Oral evidence - Long-Term Sustainability of the NHS

Robert Chote.

Q272 The Chairman: Good afternoon, Mr Chote. Thank you very much for coming today to help with our inquiry. You are the most important witness we have had, because we would like to hear about financing the health service and social care. It would be helpful for the record if you say who you are, and if you want to say anything to start with do so, otherwise we will go on to questions.

Robert Chote: I am Robert Chote. I am chairman of the Office for Budget Responsibility. The most sensible thing, by way of introduction, would be to explain the relatively limited engagement with this topic we have as a result of the remit we have been given. As you will be aware, our primary task is to produce forecasts and projections for the outlook for the public finances for the UK, which therefore requires us to look at forecasts and projections for revenues and spending. In doing that, we are instructed by Parliament to look at projections only on the basis of current government policy, in the best way that can be defined. We are not allowed to look at different policy options, we are not allowed to give advice and we are not allowed to give recommendations.

In producing the medium-term forecast, like those we produced for the Autumn Statement, most recently, and at Budget time, areas such as health expenditure, for most of that period, are laid down in clear plans that the Government have set out in spending reviews over a three to four year time horizon. We plug those in and we make a judgment about the degree to which spending across the whole range of public services will undershoot or overshoot the limits that the Treasury has set. When it comes to producing longer-term projections, which we do currently once a year—we will be moving to doing it only once every two years—we, again, have to make assumptions about how much will be spent on health, education, transport, defence, et cetera, over that horizon. Clearly, at that horizon, the Government have not set a policy statement of what they want expenditure to be, so our legislation requires us to make a judgment and explain that as transparently as we can.

In most areas of expenditure our starting point for public services and capital spending is to assume that it remains constant as a share of GDP. One wrinkle to that over a 50-year horizon is that we adjust for demographic trends. In a sense, if the representative member of the population gets older over that period we assume that the expenditure that older people consume more of rises as a share of GDP, so for things such as health and social care you would expect to see that rise as a share of GDP as a result of the ageing population.

We are conscious in doing this that, obviously, looking back, there has been a trend in health expenditure, although not in the most recent years, for it to rise as a share of GDP, not just in the UK but in other countries as well, which casts doubt for us on whether assuming a constant share of GDP adjusted for demographics is a sensible definition or interpretation of unchanged government policy when government policy is not defined. In past reports we have looked at what would happen to health expenditure if you made an adjustment for the perceived differences in productivity growth in health and the rest of the economy. In a paper which, I think, some of you will have seen relatively recently, we have gone back to that and said, “Do we need to revisit that set of assumptions by looking more broadly at the role of demographics, non-cost pressures and the relationship between demand for healthcare and changes in income?”

I would come back to the limitations we have. We are doing this to produce a line for health expenditure to go into a forecast for the public finances. My colleagues and I make no claim to be health specialists. It is not our job to say what should be spent and it is not our job to say whether what is being spent is being well spent; it is to come up with some relatively transparent assumptions about what unchanged government policy could imply, given the pressures on healthcare, and to forecast the public finances accordingly.

The Chairman: That is very helpful. I hope most of our questions are related to the finance side and not the health side. I will kick off first with your latest analysis of public spending on health. This Committee is looking longer term; it is looking at 10 to 15 years from now, 2025-2030 and beyond. On that basis, what are the key factors that you think will drive those pressures? What will the pressures be? What impact will those funding pressures have on the long-term fiscal sustainability of the NHS? What possible figures might we be talking about?

Robert Chote: I will deal with the second part of your question first. I think it is important to make the point that one has to be careful in saying that any given category of expenditure is, in its own right, fiscally unsustainable or sustainable. Clearly, if there is an area of expenditure which, over time, is going to rise inexorably to 100% of GDP, wiping out everything else you could possible spend, that is clearly unsustainable. However, with smaller increases, the judgment about sustainability is one for the public finances as a whole. If you want to spend more on health as a share of GDP your choices are to borrow more, to spend less on something else or to tax more to do that. In a sense, sustainability has to be thought of in the constellation of decisions you want to make about other elements of tax policy and spending policy rather than saying that any particular element of it is, by definition, unsustainable or sustainable because it is rising or falling as a share of GDP.

In terms of the drivers, there are three main elements that most of the studies, internationally and in the UK, have looked at, one of which is the relationship between expenditure, or the demand for healthcare, and income. Do you expect, as people’s incomes rise, that their desired expenditure, or the amount that society will want or end up spending on health, will rise more or less quickly than that? There is an argument that says that as societies get richer not only will they want to spend a larger amount of cash on health but they will also want to spend a larger proportion of national income on that. Clearly, if you assume that ad infinitum it would rise up and up and would be unsustainable. The assumption we have made, which is not an untypical one—others have made different choices, higher or lower—is to assume that, basically, it will rise in line with income. There is no specific pressure, either upward or downward, on health expenditure as a share of GDP simply as a result of societies getting wealthier. That is one issue.

The second issue, as I say, is a demographic one. Health expenditure tends to be higher on people who are older. That is not the whole story. Obviously, quite a lot of health expenditure is higher proximate to death, which will show up as a relationship with age—but there is more to it than that. In the past we have focused on the demographic pressures. It struck us, looking at other studies that people have done in different countries, that there seems to be a growing belief that pure demographic effects are not, perhaps, the major driver of the increases in healthcare spending as a share of GDP that one has seen, and therefore that it would be a source of upward pressure but not necessarily the main source of upward pressure going forward.

We base our projections on projections of the size and structure of the population done by the Office for National Statistics. We have to choose, among those, for example, to have a particular measure of net inward migration that seems sensible, given policy and economic developments, but those are implying, over the 50-year horizon that we look at, a significant ageing of the population. The proportion of 85 year-olds and above was about 0.8% in 1970 and is about 2.4% now; it will be 7.3% in 2070. The ONS projections assume that the number of centenarians is around 14,000 now but will be 450,000 in 2070. There is obviously a lot of uncertainty around these things but that is one driver element.

The third one is a catch-all term for “other cost pressures”, which are partly things such as new techniques, new drugs and new treatments that become available that may be more expensive. There is also the paradox that some treatments that are cheaper can end up putting upward pressure on healthcare spending because they can be used more widely. A good example of that is the use of stents as an alternative to heart surgery. Again, I am going well on to thin ice in terms of my medical knowledge here. As I understand it, around 30% of the use of stents is on people who would otherwise have much more expensive heart surgery, but 70% of it is being used to treat people who would not have received heart surgery and, therefore, that is still putting upward pressure on spending even though stents are a lot less expensive than cracking someone’s chest open. You have those issues.

A related cost pressure which can show up in the numbers is the perception that healthcare, and perhaps, in particular, social care, is more labour intensive and therefore less able to deliver improvements in productivity—the amount of output per hour worked—than in the rest of the economy. Having said that, the rest of the economy is not doing terribly well in delivering productivity growth at the moment. Leaving that to one side, you have estimates that hospital budgets are about 70% labour cost, and that has been rising. If you assume that productivity growth in healthcare, on most studies, rises by about 1% a year, we have in the past been used to productivity growth in the rest of the economy of about 2% to 2.5% a year. If you therefore want the output of healthcare to rise in line with the output of the rest of the economy, you have to make good that productivity gap and spend continuously more. That will be another element of the cost pressure story.

The relative importance of these appears to differ quite a lot from study to study. Some have the share of other cost pressures as low as 25%, and some have it as high as 75% of the explanation. I suspect that the numbers you end up with may be, in part, a function of which of these you try to estimate first and, therefore, which one is left to soak up all the things you have not managed to explain with the other two assumptions. How much weight you should put on the precise relative importance of those would be doubtful, but the conventional wisdom is that the major source of upward pressure is the other cost pressures and less, perhaps, the demographics that people would have focused on earlier.

The Chairman: Thank you very much.

Q273 Lord Kakkar: Lord Chairman, thank you very much. I would like to further explore how the Government might accommodate a growth in health spending because of these upward pressures beyond the assumptions of an increase in wealth in society. How might this growth in expenditure be used to provide for health? Is it reasonable to consider more and more cuts to other areas of public spending to accommodate increased requirements and expectations in health? Or do we need to consider raising taxes or, indeed, as you have mentioned, borrowing much more to support health?

Robert Chote: As I say, those are the options and it is not for us to say which of those should be pursued. If you look at the way in which the budget deficit as it stands now is being dealt with on existing plans, cuts to public expenditure as a share of GDP and, in particular, cuts in current spending, as distinct from capital spending, so the day-to-day running costs of public services and welfare spending in the form of cash transfers, are delivering quite a lot of the, roughly, 3% of GDP improvement in the structural budget deficit projected over the next few years. A lot of that is coming out of other relatively unprotected public services. As you know, if you look at what has been happening on health and you look at what the current plans imply over the next four years, there is an increase in real spending, so spending is going up faster than inflation across the economy. However, that is not keeping pace with the growth in population. Real spending per capita will shrink by about 1% on current plans over the next four years, and there will be a further fall in expenditure as a share of GDP. On our estimates you would be at about 6.8% of GDP spent on health by 2019-20, which is down from about 7.6% at its peak in the wake of the financial crisis.

You are also seeing welfare spending—cash transfers—being cut in generosity, in particular. That is focused very much on things such as incapacity and disability benefits and tax credits. The state pension is conspicuously less affected by that and is, in a sense, an upward pressure, so that is an area that remains relatively protected. The other choice, of course, would be to raise more in tax. The Government have chosen, at the Autumn Statement, to have a less ambitious programme of deficit reduction than they inherited from the Cameron-Osborne Government regime that preceded them. There has been a shift there. The Government have used up some of that but have put the money into additional capital spending plans. We have soaked some of it up by having a more pessimistic forecast for the outlook for the public finances as well. You do, in the end, come back to a choice, and Governments historically have been reluctant to go for a significant rise in the tax to GDP ratio. So far it has been cuts in other elements. How far that goes on, as I say, is a constellation of choices. It seems to me, in thinking about what one should be telling the healthcare sector are its prospects for the 10 or 20 years coming up, you cannot do that in isolation; you have to say, “What does this imply for what we might want to spend on defence or transport?”—or whatever else it might be.

Lord Kakkar: For how long is the current trajectory we are on, the 7.6% to 6.8% by 2019, sustainable in terms of what you see as the broader determinants of demand in this particular part of the economy?

Robert Chote: As I say, over that sort of time horizon this is ultimately a political choice. If you want a greater or lesser quality and quantity of publicly provided healthcare than is implied by that, you have to make your own choices. Clearly, healthcare is a squeaky wheel that can be relatively confident of being greased relative to some other areas of public services. That would be the historical record.

Lord Kakkar: I am now on very tenuous ground as I know nothing about economics and productivity, but I ask this question: how much of this shrinking proportion of GDP might be met by an increase in productivity to bring productivity of the healthcare workforce closer to general productivity?

Robert Chote: Measuring output in the health sector is not straightforward. In a relatively broad-brush way it can be done as the number of interventions, roughly speaking, which obviously begs a whole series of questions about value for money and whether you are reflecting quality in that or not, et cetera. As I say, historically speaking, recent studies suggest productivity growth in healthcare of about 1% a year. Over a longer time horizon we have been used to the rest of the economy delivering 2% to 2.5%, although it has not been delivering anything like that recently. It may be that rather than healthcare improving its productivity performance to match the rest of the economy, the rest of the economy is worsening in order to match that of health.

Lord Kakkar: Another question that has been put is whether there is any evidence that spending on healthcare has an overall impact on broader economic growth as a potential justification for this being an area that, in terms of sustainability of the NHS, should be more targeted and more protected.

Robert Chote: It is not an aggregate relationship that we have explored or attempted to explore. We do not look at what the Government are intending to spend on health and then take a judgment from that back into what we are assuming about the underlying growth potential of the economy—although I am sure some people must have done studies that do that. There are obviously elements of the delivery and effectiveness of healthcare, broadly defined, that would have an impact, not least the age at which it is realistic to expect people to continue to work. Do the improvements in life expectancy, and the changes in morbidity that come out of that, mean you are extending people’s effective working lives by the same degree? We are struck, looking, interestingly, within the welfare budget, by what the Department for Work and Pensions is having to spend on things such as disability living allowance, and by the degree to which mental illness is an important factor. Part of it is, obviously, dementia for relatively old people, but it is increasingly striking younger people as well, and that would have a feed-through to economic performance and the health and material well-being of the individuals concerned.

Lord Kakkar: One last, very short question on hypothecated tax for healthcare expenditure. What are the pros and cons of that, in your opinion?

Robert Chote: When I was running the Institute for Fiscal Studies I would have given you chapter and verse on this, which is now outside my remit. If I was still at the Institute for Fiscal Studies, I suspect that they would, in the great Treasury tradition, broadly defined, be wary of hypothecation, rhetorical or mechanical, on the grounds that money is fungible. When you link expenditure to particular receipt streams, when that receipt stream starts to dry up for some unexpected reason, do you want to reduce the amount of spending in that area? It is easy to set these things up; it is harder to deal with those things when they come out. As I say, I could not possibly comment on that subject. I suspect, if you asked my IFS colleagues, they would say something similar.

Lord Warner: The coalition Government made some rather bold claims about driving up productivity, but on closer inspection it looked as though the driving up of productivity was more to do with pay restraint than anything else. What assumptions do you make about wages and pay in these labour-intensive industries in terms of productivity over the longer term?

Robert Chote: For the projections we make, we are obviously looking at the wage and productivity picture economy-wide rather than building it up sector by sector. One point which is worth bearing in mind is that in the efforts that have been made to measure productivity in healthcare—and the Office for National Statistics has had a programme of work over this period for some time—there is an inevitable link between measured productivity and the swings of relative boom and bust in the generosity of spending at any given moment. Disentangling that cyclical effect from any long-term underlying improvement is not straightforward.

When you saw public spending in public services growing rapidly in the Brown chancellorship era, somewhat inevitably quite a lot of the productivity numbers did not look as good because you were shovelling in a lot more input and getting relatively less output. The argument could be made, “Well, that is an inevitable short-term consequence”. What you hope it will do is deliver for you a longer-term improvement that will show up later on, but by that stage the spending cycle may have moved in the other direction and you can sometimes get the productivity numbers to look quite good by pulling input out, and the system manages to keep going—as Wile E Coyote manages to keep walking as he goes over the edge of the cliff before falling down. I am not an expert on the way in which that is measured, but I think you have to be wary of the short-term interactions when you are shovelling more money in, be it to wages or other elements of input, and the timescale by which that leads to underlying changes in productivity.

Lord Turnberg: Can you help me with this productivity business and how we measure it? It seems to me that if we cure someone of a disease with some expensive treatment, they go back to work and they are productive and pay taxes for longer, and that will accrue to the Treasury but not to the Department of Health. Does that get into the productivity measurement?

Robert Chote: As I understand it, the productivity measurement tends to be a more mechanical number of treatments of a particular type. Over time you can do it in a more sophisticated way by chopping it up and trying to reflect the quality of those things in a more granular fashion. It is not quite as stark as in education where it is the number of pupils taught. If you make cuts in education and implement them in a way that does not involve not educating anyone whose name ends after M in the alphabet, you get a perverse effect on measured productivity. I do not think it is as extreme in healthcare. It is the number of hips replaced, not the number of days of useful working life restored.

Lord Turnberg: That is a shame, is it not? Productivity should be in terms of recouped costs to the community at large, not just to the health service.

Robert Chote: We are talking here about the way in which these things are measured for national accounts. One would hope that in the corridors of power decisions are made on a richer evidence base.

Lord Turnberg: Thank you—wherever they are.

Q274 The Chairman: You mentioned the “boom and bust” phase and that is how currently the NHS is financed; there are years of plenty and then there are years of not plenty. The trajectory seems to be going down now, ending up at 6.8% in 2019, as you mentioned. What is your suggestion for how to get rid of that, and what would be the effect if you had a level playing field and increasing inflation?

Robert Chote: As I say, the commitment of relative protection, as I understand it, has amounted to, “We will ensure that real spending goes up”, and there are a lot of sectors in public services where that is not the case. If people in health think they are hard done by, look at parts of the criminal justice system where the squeeze will be considerably tighter. As I say, protection is a relative term. You have spending rising in real terms, projected over the next four years, but that does not keep pace with the population, and as a share of the economy it would fall—but if you were not taking roughly 1% of GDP out of healthcare over this period you would have to be taking it out of somewhere else, taxing more or borrowing more than you otherwise would have done.

Clearly, I presume, there is a strategic choice that the Government, as a whole, have to make, which is to say, “Do we want to approach this in a top-down fashion or a bottom-up fashion?” The top-down fashion would be, given what we think people are willing to pay in tax and what we can get out of them, logistically, and given all the variety of other challenges that we have to address in areas such as education and defence, et cetera, you come up with a view of what health ought to be getting as a share of GDP. GDP can move around a lot, so that is quite an awkward thing to plan for, and obviously over time that normally translates into cash settlements. Do you then tell the NHS, “That’s your money, plan within it. Don’t come and tell us that you’ve suddenly discovered that winters are cold. That’s the budget you’ve been given”? Or do you do it on the basis of, “Health needs this”—and each year they will come back and say it needs a bit more and then you have to go away and make decisions about what is elsewhere?

In part, I presume it is a decision about how clear a top-down signal you send the sector. Do you say, “Like it or not, that is the share of GDP you’ve got, given all the other challenges we have; deliver the best healthcare system you can within that”? Or do you say, “This looks a particularly bad winter, there are pressures here, there and everywhere”? That is well above my pay grade. As I say, it comes back to this point that, even with fiscal sustainability, you can look at demands and cost pressures—but there are different sorts of pressures in all sorts of other parts of the public sector and choices to be made about how much people are willing and able to pay in tax. At some level you have to make big decisions that bring all those things together; not just looking at the drivers in each particular sector.

The Chairman: Lord Warner, I pinched half your question.

Lord Warner: Can I pursue the impact of boom and bust on forecasting? Some of us have had a look at what has happened over the last 25 years to health and social care expenditure. It has been extremely volatile, on both of them, year on year, with spikes all over the place. There has also been no synchronisation, even though the demography is pretty similar, between what has been allocated year on year to social care compared with health. Indeed, in some years social care has done better than health, so it has been a rollercoaster for both these services. Given that that has happened over the last 25 years, how does that rollercoaster approach to funding these services affect your forecasts over the next 25, 30 or 40 years? Would smoothing those things out change your forecast of future expenditure? Is it likely to?

Robert Chote: Certainly over the much longer time horizon we are, effectively, smoothing by making broad-brush judgments about the share of GDP. To date there has been some adjustment purely for the demographic effect and presenting a representative implication of weaker productivity growth. As I say, we explore in this paper whether we should take a somewhat less pessimistic view of the pressures from morbidity and a somewhat more pessimistic view of other cost pressures.

The implications for our forecast are that over the five-year horizon we produce the forecast on the basis of government policy as it is stated. The Treasury sets what are called departmental expenditure limits, or DELs, and we look at the overall size of that DEL envelope for public services as a whole, informed by whether there is obvious evidence of particular pressures in particular sectors. Relative to other countries, the Treasury is powerful, relative to spending departments, in the management of public expenditure. We do not produce a bottom-up forecast department by department because we know that the Treasury has a sufficiently firm grip at the top so that it is very rare that the aggregate envelope for all these services is overspent, even if they have to take some money out of one and stick it into another.

One element of those departmental expenditure limits is the grants that the Government provide to local authorities. Local authorities also raise their own money and some local authorities are more exposed to social care costs than others. That is an area of complexity that goes beyond the way in which we look at it. Of all the many difficult things we have to forecast, forecasting local authority expenditure in aggregate is not a straightforward one. We had assumed for years, as the cuts were tightening, that local authorities were going to start running down their reserves in order to keep expenditure going. What has, in fact, happened is that they have ended up, perhaps until very recently, shovelling more money into reserves. Tony Travers at the LSE famously says that local authority finance officers share genetic material with squirrels; when confronted by any uncertainty their instinct is to put the hazelnuts away for a rainy day. It has certainly been a source of error in our forecasts.

As I say, we do not go into a sufficiently disaggregated look. There are clearly the local authority experts we would talk to, to say that there are different pressures facing local authorities for whom social care is a significant part of their budget versus those for whom it is less so. The fact that the Government are explicitly allowing authorities to raise more council tax to address some of this is obviously a recognition of that. I read, as others do, about the potential benefits of thinking about social care and health expenditure more holistically, which I am sure is a very important topic but is not one that is core to the forecasting job we do.

Q275 The Chairman: The OBR does a superb job on the independent analysis of spending in health and gives an important insight into the sustainability of healthcare funding. Given the pressures on the health and social care systems, do you think further independent analysis and oversight of the funding, the workforce and impact on the healthcare system based on demography and medical advances would be helpful? If so, who might do this and how could it be implemented?

Robert Chote: The fact that you are all taking an interest in this with this inquiry indicates there is a level of knowledge. I do not think, certainly relative to sectors such as education, that health is particularly under-think-tanked. For a lot of people, in some cases, the sound of a grinding axe can be heard in the distance, and in some cases not. In a sense, some of the most interesting work is done by people such as the OECD who are able to look across countries and pick up things you would not necessarily look at, or interesting patterns that you would not get confined to analysis there. There have been exercises in the past—including the Wanless review—looking at definitions of particular need and how that could be done, and I am sure there would be many researchers who would very happily accept funding for doing this sort of analysis.

The Chairman: The difference is that you were set up by the then Chancellor to be an independent body to give this independent advice. The think tanks do a good job but they are not listened to because they are not set up by the Government. If we had an independent body that does in-depth analysis that you do not or cannot do, and then says, “Based on these findings we think this is what the funding settlement should be; this is what the workforce should be”, et cetera, would that not be better?

Robert Chote: Again, I am not close enough to the sector to know. As I understand it, Wanless was relatively highly regarded as a proper piece of work. Presumably, that could have been given an ongoing life in the way in which we have mechanisms, for example, for the periodic review of the minimum wage or the state pension age. You do not necessarily have to have a large organisation that does it all itself; you could have a body such as the Low Pay Commission. Their working model is that they are able to commission research but come back to it and review in on a regular basis, and obviously they have a formal role in putting in commitments.

We now have the National Infrastructure Commission, which is supposed to be coming up with an overall approach to infrastructure spending, although I think it was the Chancellor’s announcement in the Autumn Statement that there is a top-down approach there of saying, “This is the amount of money you have to play with; within the limits, go off and think about how that ought to be done”. If you were setting up a body in health in this area, again, you have that choice between saying, “Do you want them to go away and work out what we need?” or do you want to say, “Health can have 9% of GDP to spend in 20 years’ time. What can you deliver for that?” It could be approached in either or, indeed, both of those ways, if you wanted to. I would have thought models such as the Low Pay Commission or the National Infrastructure Commission would be possible ways of going at this. The fact that Wanless did what it did showed that a one-off exercise in that sector could get quite wide appreciation and purchase.

The Chairman: In the Autumn Statement the Chancellor said this about you: “Let me turn now to the forecast. Since 2010 the Office for Budget Responsibility has provided an independent economic and fiscal forecast to which the Government must respond. Gone are the days when the Chancellor could mark his own homework, and I thank Robert Chote and his team for their hard work”. My question, therefore, is: should a body be established to ensure that the Health Secretary does not mark his own homework?

Robert Chote: Again, you are taking me into areas of sectoral institutions. Does Monitor have any role in the performance element? Is this about whether the Health Secretary or the health department are delivering what they can with the money they have been given, or is it something you are tasking us with, saying, “How much money should these people be given”? Are you doing the top-down or the bottom-up? The advantage we have is that while we have a broad job our remit is very tightly defined. It is a great help that I can say, “I’m awfully sorry; I’d love to give policy advice or to shoot my mouth off over whether this is a silly or sensible reform to stamp duty or capital gains tax but that’s not my job.” There are other people at the Institute for Fiscal Studies, for example, who can do that. Having a deep but narrow remit makes our life a lot easier because it is just as important to be able to say, “That’s not a question I can answer”, as opposed to—pleasurable as it is—turning up here and offering my unfocused, uninformed thoughts on a whole variety of things. That is one of the choices you have to make about setting up this body, if you want to go down that route. Can you give it a mandate that does not leave it having to produce airy-fairy reports about, “Wouldn’t it be lovely if we had 15% of GDP to spend on healthcare”? Wouldn’t it—but there are other things to spend it on elsewhere.

Lord Warner: Can you say a little bit about how long it took you to get yourselves set up to do this, be effective and get accepted? Can you give us a feel for that?

Robert Chote: We were set up in 2010. The major advantage that we had was that, effectively, the Government chose to outsource a function that already existed in the Treasury. When we were created, the bulk of our staff were simply the people who were responsible for doing the main macroeconomic forecasts and the main revenue and spending forecasts in the Treasury who were, not quite physically, lifted out of the organisation, put into a quango, had myself and two deputies, like the figures on a cake, stuck on top and charged with making the decision. I inherited people who had been doing this work, mechanically, for many years and with established relationships between us and, in particular, Revenue and Customs and the Department for Work and Pensions that had been there for ages. We were able to get up and running very quickly. We have expanded in number, we have expanded in the remit somewhat, but the core bit of the machine was taking out a bit of the Treasury, moving it into a quango and bolstering its independence and transparency—and away we went.

If you are creating something new and trying to establish its reputation—which is more, for example, what is happening with the National Infrastructure Commission—there is a whole series of choices about the model you want. We have a large staff/small council model; the infrastructure commission has more commissioners and staff but the ratio is different. It was easier for us because the Government was outsourcing a task that they had already undertaken rather than creating a new one or duplicating an existing function.

Lord Warner: That is very helpful, thank you.

Q276 Lord Willis of Knaresborough: Can I come back to your initial remit about forecasts and projections on current government policy? That is what you said as you started. Could you tell us, for interest, what you regard as the current Government’s policy on the social care settlement? In fiscal terms, what is the settlement? What do you understand to be the settlement? My second question is going to be: can you project that through to 2020 and beyond?

Robert Chote: On social care, as with healthcare more broadly, over the longer time horizon you are looking at a constant share of GDP adjusted for demography. For social care there is a more sophisticated analysis that can be done, but on the amounts of money involved the scale of difference is that much less.

Lord Willis of Knaresborough: You are saying that it is linked to GDP and demography? The figures do not seem to indicate that because it is going in the opposite direction.

Robert Chote: I can give you more detail on that. Obviously, I was focused on health for the purposes of the invitation.

Lord Willis of Knaresborough: It is tied in to the same thing, is it not?

Robert Chote: Yes. The greater complexity there, as well, is what has been agreed in terms of local authorities’ spending and their ability to raise revenue.

Lord Willis of Knaresborough: Can I try and help you here? I deliberately went back to forecasts and projections based on current government policy. Current government policy is, in fact, to integrate health and social care, and to do it from now, not in five or ten years’ time. Are you saying you do not do any work at all on projecting what that overall cost of health and social care will be?

Robert Chote: We do the projections on the basis of the Government’s stated decisions about the amount they are spending in particular elements of departmental expenditure limits, some of which will include estimates of the grants that they provide to local authorities.

Lord Willis of Knaresborough: The answer is no, is it not? The answer is no, you are not.

Robert Chote: I am sorry?

Lord Willis of Knaresborough: The answer is no. You are not aggregating those two on the basis of current government policy to project expenditure.

Robert Chote: Explicitly a combined health and social care budget?

Lord Willis of Knaresborough: Yes.

Robert Chote: No.

Lord Willis of Knaresborough: Do you not think that is a mistake and perhaps something you should be doing without having government Ministers asking you to do it, as an independent body?

Robert Chote: At the end of the day, remember, the job we have been given by Parliament is to get an estimate of what this implies for the public finances, hence the aim of looking at the total expenditure. From our point of view it would be irrelevant whether you spent what is defined as social care versus health.

Lord Willis of Knaresborough: I understand all that. What I am trying to get you to say is that there is a job to be done by your organisation to look at current government policy and its projection in terms of finance, linking together government policy on combined health and social care.

Robert Chote: Whether that would get you to a better answer on the aggregate expenditure, I would be very interested to see. If you think there are particular ways in which we could slice this cake that would be more effective, I would be very happy to look at that.

Lord Willis of Knaresborough: The cake, at the moment, is being sliced, as you have said, quite rightly, so that there is a notional increase in spending on health, as agreed with the NHS, up until 2020 and the five-year plan. There is not the same settlement for social care, despite the fact that the two are interlinked in the form of government policy. All I am trying to get you to say is, “Yes, Lord Willis, this is a deficiency which I will take away and start working on tomorrow”.

Robert Chote: I am not going to say that. As I say, we are producing the different elements of what the Government are spending both directly in central government and the allocations they make to local authorities. We make judgments on the ability of local authorities, for example, to raise their own revenue and then we have to make judgments about the degree to which they spend the money they raise or whether they add some of it to reserves or subtract some of it. At the end, we get to an overall set of forecasts for revenue and expenditure which drives our public finance forecast.

As I say, I am very happy to look at whether we are picking up the financial consequences of the various flows implied by statements of government policy as effectively as we can.

Lord Willis of Knaresborough: I think this Committee would be very interested if, in fact, that were the case. A simpler question to finish my section with is: Dilnot is current government policy. Have you factored that in to future spending requirements for social care, to your knowledge?

Robert Chote: To be honest, I cannot remember the current timing on Dilnot. It is not going to show up very much in the course of the five-year forecast that we are producing, for which plans are set out. We have responded as the timetable for that has moved out.

Lord Willis of Knaresborough: You did also say in your opening remarks that, in addition to the five years, you were forecasting beyond that, based on government policy.

Robert Chote: We do projections beyond that, yes.

Lord Willis of Knaresborough: Dilnot is not part of that?

Robert Chote: I think at that stage it is part of that, but it does not show up much in the five-year forecast.

Lord Lipsey: I would like your view as an economist as well as the head of the OBR. We have heard a lot of arguments in this Committee which say, spend a bit more on public health and people will give up diabetes and save the health service £4 billion a year, and they will stop dying of smoking and save the health service £8 billion a year—I am sorry, those are imaginary figures. Members of the Committee are slightly concerned about, “If I do not die from smoking today, am I not going to die, perhaps even more expensively, from dementia in five years’ time?” Have you looked at those kinds of putative savings at all? How do you think the logic of that pans out?

Robert Chote: We have looked at some of the studies, in particular, about the issues of morbidity, and the degree to which the extension in life expectancy results in an increase in healthy life expectancy and how much not. Some of the work that we cite, looking at the other cost pressures, would note, for example, the chronic diseases that you are likely to see, partly as a result of ageing but not just as a result of ageing, as chronic diseases become more important as a source of illness. Then there is the issue of co-morbidity. If that is a greater issue then the chances of other potentially expensive, complicated things happening at the same time is greater as well. They are subsumed in different ways into parts of the analyses people do on other cost pressures. Dementia, obviously, stands out as a particular issue. As I say, there is a broader mental health issue that is showing up in a way that we have not—and people had not—anticipated in the welfare budget, and at younger ages as well as older ones.

Lord Turnberg: If we are going to cure Alzheimer’s, how do you calculate that?

Robert Chote: I do not think we will be making explicit adjustment for that. The question then is: if you cure that, presumably, what else are people going to have instead, and is it going to be cheaper or more expensive over the longer term? If you are not killing people off with something else more quickly and they are ill for longer in such a way that is expensive but yet has an overall improvement in quality-adjusted life years, what choice does society make about whether it wants to spend a lot of money in that sort of area?

Lord Warner: Can I take us back to make sure I have understood your answers to Lord Willis? I do not want to go too much into the detail. As I understood what you were saying, the problem for you in forecasting ahead for social care is the fact that finding out precisely what the local authorities are doing on their bit of the budget is extremely difficult because you have to make a set of assumptions about what they may or may not spend. Is that a correct interpretation of what you are saying?

Robert Chote: That is the overall issue in terms of our forecast for local authority spending as a whole. Again, this is not an area where we are producing bottom-up forecasts authority by authority; we are looking at the overall amount of money, essentially, that local authorities are given in grant. Then there is the amount of money that local authorities raise for themselves and then there is a judgment about the degree to which they add or subtract from their reserves to do that. We are conscious that, in talking to local authority experts about the anecdotal evidence, you might hear, “Oh, there are particular pressures in this sector and these are an important driver”, and there may then be explicit permissions for authorities to raise more council tax to spend more money in these particular areas. That can be a factor.

Lord Warner: We only find out at the end of each year. Effectively, we only find out at the end of the process, do we not?

Robert Chote: The issue for both the NHS and healthcare, at central government level, and for local authorities, is that it takes some considerable months before we know exactly what local authorities have spent and exactly what the NHS has spent. That can be some months after the end of the fiscal year.

Lord Warner: That is very clear.

Robert Chote: I would have boned up more on the social care forecast if I had known you were going in that direction.

The Chairman: Based on your analysis, are you able to say what you think the share of GDP would have to be in 2030-31 for health?

Robert Chote: We tend to look over a 50-year horizon. You could probably chop it off at the relevant bit and divide the number accordingly. The most recently updated set of long-term forecasts we did had health spending rising to 8.8% of GDP by 2060-65, compared to 6.8% at the end of the medium-term forecast for 2019-20. That was a bit higher than we had forecast previously, partly because since we did the previous set of long-term projections the Government have decided to spend more in total on public services and more on health at the starting point, so there is more healthcare for the demographic pressures to bite on over the subsequent years, and there have been changes in the population projections.

In addition, if you look at our standard, “Let’s adjust this for lower productivity growth and let us assume that you spend sufficiently more on health to deliver the same increase in output year in, year out, as you get in the rest of the economy”, that would get you up to about 13.5% by the end—an additional 5% or so. A couple of projections that we look at in this paper would be, “Let’s assume that the other, i.e. non-demographic, costs are rising at the sort of rate we have seen recently and they continue to do so throughout that 50-year horizon”. That would get you up to something over 18% of GDP. If you make what most forecasters in this area would regard as, perhaps, a more sensible judgment, as the Congressional Budget Office does in the United States, of saying, “Let’s assume that this additional cost pressure moves to about 1% a year rather than remaining as high as it has been more recently”, that gets you to about 15% of GDP.

The other area where our forecast looks relatively pessimistic is on the assumptions we make about morbidity. Most other projections will be more optimistic on that, so you could take, maybe, 0.5% to 1% of GDP off that. You can see that there are very wide variations between these, depending on exactly what sets of assumptions you make about them.

Lord Turnberg: That is health and not social care?

Robert Chote: That is health. With social care, as I recall it, again, you obviously have a demographic pressure, but the starting number is much smaller, so its impact on the public finances is much less important. Proportionately. I suspect it may be even larger than it is for health but the fiscal consequences of it are much less.

Q277 Baroness Redfern: Local authorities can raise social care to 2% each year, as such. It is disappointing there is less on social care. My question, Robert, is: what is your key suggestion for a change this Committee could recommend which would support the long-term sustainability of the NHS?

Robert Chote: As I say, it is you, collectively, as in Parliament, who told us not to offer policy suggestions. This is an issue of striking the right balance between thinking about and planning healthcare spending on a bottom-up basis and living in a world in which the NHS can come back, the wheel can squeak and more grease can be applied as and when asked for, versus a more top-down assessment of the choices about healthcare spending along with a whole lot of other choices about what we want our criminal justice system to look like, what we think our defence needs and diplomatic needs are going to be, et cetera, and to provide a more overarching sense of the share of GDP that health should reasonably be thinking of. We should not be ignoring that element of it to the degree that as a complete amateur and outsider one gets the sense of health being thought of in isolation—not in isolation from social care but from the rest of it. My suggestion would be thinking about what longer-term signals you want to be sending to the sector—not just because of a variety of exciting studies on drug price pressures but what are we going to want to spend on defence in 15 to 20 years’ time?

The Chairman: Thank you very much, indeed, Robert, for coming today. It has been most helpful. We appreciate what a busy person you are.

Robert Chote: Thank you very much for the invitation. I hope it has been useful.

The Chairman: It has been extremely useful. You might want to know that the Government are to reveal a Brexit plan before the EU exit begins.

Robert Chote: Very good. The pleasure of an hour not talking about Brexit is not to be underestimated.

The Chairman: That is why I told you, because you told me you did not want to talk about Brexit.