23
Examination of witnesses
Laura Gardiner, Tom Josephs and Professor David Miles.
Q1 The Chair: Welcome to the Lords Economic Affairs Committee’s one-off session on the publication of the Office for Budget Responsibility’s Fiscal risks and sustainability report. We recommended in our last report on the fiscal framework that this very important document get more publicity, so we thought we would lead by example and get it more publicity by inviting the authors back in to discuss various elements of it.
We are really grateful to the three of you: Professor David Miles, a member of the Budget Responsibility Committee of the OBR, Tom Josephs, also a member of the Budget Responsibility Committee, and Laura Gardiner, chief of staff at the OBR. We are screening this live on parliamentlive.tv and a full transcript will be taken. We will circulate that to you so that you can make any factual corrections afterwards. Thank you again for your time. We are very grateful.
Can I start with a general question on spending and the reform of spending? Which spending areas—there are large substantial areas, such as the National Health Service, pensions, social care, et cetera—would you highlight where reform could have the most pronounced effect on the long-term trajectory of spending and the public finances?
To put this question another way, which areas pose the greatest risks? Which areas do you have most concerns about in the medium to long-term trajectory, where the Government have more credible control over the levers of controlling that spending? David, I will start with you.
Professor David Miles: Thank you for giving us the opportunity to do something that you were keen that we do, which is to help to publicise the Fiscal risks and sustainability report. We are very grateful that you have found time to see us quite soon after we published it.
If I could ask a related question, where are the spending pressures that we have identified that generate the largest increase in overall spending over the next few decades? It is probably no surprise to say that they are in the areas of health and pensions, or welfare support for the elderly, which account between them in roughly equal measure for pretty much the whole of the projected increase in government spending of about 10 percentage points of GDP. That is on an assumption of what we call unchanged policies.
Unchanged policies on the tax side tend to generate—it depends a bit on which scenario you look at—a profile for tax revenues that is relatively flat. This opens up a large gap the further ahead you look and is behind the trajectories for debt-to-GDP, which on pretty much all the scenarios, some of which are more optimistic than others on productivity or labour market outcomes, et cetera, generate trajectories that are ultimately unsustainable and ever increasing in profile.
Perhaps it is in the areas of pensions and health, neither of which are easy, where one might argue that one needs to do most. On the pensions side, we calculate how costly it is to stick to the triple lock over the course of the next several decades. It is both significant and rising in terms of the difference in spending on pensions under the triple lock. If you tie state pensions to average earnings, I think I am right in saying—my colleagues will correct me—that a few decades from now it adds about 1.5% and sometimes even nearer to 2% to government spending over a 50-year horizon. One can appreciate that 2% on government spending every year for several decades is a non-trivial part of the very big increase in debt-to-GDP that we get.
On the health side, it is a bit more difficult to identify a specific policy that you can cost to make particular savings. The increase in health costs relative to other costs in the economy has been running at a quite substantial level for some decades, partly because productivity in a very service-heavy sector of the economy such as health tends to fall behind productivity in the economy in general. If, potentially as a result of AI, that could be turned around and productivity in the health sector was at least to match that in the rest of the economy, that is another area where you would, over decades, generate really quite substantial savings.
I do not know, Tom and Laura, whether you want to elaborate on any of that.
Tom Josephs: First of all, I would echo David’s thanks to the committee for the opportunity to talk about the report. Thank you for highlighting it in your report. We see it as one of the most important reports that we do. We have been looking more widely for ways in which we can raise its profile and promote it.
I have one thing to add to David’s overview. In addition to health and pensions, the increase and the pressure on those from demographics and this wider cost pressure on health, the new pressure on spending that we have identified this time is defence. We assume in our projections that the Government meet their commitment to get to 3.5% of GDP by 2035. That is quite a significant new pressure on top of the ones that we have known about for a while. It contributes to a quite material increase in the upward trajectory of debt in the projections.
Of course, there is quite a lot of risk around that number. The NATO target is 3.5% of GDP, but, unfortunately, there are clearly scenarios where you could envisage defence spending needing to increase more than that. That is the new pressure that we have identified in this report.
The Chair: I am interested in the relationship between spending, growth and debt. One way of thinking about that is, “If spending trajectories are as you envisage over the next few years and are not changed, what rates of growth will be needed in order for the debt ratio to remain stable over the long term?” Do you have a ballpark figure for that?
Professor David Miles: We show one scenario in which productivity growth is substantially higher—not just higher than it has been over the last 15 or 20 years, when it has been dismally low with barely 0.5% productivity growth, but even higher than it was in the couple of decades leading up to the financial crisis. We show a scenario in which productivity growth is pretty close to 2%. Holding everything else constant, that is not quite enough to keep the debt-to-GDP ratio flat, but it is a lot less steep than under what we sometimes call our base or central scenario, where productivity growth is about 1.4% as opposed to closer to 2%.
That calculation tells you that you would need something probably a bit above 2% for productivity growth to hold the debt-to-GDP ratio where it is now, just under 100%. A number of about 2.2% would roughly be the average between the end of the Second World War and the turn of the century. It is not exceptional in British history, but it is very dramatically higher than we have had at any point in the last 20 years.
Laura Gardiner: There is just one thing to add to that scenario David was explaining. If you were to achieve those rates of productivity growth, there is a big question, which we explore in alternate scenarios, about whether Governments would choose to invest the proceeds of that growth in improving public services or bank them to hold debt down.
The scenario that David described is true if spending does not rise as a share of GDP to match the growth in GDP that the productivity delivers. In the period David was describing, the second half of the 20th century, for the most part spending on public services did increase to match the relatively healthy growth in GDP. Spending-to-GDP stayed roughly constant. We show a scenario where you get that higher productivity growth, but primary spending rises with the GDP growth. In that scenario, debt is not very different from the baseline scenario at all.
Productivity growth is great and would make everything easier, but there are still choices for Governments in terms of where the proceeds of that growth go, whether that is into public services or holding down debt.
The Chair: That is absolutely understood. Just to recap on what the three of you have said, in very crude terms, your assumption is a long-term productivity rate of 1.4%. Broadly speaking, in order for debt to be stable-ish, productivity would have to be 50% above that, at 2.1% or 2.2%.
Professor David Miles: It would have to be something like that, yes, and we would have to not spend it all.
The Chair: To Laura’s point, we would have not to spend it all in the way that we did in the second half of the 20th century.
Lord Reid of Cardowan: It is just a factual point. When Professor David was speaking about the triple lock, you said at one stage that this meant an annual increase of 1.5 to 2. I did not know whether you meant billions or percentages.
Professor David Miles: Those were percentages, so big numbers.
Q2 Lord Prentis of Leeds: You were referring to health and pensions. It is very difficult to define productivity in the health service. Say there is a 30,000 shortage of nurses, which has been talked about. If you fill those jobs, productivity goes down and yet it provides a better service. If you are looking at future costs, if the national care service is developed, that will be very expensive. It could be developed as part of the health service or linked to it. You could find increasing pressures on that health spending.
Laura Gardiner: That is a very good point. I would add two points. First, you are absolutely right that defining productivity in the health sector and in public services more generally is extremely difficult. Our assumptions for these long-term projections are based on a range of international evidence and studies showing that it has tended to be the case that labour-intensive public service sectors such as health can lag some of the productivity gains in other sectors where there is more potential for the implementation of technological advancements and things like that. In the long run, wages have to keep up with those in the private sector. That is the core point of where these things that we call “other cost pressures” in health come from.
Your point in relation to the national care service is a good one. We model both health and social care, but we do so a bit differently. One of the things that is a bit different is that we have these increasing non-demographic cost pressure assumptions in health. We do not have that to the same extent in our social care spending projections. That is reflective of things such as the system of restricting eligibility for care. Means testing via local authorities has been a mechanism that can hold down costs to a greater extent than in health.
There is definitely lots of overlap between spending on health and spending on social care in the long run. It is possible that a different model of delivery or provision for care might put upward pressure on our projections for care spending, but that might take a bit of pressure off our projections for health spending. We have thought about them in parallel, and it may be that some of the things we are calling other cost pressures in health may emerge in the care bit rather than the health bit, which is where we have put them all.
Q3 Lord Newby: You said that, instead of having the triple lock, you could increase pensions by average earnings. Why did you choose that rather than inflation? Benefits rise by inflation, not by average earnings.
Laura Gardiner: That is a very good question. We showed both in the report. There is a chart that shows state pension spending under the triple lock, earnings uprating and CPI uprating, or inflation uprating. The gap between the triple lock and earnings—as David said, by the 2070s it is close to 2% of GDP of spending—reflects the historical volatility of inflation and earnings, which allows us to estimate the years in which the non-earnings bit of the lock bite over 50 years. If you go down to inflation uprating, that saves you relative to the triple lock a much bigger 5.5% of GDP.
A common theme across probably every answer that we will give you is that there is a huge amount of uncertainty in all the assumptions that we make on policy, and that is why we show lots of different scenarios. We have shown all three.
Across state pension spending, non-state pension welfare spending and areas of tax threshold uprating, which we might get into later, we have tended to make as the assumption in our baseline scenario, in the triple lock, the state pension bit and other areas, earnings uprating as our default assumption. As you point out, that is a little bit different to some of the default policy commitments in the medium term.
We do that for a range of reasons, but, in essence, on our economic assumptions for inflation and earnings over 50 years, if you assume CPI uprating, the value of benefits in relation to the economy would be reduced by more than half. On the tax side, if tax thresholds go up by inflation for 50 years, under our economic assumptions over two-thirds of workers, including minimum wage earners, would pay the higher rate of tax. Our baseline assessment is that that is quite a weird way to interpret unchanged policy, but we recognise that it is a very finely judged choice in our assumptions, so we show both.
Coming back to the state pension and the triple lock, earnings uprating might be considered the most neutral alternative assumption to the ratchet effect of the triple lock that we have, which would hold the value of the state pension constant in relation to earnings in the economy over a 50-year period. That is consistent with our long-term assumptions on other bits of welfare and tax threshold uprating.
We recognise that there is a very valid case to make for inflation uprating as another alternative, given that it is in the medium-term policy settings, so we show that too, as well as discussing its implications for the economic incentives to work on the tax side and poverty on the welfare side, which is what you would get if you had a 50-year wedge between the rate at which benefits are growing, the state pension and working age and the rate at which wages are growing.
The Chair: We have to make some progress. That was a long first question, which is my fault because I asked all the follow-ups; I apologise. We may have a vote in about half an hour. If so, we will suspend for 10 minutes and come back. I apologise in advance if that happens.
Q4 Lord Young of Cookham: Can we move on to the section of the report that deals with fiscal sustainability? You make the point in the first part of the report that, if we carry on as we are, debt eventually becomes unsustainable and you have a guess at the sort of adjustment that would be needed to get things into balance, which is the equivalent of the education budget. You also make the point that the later we take fiscal tightening, the more difficult it becomes. Can we just deal with one related issue? Is there no scope at all for reducing any of the taxes that we have that in turn might promote growth? Should we discard that? Are there certain taxes that, if they were lowered, might generate growth?
Professor David Miles: It is certainly conceivable. There are some features of the UK tax system that generate particularly high marginal tax rates for people earning in particular bands, for example, which are pretty likely to be a major disincentive either to work or to save. They have some damaging features on the wider economy and potential output and wealth in the country.
There may be ways of getting rid of some of the most egregious damage to incentives that some parts of the tax system generate. If you did that, it would more than pay for itself in terms of a minimal loss in revenue but a substantial positive impact on incentives to work and save. It was not our focus in the report, but that is not to say we do not think that is impossible.
Our strategy really here was to say, “Take the tax system as it is. If it were unchanged, how might that play out?” The message is, “It does not play out very well. You need to make some changes somewhere down the road”. I would not want to suggest that the need to tighten fiscal policy and close the gap between tax revenue and spending necessarily precludes cutting some taxes. If you can identify them, those would be the ones that do the most damage relative to the revenue that they bring in.
Lord Young of Cookham: Would that be the change in the personal allowance once you hit £100,000. Is that the sort of tax?
Professor David Miles: That is the kind of very high marginal tax rate that does the most damage to incentives. It is at that point in the earnings distribution. There are some points at the lower end of the earnings distribution where the withdrawal of benefits can generate some very high tax rates.
Lord Young of Cookham: I do not know whether anyone else wants to come in on that. The Government have been tightening fiscally over the past few years. Could they have done it in a different way that would have had less impact on growth?
Professor David Miles: The primary way that they have done it—this goes back to the previous Government; it has been the policy for some years now and it is set to run a bit longer, so it will turn into a decade—is freezing the thresholds for income tax. I think I am right in saying, though Tom will correct me if not, that that has been behind much of—it is by far the biggest single measure—the increase in the tax take out of GDP, which is what is going to probably mean, on our projections, on the central forecast, that the years of increasing debt-to-GDP at least stop within the next few years and we level off for a while.
Were there other ways of doing it? Almost certainly there were. There is not an obvious open goal that has been missed. If you are going to do this primarily through trying to raise taxes relative to GDP, it is pretty hard to think of a painless way of doing it that will, in the short term, bring in a lot of tax revenue.
Tom Josephs: I just have one thing to add. It is the case that since the end of Covid Governments have had in place plans to consolidate the public finances, but those plans have not been delivered. Over the period since Covid, we have not seen a fiscal consolidation being implemented, in the sense that borrowing has been stuck at around 5% of GDP for the past four or five years. Governments in each of those years have had plans to reduce borrowing over the subsequent five years to around 2% of GDP, which is roughly the rate that you need to get debt stable and therefore have the public finances on a reasonably sustainable path, but, as I say, that has not been delivered.
One important result in the report is that, if Governments do not do anything about the long-term pressures that we identify in our baseline scenario, they start to push debt on to a clearly unsustainable upward trajectory in the 2040s. The starting point for that projection is an assumption that the current consolidation plan is achieved. If it is not achieved and these long-term pressures start to hit from the current level of deficit, that upward trajectory in debt starts much sooner, in the 2030s. It underlines the importance in terms of fiscal sustainability of delivering the current medium-term fiscal consolidation plan.
Q5 Lord Burns: I would like to explore a little bit about whether there might be different outcomes even without major policy changes. What do you regard as the error margins around some of these numbers, particularly as one is looking so far ahead?
Reading the report, it seems to me that what you are highlighting very much is the demographic situation and that most of the pressures that are coming are from the health service, social care and pensions. The demographic factors that are pushing this are now largely baked in. We have a very good idea, do we not, of what is going to happen to the number of elderly people? It will be 25 years before the birth rate can make any difference to the number of people who are in the labour supply. Therefore, it seems to me that it is quite difficult to avoid the conclusions that you reach about the pressures upon health care.
You say that there could be some great breakthrough with regard to health care, but—I am putting forward this hypothesis—the underlying pressures that you are talking about are not likely, for the next 30 years or so, to be very far removed from what you have presented. Is there much larger scope for error in reaching those conclusions?
Professor David Miles: You are right. The demographics are quite predictable and reliable, in a sense, over the decades, in terms of how many people are going to reach the point in their life when they use the health service much more intensively. It is very hard to imagine that playing out much differently.
In a way, it might come back to cost pressures. Who knows quite how they will play out in the health system? It may be that there are productivity improvements or ways of doing things more effectively or treating patients at lower cost that turn out to be very helpful. To the extent that that does depend on technological things, they may be something that you hope happens, but there is not much that a Government can necessarily do about them. They either come along and you can use them or they do not. The health pressures are predictable and quite difficult to avoid unless one gets lucky on the technological front.
Laura Gardiner: There is just one thing to add to that on the health side. We have talked a little bit about the assumption that, as well as demographics, what is driving health spending is growing non-demographic cost pressures.
We show a scenario where those other cost pressures do not exist. In that one, health spending just rises very modestly, to 9% of GDP by the 2070s rather than 13%. That is four percentage points; it is a big gap. That could be driven by technological breakthroughs having a transformative impact on health, as we have discussed. It could also be driven by policy choices to contain those costs persistently in a way that has not happened in the past. That would certainly be challenging, but there are different health systems in other countries that manage that, and the balance between public and private spending is different.
As well as the number of people in the population and their age, about which, as you say, there are some good projections that seem certain they are going in one direction, there is also the health of the population. We do show some scenarios, for example, around the years of life spent in good health. We do this especially because there has been quite a lot of data on healthy life expectancy in recent years. It is quite contested; it is very difficult to measure.
For example, we show that in a higher healthy life expectancy scenario—we went back to an old model where healthy life expectancy tracked overall life expectancy, which does not appear to have been the case recently—you could save another 2% of GDP in health spending just via the population that you have at the ages they are being a fair chunk healthier than in our baseline scenario. That is uncertain. That could come about because of efforts of policy or just because of other factors, such as revolutions in drugs around the world and other things.
I agree with your general point, but there is enough uncertainty in both what could hold down other cost pressures and how underlying health trends could contribute to health spending.
Lord Burns: The state pension numbers must, to a large extent, be a fairly mechanical calculation without some significant change in the ages at which people die. In a sense, we know the number of pensioners who are going to be coming through and we know the difference in the triple lock, whether it is earnings-related or price-related.
Laura Gardiner: I am sorry to say that everything is uncertain, but what do you expect from us? The triple lock, as you know, has this slightly random ratchet effect. Even an estimation of the extent to which the ratchet effect happens and drives a wedge between what would happen with earnings growth is quite uncertain.
In last year’s report, we looked at an assumption in which the triple lock ratchet effect followed a more volatile period for inflation and earnings, so the period since 2010, rather than measuring it over the whole period since 1990, which we did in the baseline. That would add another 1.5% or close to 2% of GDP again. Even how much the triple lock forces things up relative to earnings is pretty uncertain.
Lord Burns: I was concentrating more on the things that could make this less of a problem rather than—
Laura Gardiner: That is a two-sided risk. We could have a very stable period where earnings are always higher than inflation and always higher than 2.5% over the next 50 years. In that scenario, the pressure would be much less. It could go the other way.
Lord Burns: What about any of the other assumptions? Are there any great margins of error on things such as immigration or participation in the labour force?
Professor David Miles: On immigration, we have taken the latest Office for National Statistics central projection, which is not much more than them taking the average levels of net immigration over some past period, maybe 10 or 15 years, and saying, “That is the new norm in the future”. Their number is 230,000 a year, which would be simultaneously a lot less than it has been in recent years and a lot more than it has been over the last 40 or 50 years. It is a very high level. That is what we do.
In a way, that is more under the control of the Government than almost anything else that we have been talking about so far. If you could pull a lever and change the number of people arriving in the country, it would still be very unclear quite what the fiscal benefits or costs might be because it would depend on what kind of people arrive, what proportion of them go into the workforce, what wages they earn, at what point they become eligible for welfare and whether they stay in the country.
Ideally, for fiscal purposes, you want people to arrive at 25, stay for 20 years and leave while they are still in good health. They have not drawn on the health service; they have paid tax; they do not get the state pension; and they go again before they become fiscally very costly. You cannot guarantee that. If most people arrive, become like the rest of us almost immediately and stay until they die, although there is a short-run fiscal benefit, it is much less obvious as you look further down the road.
On the impact of significantly higher or indeed significantly lower net immigration over the long-term, my own view—we do not focus on it too much in this report—is that in the long-term immigration is not really an answer to the question about how we become more sustainable.
Lord Burns: One other area of uncertainty—this goes the other way—is the point that you make in the report that, on average, we have had some kind of crisis once every nine years that has been a shock that has required some significant budgetary support from the Government. Could you say a bit more about how you have modelled that and what the consequences of that are?
Tom Josephs: I can, yes. In our baseline scenario, we assume smooth and steady economic growth all the way through 50 years. That is clearly very unlikely. What we have seen over the past three decades is that the economy has been hit by a very major shock roughly around every 10 years. As you say, that has had big fiscal costs both in terms of the impact that it has had on the economy and the cost of government policy support to address the shock.
We have a scenario where we assume that, instead of steady growth, the economy gets hit by a major shock every 10 years roughly. Unsurprisingly, that means that, again, the starting point of the upward trajectory in debt happens more quickly than in the baseline scenario because you are pushed on to an unsustainable path by the shock more quickly and the upward path in debt is then steeper, if Governments were to do nothing about it. The reality is that that is a more plausible central scenario than the one that we call the baseline.
Q6 Lord Razzall: We have rather done to death the issues of the triple lock. As I understand it, what you are saying is that, if the triple block is maintained, it is a 9% increase during the period that you are measuring. It is 7% if you go to average earnings and 5% if we go to inflation. First of all, is that correct? You only need to nod.
Laura Gardiner: It is a bit lower on the last one. It is 3% or 4% if you go to inflation. Yes, that is spot on otherwise.
Lord Razzall: What I am not quite clear about, and neither was Lord Reid in his earlier question, is the actual monetary impact of those three scenarios on your calculation of GDP.
Laura Gardiner: Sorry, are you asking for a figure?
Lord Razzall: What does it do to the debt-to-GDP ratio?
Laura Gardiner: The one of those that we have fully flowed through to the debt-to-GDP ratio is the difference between the triple lock and earnings, so the first two. Our baseline projection has the triple lock in it and has debt rising to around about 300% of GDP by the 2070s. In the alternative scenario, where instead of the triple lock you uprate with earnings, it rises to about 260% of GDP. It takes close to a fifth off that. That is a difference of 40% of GDP.
Lord Razzall: What about inflation?
Laura Gardiner: We have not modelled inflation right through to debt. I do not have those figures to hand, but it would be a sizable chunk lower.
Professor David Miles: If I may add, it makes a material difference, but there is no sense in which—
Lord Razzall: It does not solve the problem.
Professor David Miles: Yes, that is exactly right. It makes it less, but there are a lot of other things that you need to do.
Laura Gardiner: That goes with the theme across our scenarios. On both the tax and the spending side, across the scenarios that we look at, lots of things get the debt path a bit lower and some get it a bit higher. In almost all the scenarios that we model, debt eventually reaches an unsustainable trajectory over the next 50 years. We have not come up with one silver bullet on either the tax or the spending side.
Lord Razzall: No. The triple lock is just one factor in all that.
Q7 Lord Liddle: One of the shocks that we may be experiencing at the moment and that might come again in a troubled world is an energy shock. This may compel Governments to produce some sort of support package, however generous it is. Have you thought about this question of how you deal with energy shocks? Could you design any action by Government that would mitigate the long-term fiscal damage?
Tom Josephs: I can take that one. What we have not done and what it is not our role to do is design policy responses to particular shocks. Our job is to cost the policy responses that Governments choose to use. In the face of the current energy shock that we are facing, rather than thinking about what the policy response will be, our focus is to think about what the impact will be on our next medium-term forecast, which we are likely to do alongside the next Budget. The previous forecast that we did in March was done too late to take into account the impact of the energy shock from the start of the conflict in the Middle East.
The other thing to say is that, compared to the 2022 energy price shock, the current one so far is much less severe, especially in terms of the increase in gas prices. The energy package that the Government announced in 2022 was designed to deal with that much larger increase in gas prices, in particular, and was very expensive.
In terms of the general principles of policy advice that you would give in response to a temporary shock, to minimise the fiscal cost of that package, you would usually make it temporary so that it deals with the short-term shock but is not a permanent policy giveaway; make it as targeted as possible in terms of reducing cost for those who are particularly affected and perhaps need the help most; and make it timely, so that it gets that support to people quickly. “Temporary, targeted and timely” is the classic advice for policy responses to this form of shock.
Lord Liddle: Have you done any work on the policy costs of trying to insulate ourselves from future oil price shocks or, in other words, having an energy system that is not dependent on the Gulf? Could you justify a big capital investment in order to reduce that dependence on the basis that it will prevent these shocks being such a problem in future for the public finances?
Tom Josephs: That is not something that we have specifically tried to model. You are suggesting an investment in renewable energy in order to move away from the dependence on fossil fuels and nuclear. We have not tried to model that specific trade-off.
What we have done in our reports is looked at the cost overall of the transition to net zero on the basis of the Government’s current timing for that, i.e. the net zero commitment for 2050, and the Climate Change Committee’s baseline path for a transition to net zero over that period and what policy action would be needed to deliver that, which does include transitioning away from fossil fuels.
We find that there is quite a material fiscal cost to that. It is around 20% of GDP over 50 years. Most of it comes from the loss of emissions-related tax revenue, in particular fuel duty. The cost of the public investment that the Government would need in order to support that transition is quite a lot lower. We roughly have around £10 billion a year, in today’s terms, of extra investment in things such as the energy transition and transport systems. A big chunk of the cost is switching homes over to heat pumps and that kind of thing.
Q8 Lord Reid of Cardowan: We have spent quite a lot of time discussing the costs of an ageing population in terms of demographic factors, pensions and health. People are living longer, as you said, but they are also more liable to suffer from chronic diseases. They are more costly. One other thing, which I am not sure was mentioned, is human ingenuity. People are constantly inventing technological solutions for health problems and pharmaceutical solutions that were hitherto unknown. Certain diseases were regarded as something you just had to accept.
Can I switch it around and ask about age and revenue-raising rather than the cost side of it? To some extent, this follows on from the questions that Lord Burns was asking. What changes, if any, to the UK’s tax base would be most suitable and beneficial in the context of an ageing society, if your aim is to promote a more sustainable fiscal backdrop, apart from an annual cull?
Professor David Miles: There is something that would help greatly. It is not necessarily a fiscal measure, but things may play out this way, and I am quite optimistic about this. Currently, we live in a world—it is the world that we have been in for some time, but it may be that the future will look different from this—in which people are living longer but expect to retire in their 50s or early 60s. In many ways, a change in expectations would be beneficial on many different fronts.
First, people are clearly living longer than they were 20, 30, 40 or 50 years ago and yet are retiring on average at about the same age as they did in the past. That may change. It may change as a result of people realising that they feel okay. Work is less physically demanding, certainly, than it was 50 or 60 years ago and probably less physically demanding than it was even 30 years ago. That may have still quite a long way to go, which would mean that people are able and, on the whole, willing to work a bit longer.
Employers might find that they can no longer take a view, which may have been prevalent in the past, that they do not really want to offer a new job to somebody in their 50s or 60s because there are lots of people in their 20s, 30s and 40s who are willing to work. If the demographics change, as they almost certainly will in the UK, the relative number of people in their 20s, 30s and 40s to the number of people in their 60s, 70s and 80s, employers might naturally, in their own economic self-interest, consciously try to hire more people at an older age than they did in the past and, to induce them to work, may offer more flexible and part-time work. This plays out to everybody’s advantage in many ways and certainly has fiscal benefits as well.
One of the reasons why, on our projections, having a rising proportion of the population in the 70s and 80s generates some fiscal pressures is that, on the whole, at the moment people in those parts of their life are not paying an awful lot of income tax. That may be different down the road. If it is, that may be one of the most fiscally advantageous things. None of that is in in our projections. We do not take account of those possibilities.
I am personally quite optimistic about that. I am also optimistic about the other thing that you mentioned, which is technological progress. It is finding drugs that hopefully are not terribly expensive, do not have bad side effects and are very effective if you start taking them at a younger age as protection against some of the big killers today, such as cardiac conditions, cancers and dementia.
I have heard medical people and people in the pharmaceutical industry wax lyrical on the benefits of statins, which are pretty well known, but there are other drugs that people are discovering were never designed as protection against dementia or cardiac conditions but seem to be very effective against those conditions. They are not that expensive either, but they are not widely used at the moment. Maybe that is coming.
Lord Reid of Cardowan: You said at one stage that people were living much longer but retiring at the same sort of age. My impression—it is completely impressionistic—is that more and more older people are continuing in some sort of gainful employment. They are working longer in their lives. Maybe it is because I am in the House of Lords that I think that. I got the impression that that was evolving, but let us leave that aside.
What you are suggesting is social and cultural change. I was actually asking whether, assuming the present set of circumstances of an ageing population, a change to the UK’s tax base in terms of mechanisms might be more suitable for an ageing society in order to reach a more sustainable fiscal backdrop?
Professor David Miles: It is a tricky one. It deserves some more thought. Off the top of my head there is not anything obvious. Tom and Laura, do you have any thoughts on that?
The Chair: Lord Newby has an idea.
Lord Newby: An obvious thing that you could do would be to extend national insurance to everybody. Does that raise much money?
Professor David Miles: Yes, it is quite a big generator of tax revenue and has grown over the years.
Laura Gardiner: The specific policy you are referring to would be employee national insurance on people working above state pension age. Consistent with the fact that we cannot model alternative policies, that is not something we have looked at in particular, but it clearly is a big cliff edge in the tax system.
On the general question about the tax base in an ageing society, one thing that is quite clear from our work in this report is that demographics have a very large impact on the spending projections. We put those same demographic projections through lots of the largest taxes to produce our projections and they do not have anywhere near as large effects.
Some of the things that drive changes in tax are not related to the ageing society, such as the transition to net zero, which Tom has talked about. There we highlight the loss of revenues, but also the things that have happened in recent years, in particular the introduction of electric vehicle excise duty, which partially offsets the loss of fuel duty revenues.
David is right that the question of redesigning the tax system for an ageing society is a very complex one. We probably do not have a prescription that we can offer to you now, but there are other things driving changes to tax over the next 50 years. Net zero is one, but, as I said, there have been some policies introduced that partially address some of the loss of revenues.
Another thing that we look at, which we might get into later, is the risk around artificial intelligence to the tax base and the potential loss of revenues if AI mainly substitutes for labour. Because corporate profits tend to be taxed at lower rates than wages and salaries, there is a risk there. That is another complex set of issues to think about in terms of making the tax system fit for the future.
Ageing is one aspect to think about in relation to the tax system, but the transition to net zero, the impacts of AI and the other things that we flag in the report are probably just as important as demographics on the tax side.
Q9 The Chair: Can I ask you about inactivity? We have had the Alan Milburn report on NEETs and the growing challenges we face with the cohort of people who are 18 to 24 and younger. In our report on the ageing society, we looked at inactivity in the 50-to-65 group, which we were surprised was as significant an issue as it has been. The general debate about inactivity focuses on the pension age itself and how far up that should go. From a fiscal dividend point of view, what is the most profitable cohort to focus on? Where is the biggest bang for our buck, from a fiscal point of view, in reducing inactivity?
Laura Gardiner: I will have a first go at that. There are two parts to it: what could the fiscal impact be, and where should you focus? The first of those is something that we looked at, around when you were doing your ageing society report, in our 2023 Fiscal risks and sustainability report. As always on this topic, I should caveat that. As you probably know, the data from the ONS Labour Force Survey has been challenging over recent years.
The Chair: We are very aware of that, yes.
Laura Gardiner: It has been hard to form a clear picture. In the latest data—the survey has been slowly improving—two things are clear. First, it is even more the case that health-related reasons are dominating the rise in inactivity. There does seem to have been a rise since the start of the pandemic. It is maybe around 700,000 people since early 2020 from the last stats that I looked at.
In our 2023 report, we said, “What if you reduced health-related inactivity by 500,000?” That would be most of the post-pandemic rise we have seen. Looking over the five-year forecast period, back in the medium term, that could reduce borrowing by about £20 billion and debt by about 3% of GDP. Two-thirds of that comes from higher tax revenues from more people working and one-third of it comes from welfare savings. We did that a few years ago now, but that gives you an idea. Yes, there would be a fiscal benefit from reducing inactivity, and that could play out in amplified ways over the longer term.
On the question of where to focus, as I have said previously, it is not for the OBR to advise on specific policies, but there are some general points that I would make. Age itself is not always the best prism to think about what to do. It is probably the underlying reason for inactivity where health seems to be a really big thing, but the duration of inactivity, the skill level and the job readiness of inactive people are as important.
The other caveat that I would make is that in this area it is really tempting to look at who is inactive now and think, “How far away are they from the labour market? What could be done to get them into work?” It is just as important to think about preventing the outflow from work to inactivity. Just getting people back into work is dealing with the problem after it has occurred. The rates of entry to work from inactivity are very low, so you can probably do just as much to try to prevent people from getting into that state in the first place. That might be about when people get ill in work, the transition from education or other things.
Those are all my caveats, but now I will answer your question. Since we did our work a few years ago and you did your report, the evidence has got very much stronger around young people being a real source of the issue. I looked at the data this morning. Since 2023 inactivity among young people has carried on rising whereas it has been falling among the over-50s. The picture has substantially changed since you did your work and we did ours a few years ago.
On top of that, there is the intuitive point that young people have much more of their lives ahead of them. Anything successful that you can do to intervene could have social benefits for the individuals and fiscal benefits for all of us that extend over many decades rather than just years.
That is backed up by some of the evidence in Alan Milburn’s interim report around scarring. He gathers a very impressive collection of academic work over a number of decades showing that periods of inactivity when you are young affect your earnings and employment over maybe 10 or 20 years, maybe even longer, but they also affect things such as your health outcomes over that period of time.
I would normally say that it is a bit simplistic or binary to pick an age cohort and focus on it. I emphasise my point that you want to think about the reasons and the duration and not just age, but it does seem increasingly clear that the inactivity problem is being concentrated among the under-25s. There is a growing body of evidence that, though it might be hard, arresting those challenges will have much longer-term impacts because they have so much more of their lives ahead of them.
The Chair: That is very clear. Thank you very much.
Q10 Lord Butler of Brockwell: I think it is the case that your projections show that, even if the Government’s present medium-term fiscal objectives are met, we are still on a path to an unsustainable position. Is that right?
Tom Josephs: That is the case, yes. It is also the case that, if the Government deliver on their medium-term consolidation, the unsustainable rise in debt happens quite a bit later than if that medium-term consolidation is not achieved. The starting point for the projections is the current level of the deficit, which is relatively high by historical standards.
That is an important finding in our report. Action now to get the public finances on a sustainable footing is really important in terms of being in a better position to be able to face these longer-term pressures.
It is also true that, even if the Government deliver on that consolidation, a further adjustment over time would be needed to prevent debt moving on to an unsustainable trajectory. We show that taking additional action earlier rather than later is much less costly than delaying.
Lord Butler of Brockwell: Is it not therefore the case that the medium-term fiscal targets are in danger of producing a sense of complacency that is not justified? Are they, in a sense, misleading because they do not prevent the country being on the path to an unsustainable debt position. Is that correct?
Tom Josephs: It is certainly the case that the current medium-term consolidation plan is just about sufficient to prevent a rise in debt in the medium term and get debt on a stable and slightly falling path in the medium term. It is essentially the minimum that you might think is consistent with sustainable public finances. Yes, as you say, it is not sufficient over the longer term to do just that and think no other fiscal adjustment will be necessary. It will be necessary, if these projections are right and start to materialise, for Governments to take further action after the end of the medium-term forecast.
It is also the case that, as I said earlier, the current medium-term fiscal consolidation plan is one that would deliver a fairly significant reduction in borrowing. It would get the deficit down to a level that would be its lowest since around 2000. That is not to say that the current government plan is not making any progress towards getting the public finances on a more sustainable path. It represents a reasonably material fiscal consolidation, but, yes, our analysis suggests that more would need to be done subsequently.
Lord Butler of Brockwell: The conclusion I draw is that, though the current fiscal framework is necessary, it is not sufficient to prevent what will in the long term be an unsustainable position.
Tom Josephs: Yes, I would agree with that.
Q11 Lord Prentis of Leeds: My question follows on from Lord Butler. We have talked for many years about the sustainability or otherwise of our debt. It has been a feature of the last 20 years at least. It is now almost touching 100%, but it is continuing to rise. We have also talked about what action could be taken, both in the short term and the long term. David, you said that it is a matter that has to be dealt with today rather than tomorrow. It is a serious issue. In the FRS, you also emphasise the fiscal risks of not taking action and the risks involved in government inaction.
My question really relates to that and the point that Lord Butler was making. Are there any examples that you can point to of any Government dealing with an unsustainable fiscal pathway of the scale we face in the UK without that Government being forced into doing so by a monumental crisis?
Tom Josephs: It is certainly the case that, unfortunately, a lot of countries have only confronted these issues properly when faced with a fiscal crisis, but there are also examples where debt has successfully fallen, quite significantly, outside a crisis. An example is the UK after the Second World War. Debt was at a very high level coming out of the war. As a consequence of both the First World War and the Second World War, debt had been pushed up to very high levels. It came down fairly rapidly in the three decades or so following the Second World War.
That was driven by a few different factors. First, for much of that period we had economic growth that was significantly higher than the interest rate that the Government were paying on their debt. That is a big factor that allows debt as a share of GDP to fall, if you have a strongly rising economy and the cost of your debt is lower. That was driven by a combination of strong growth after the war, favourable demographics, lots of new people entering the labour market as well as, in part, various regulatory factors that kept interest rates down.
Lord Prentis of Leeds: Would you not say that that was after a crisis—the Second World War?
Tom Josephs: That is true. In that case, what happened was the rise in debt was caused by the crisis. We did not have a fiscal crisis that forced the Government to address the large stock of debt. It managed to get debt down through, as I say, a combination of higher growth, lower interest rates and running very low deficits for quite a substantial period of time.
It did help over that period as well that Governments were able to reduce defence spending quite substantially. It was at very high levels during and after the war. The fact that Governments were able to reduce spending on defence allowed room for other government spending priorities to be delivered while also reducing debt.
The Chair: I am afraid we have to suspend because we have a vote. I will suspend the meeting for 10 minutes.
Sitting suspended.
The Chair: Thank you for your patience. Welcome back after the tea break—or voting break, I should say—in our session. Tom, you were answering Lord Prentis’s question.
Tom Josephs: Just to finish off, the period after the Second World War in the UK was a period when debt fell quite rapidly. It was driven by favourable demographics, strong economic growth, low interest rates and falling defence spending. All those factors are not ones, unfortunately, that we are experiencing now. As the report shows, the demographics are rather unfavourable from a fiscal point of view; we have had a period of low growth; most recently we have had high interest rates; and, after a period of defence spending falling, we are now entering a period where defence spending is rising again. Unfortunately, we are not able to benefit currently from the factors that helped bring down debt in the post Second World War period.
Lord Prentis of Leeds: I get the impression that Governments of all different colours put off dealing with the unsustainability of debt—it is for another day—whereas the OBR is saying that the trajectory is so great that it should be dealt with now, otherwise future generations are going to suffer. The report refers to this—that the young people of tomorrow will suffer unless we take action now. It is very difficult for a Government to take action of such magnitude if they are seeking to deal with everything over a five-year period in the normal way of government. From what you are saying, it seems like we need to deal with it now. How do you get the will to actually do that?
Tom Josephs: The report is saying that the cost of taking action early to deal with these pressures is much lower overall than the cost that would result if you were to delay dealing with the pressures to much later down the line. In fact, we have some analysis that shows that, if you deal with the pressures early, the cost is around half of what it would be if you were to delay for a couple of decades and then try to deal with it, when you are in the midst of the upward trajectory of debt.
The other thing that I would say is, consistent with the discussion we have been having at this session, our job is not to recommend a particular policy prescription for dealing with these pressures. The report does recognise that there are difficult trade-offs and costs and benefits to all the potential ways of dealing with these pressures. Therefore, that suggests to us that Governments will need to take action across a number of different fronts, spending and possibly taxation. It would be very difficult to deal with it through just one route.
Professor David Miles: Your question is a profoundly important political one. How do Governments get the support of the public to do something that is very difficult right now? The answer to that difficult question is that they have to be honest with the public about what the risks are if they do not do that, which will affect the young people of today who are already here, their children, their grandchildren and the unborn, frankly. If you could build a consensus among the public that it is the right thing to do, almost morally, it becomes politically feasible to do it.
Without that, it is very difficult because people say, “Why are you cutting back on services? Why are you increasing taxes? Why are things getting worse for us right now?” There will always be another political party that will say, “The people who are in power now are hopeless. Vote for us at the next election and you will not need to do these terrible things, because they are incompetent”.
It seems to me that you have to level with the public about what might be down the road. No one is certain, of course, and we could get lucky and not need to do these dramatic things because technology and productivity go in a certain direction, but you would not want to bank on that. For the next generation, most of whom are already alive now—they are young people—you would not want to gamble that we will get lucky, things will turn out okay and we will not need to face these very difficult fiscal situations 20, 25 or 30 years from now.
Q12 Baroness Penn: On the topic of getting lucky or a more optimistic potential scenario, Laura, you touched on the potential risks from AI in terms of its impact on the labour market and therefore the tax base. There is also potential upside from technology and AI in terms of its cost-reducing potential and productivity-increasing potential. Do you see any signs of that more optimistic narrative? If you were looking for evidence for or against that, what would you be looking for? Of course, there may be a mixed scenario where you get positive sides for future fiscal sustainability and some of the risks that you touched on as well.
Professor David Miles: AI is one of those things where there are intelligent people who know a lot about it who have completely different views on how it will play out.
The Chair: That is not like the economy in general, then.
Professor David Miles: I hope I have not bored you with this before, but two of the most recent winners of the Nobel Prize in economics, Daron Acemoglu and Philippe Aghion, both got their Nobel Prizes for work on growth and technology and have done lots of work on AI. One of them, Acemoglu, thinks that it will have a negligible impact on productivity and quality of life for most people and there may be more downsides than upsides. Aghion, who is the latest winner of the Nobel Prize, thinks it will be transformational. He thinks it will boost productivity enormously and have hugely positive effects on the standard of living and probably huge fiscal benefits. They are just all over the place.
We have a central scenario where we allow for a meaningful but definitely not transformational impact of artificial intelligence on productivity. It is one of the reasons why, in our projections, we have productivity growing a lot faster than it has been for the last 20 years. Relative to the optimists, that is ludicrously pessimistic.
It is one of those things where the range of opinions is so wide, and well-informed people cover the whole spectrum of views, that all one can end up doing in terms of these projections is to say, “Here is a central projection and here is a significant upside”, but the significant upside, as we have called it, may turn out to be more pessimistic than the actual outturn.
All of that makes it very difficult, coming back to Lord Prentis’s question, to know what to tell the current generation in terms of what we should do on fiscal policy, taxes, public services and all of that, when there is a chance—it may be a quite high chance—that you do not need to do all that fiscal tightening because 20 years from now standards of living will be way higher and it will all be very easy to deal with.
While that might be true, do you really want to roll the dice for a future generation and say, “There is a chance that it will all be okay for you, but there is a significant chance it will be really difficult”? That becomes almost a moral question about what risks you are willing to impose on young people and the unborn.
Baroness Penn: One of the challenges that we have heard in other evidence sessions covering some of this area is that the potential pace at which AI might bring about change could be significant and therefore the gap between having signals about what direction it may go in and being quite far down that path could be quite small. If you were looking for signals of one impact or another the next time you update this report, where would you be looking? Whether that is positive or negative, what evidence would you be looking at? Is it still too difficult to see the impact on day-to-day life in the numbers?
Professor David Miles: It might show up in employment trends and whether there are companies that have decided, not because they were pessimistic about the future demand for their product—they could be very optimistic about that—to cut back enormously on hiring because they are clearly switching AI in place of people. Although people worry a lot about that, I am not sure there are any obvious signs right now that that is happening. It might be one of the first signs. If it is, it is a sign that is a two-edged sword. It is bad news for people who were hoping to get jobs with those companies, but you might say it was good news in terms of labour productivity.
That may be where you see it coming most clearly. It has not emerged yet. Although there are lots of stories in the media about, “This is the future”, they are stories about the future at the moment.
Q13 Lord Newby: When we are not talking to you, we are at the moment thinking about fiscal devolution and its potential impact. There are a number of arguments for and against fiscal devolution, but one of them, which is very much in your purview, is in terms of growth. Have you given any thought to whether fiscal devolution, and if so what kind, might have a material impact on your projections?
Tom Josephs: That is not something that we have done a specific piece of analysis on at the current time. The new Government are coming in with this being a big focus of their policy agenda. We will assess any policy that is given to us by the new Government on devolution and look at the evidence there.
There is clearly a wide range of possibilities under the banner of devolution, from changing the structure of subnational government to more devolution of public service delivery or the devolution of tax-raising and borrowing powers. There is a wide range of potential policy options that might have very different economic and fiscal implications.
For us, one factor will be the degree of certainty around which you could assess the potential economic impact of those kinds of reforms. By their nature, they are reforms that are likely to take quite a long time to implement through changes at the national level, changes at the regional or local and then policy choices at the regional and local level. At the start you will not necessarily know how those will pan out and, therefore, you will not know the economic and fiscal implications of them.
An example is the devolution of tax-raising powers to Scotland and Wales. In the case of Scotland, it took about nine years from the first recommendation to do that through to the point at which Scotland had the ability to change income tax rates and did so. In Wales, it was around seven years. So far they have decided not to change rates at all.
From the perspective of trying to forecast the impact of those policies, there will be quite a lot of certainty around both the nature and timing of those changes and then what the actual final impact of devolving those powers would be.
It may also be that devolution has a range of different policy objectives. Some of those may be around growth while others may be more around having a better ability to target and deliver public services, which could be very important but is a bit outside of our area of focus, at least in the medium term. That sort of thing could have longer-term benefits for economic growth but, again, those might come quite far in the future.
Lord Newby: Have you done any work about the impact of devolution on growth and revenues for Scotland and Wales?
Tom Josephs: We have a role in forecasting Scottish and Welsh tax revenues. That is a role that we have had for a few years. We do not do a separate economic growth forecast for either Scotland or Wales. We do a UK-wide macroeconomic forecast. We have not done any work specifically looking at the impact that the devolution of income tax has had on economic performance in those countries.
Lord Newby: They are the closest worked examples of devolution that we have and the evidence seems to be ambivalent.
Tom Josephs: From a fiscal perspective, which is our main focus, there are a number of risks that you might think about in relation to devolution. One is the risk that you devolve responsibility for public service delivery but potentially not the funding necessary to deliver that, which therefore could create additional pressure on public spending because a local area will be incentivised to deliver more but might not have the funding to do it and eventually will call on central government for additional funding.
On the tax devolution side of things, there is clearly a risk that you can create incentives for tax competition between regions. That is particularly a risk in the UK, where we are quite densely populated and a lot of these areas are quite close to each other. If you have different tax rates in areas that are very close to each other, that could create incentives around household and business location and quite a lot of complexity, potentially, in the tax system. That is a risk that policymakers need to think about from a fiscal perspective.
The devolution of revenue brings with it a risk around revenue volatility, which can be greater in a smaller area where the tax base is more concentrated. That is certainly an issue that the Governments in Scotland and Wales have faced. If the revenues from devolved income tax do not come in as forecast, they have quite limited borrowing powers that allow them to cushion that revenue volatility. That has been an additional challenge for them to deal with.
Q14 The Chair: Can I ask you a small question in closing and then a bigger one? Tom, you mentioned defence spending earlier on. You said that you take seriously the Government’s commitment to reach 3.5% by 2035. Is it just a straight linear increase in the way that you model it?
Tom Josephs: Yes. In our long-term projections, we just draw a straight line. We have a couple of scenarios around that for what it would look like if you went to an average that was more like the Cold War average, which is higher. Yes, we basically just draw a straight line.
The Chair: With your Fiscal risks and sustainability report, you pick themes to look at or focus on. I know it is a long way off, but do you have any idea what the next shortlist of possible long-term themes might be?
Tom Josephs: The way that we tend to do these reports is that one year we focus on the longer-term projections, which is what we did this year, and then the next year we do a more in-depth look at particular sources of risk. We have flagged in this report a couple of areas that we might look at next year. One is on climate change. We have done work in the past on the fiscal risks from climate change damage and the costs of the transition to net zero. The missing piece of the analysis is around the cost of adaptation to deal with the damage from climate change. We may do that next year.
We have also talked about doing a deep dive into education spending because it is the main area of significant public spending that we have not done in the past. We have done pensions, health and defence, so we thought we might look at education in more detail next year. We are also open to ideas.
The Chair: That is a good early indication of possibles. Laura, David and Tom, we are very grateful. Thank you so much for your time. We really appreciate it. I hope this is the beginning of a wider effort on the part of Parliament to take the fiscal risks and sustainability seriously, as we have talked about a few times now. Thank you again for your time. With that, the meeting is finished.
