will take. There will be choices to be made at any one point. Maybe there will be this breakthrough in productivity at some point that will change all the numbers again. My plea would be for somebody—it could be the OBR, appropriately resourced and with an expanded remit—to set out what the numbers are telling us about the future for health and social care.
The Chairman: We have run slightly over time, but thank you very much. You promised to send some key figures and some data and we would be very grateful for those.
