Treasury Committee
Oral evidence: Spring Statement 2018, HC 885
Tuesday 20 March 2018
Ordered by the House of Commons to be published on 20 March 2018.
Members present: Nicky Morgan (Chair); Rushanara Ali; Charlie Elphicke; Stephen Hammond; Stewart Hosie; Alison McGovern; Catherine McKinnell.
Questions 1-81
Witnesses
I: Robert Chote, Chairman, Budget Responsibility Committee, Sir Charles Bean, Member of the Budget Responsibility Committee, and Graham Parker CBE, Member of the Budget Responsibility Committee.
Witnesses: Robert Chote, Sir Charles Bean and Graham Parker.
Chair: Good morning. Thank you all very much indeed for coming before the Committee again, following the spring statement based on your recent forecasts. For the purposes of those watching, may I ask you to introduce yourselves?
Robert Chote: I am Robert Chote, chair of the OBR.
Sir Charles Bean: Charlie Bean, OBR.
Graham Parker: Graham Parker, OBR.
Q1 Chair: Thank you. We will start, probably inevitably, with Brexit, but I promise you we will move off it at some point, although I cannot tell you exactly when. Before I ask about Brexit forecasting difficulties, let me ask what I think is a very standard question: Robert, can you confirm that the OBR has come under no pressure from the Treasury or other Government Departments to alter your economic and fiscal forecasts?
Robert Chote: That is correct.
Q2 Chair: Okay. Moving on to Brexit forecasting, you state that the cross-Whitehall briefing on the economic impact of different Brexit scenarios did not affect the forecast, but that “further scrutiny of this analysis—and the assumptions and judgements that underpin the results—will no doubt be helpful when we come to update our provisional Brexit assumptions for the eventual withdrawal agreement.” I suppose the question is really whether you plan to ask the Government for further information about the assumptions and judgments that underpin that cross-Whitehall analysis, and what you think the key unanswered questions are.
Robert Chote: Yes, we will. As you say, we have not yet used it in anger for a forecasting purpose, because we are not yet in a position to identify a particular outcome that we want to use as a focus for the forecast. But for the sorts of judgments that we have made so far and the sorts of updates to those judgments that we might need to make in the future, the cross-Whitehall study clearly provides a variety of interesting approaches to some of those questions, which you would want to look at. By the way, we have seen no more than the slides that have been shown to parliamentarians.
Q3 Chair: So you have not seen the underlying modelling or the methodology?
Robert Chote: No, although under the legislation we have a right to probe that further in due course, which is what we would certainly do. A particular set of numbers may have emerged from the process, but the key thing when looking at this sort of analysis is to see the key judgments and assumptions driving the size of those numbers and the differences between the scenarios. I would not place huge emphasis on the particular numbers that have come out; it is about what is driving them. Those are the sorts of issues that we will need to think about.
One example of what appears to be quite a big driver of the numbers that have appeared is your assessment of what the impact of non-tariff barriers would be. That is something that is very hard to do definitively, and I think we will want to look quite closely at the sorts of judgments that the cross-Whitehall team have made in that respect and whether we think that that is a good or bad guide.
Another issue would be the impact of other trade deals with non-EU countries. Views would differ on what the impact might be of a trade deal with the US. It has to be said that, within that study, the impact of those other trade deals is relatively small, so even doubling all the effects would not make a huge difference to the overall numbers as they have been set out. Charlie will give a more informed figure, but those are the sorts of things you would want to probe, because you do not take the numbers at face value. It is not simply a question of saying, “Well, they’ve done all the work and all we need to do is work out at which bit of their dartboard we need to aim the dart,” with that being the number you end up with. It is, “What useful food for thought did this provide us in reaching those judgments?”
Sir Charles Bean: Absolutely. I think we would approach it the same way we approach a regular forecast. It is often the Departments who are doing the detailed work on taxes and spending, and our job is to kick the tyres and say, “Do you really believe that?” and “How have you come to this number?” and so forth. I would expect us to do exactly the same here.
On the basis of what we have been told so far, we could not make a judgment on whether we think the methodologies are entirely appropriate or not, and reaching a judgment about the robustness of the numbers would certainly require quite a lot more work on our part.
Robert Chote: The other thing is that wherever you come out looking at the headline differences between the sets of numbers that come out, as everyone naturally does, the degree of uncertainty that would lie around any one of those is very large in the context of the differences you are drawing between the two of them. So while at the moment we are talking about broadbrush assumptions, we are still going to end up with pretty broad brushes at the end of the process, because perhaps you can be more precise about how you have reached the judgments, but there is a lot of stuff here that is unknowable and very difficult—there is no past evidence and no past experience of events to draw from—so there is still going to be a lot of art as well as science in this, right up to the endpoint.
Q4 Chair: Let’s explore that timescale over the course of this year. I am particularly thinking about the likely parliamentary timetable of a vote on the withdrawal agreement. At the moment, I think the Government’s intention is still that Parliament will have a vote in let’s say October—around the autumn—which is likely to be before the autumn Budget. I want to unpick this a bit more. When do you expect to have to ask the Government for more details that underlie the cross-Whitehall analysis? What more details do you need from “really broadbrush” to “perhaps slightly less broadbrush” but still making assumptions?
I take your point, Robert, that this has not been done before. As you know, in our report on the Budget we recommended that, depending on when the vote happens, “the OBR should prepare a special forecast” on the withdrawal agreement. Perhaps you can talk me through the thinking in your response. You say: “The Government will presumably take the parliamentary timetable for consideration of the Agreement into account when deciding the dates of the Autumn Budget and Spring Statement.” My view as a parliamentarian is: don’t bet on it. Then you say: “It would not be practical to produce an additional forecast between the two”. It is right that this Committee should point out that parliamentarians will need more guidance from the OBR and more questioning of the assumptions made before they vote on the withdrawal agreement. Could you talk us through how you think this year will develop in terms of asking for more detail, votes on withdrawal agreements and special forecasts?
Robert Chote: Sure. On your first point about when we would start to look at the material, we can get under way with that. We have got this forecast out of the way and obviously we have other publications that we are producing in July time, as we normally do, but we can start looking at that over the summer, well ahead of whenever you need that.
Obviously, you will have a clearer sense of what the likely parliamentary timetable would be. That is an earlier date than anybody has said to me.
Q5 Chair: Really?
Robert Chote: Yes. But I am not saying that anybody has given me another date that is later; I am just saying that that—
Chair: Sorry to be the bearer of bad news—
Robert Chote: I think that in the back of people’s minds was the issue of, “Does this fall awkwardly between the autumn Budget and the spring statement?” That is a slightly different issue about whether—
Q6 Chair: So between November 2018 and March 2019.
Robert Chote: Yes, but, as I said, you are the experts in this area; I do not know what the timetable would look like.
The thing to bear in mind is that when we do a forecast of this sort, it is not just us who does it: we rely on the kindness of strangers in HMRC and DWP in particular. To give you some sense of what the timetable needs to be, the legislation sets out that the Government has to give us 10 weeks’ notice of when it requires us to produce a forecast for a fiscal event. Even in the absence of a set of policy measures to go through, this is a multi-week exercise that involves us having a lot of interaction with people at DWP and HMRC, whose own working year is set up in the knowledge that there are particular fiscal events to which they need to devote time. They have other things they need to devote time to.
As I was saying, given that it is a crowded year, I would assume that Parliament would make it clear to the Treasury that you would find that sort of analysis useful and that you might find it helpful to have the fiscal event prior to, rather than after, a judgment of that sort. That is something well above my pay grade as to whether that is possible in other elements of the parliamentary timetable.
We would approach it as we did in November 2016. When we produce a forecast, we do it on the basis of stated Government policy at the time, in a sense of genuine Government decisions, as distinct from aspirations. That means that if you have an agreement that is in the public domain, or will be by the time we publish, that is something we can take into account when we are preparing the forecast over that period.
We would have to know what the details were—but not the day before we produce the forecast, because we would have to have been working on it for that period of time. In the back of my mind was that this is presumably going to be one of the factors that determine when the autumn Budget is going to be, but that is a matter for the Chancellor, if that is how he is going to approach it.
Q7 Chair: But the forecasts don’t have to be tied to a fiscal event. You could be asked by the Treasury. We expect to ask Parliament to vote on the withdrawal agreement in October 2018, so counting back, they need to ask you by sometime in August to have enough detail so that you can then start working on the implications for your forecasts of the terms of the withdrawal agreement.
Robert Chote: The practical ability to do three forecasts in one year is pretty difficult—less for us than, as I say, the group of other people who mark out time in the year basically to crunch this. If we were the Congressional Budget Office or the Dutch CPB, we could do the whole lot in-house and cancel other things and arrange it. That is not the model that we have here; it is a more parsimonious approach to this, as you know. That is the sort of constraint in which you are operating.
As you say, in a sense it does not have to be tied to a fiscal event. We have just had what I’m never clear is a non-fiscal event or a fiscal non-event but without policy measures, so we have now moved to having one forecast that is disjointed from a fiscal event and one that accompanies one. But there are decisions to be taken in the autumn around things like envelopes of the spending review and so on. Brexit will doubtless be in the Chancellor’s mind in deciding that timetable and probably won’t be the only thing in his mind.
Q8 Chair: I’m sure. In terms of resources, we have the Chancellor come before us. Would this be helpful? This is perhaps not for you to answer, but I will put it out there. We might well ask the Chancellor about giving you additional resources and time in Government Departments’ diaries to help you with this particular added burden of a forecast.
Robert Chote: That is very generous. That is easier said than done, in terms of the experts in particular bits of DWP and HMRC who are cheerfully trying to get on with getting universal credit sorted out and so on. Their hearts may not leap at the thought of being badgered by us for another few weeks.
Q9 Chair: Or at the thought of another forecast. Would it be helpful if I wrote to you, so that you were able to set out thinking on timetables and that sort of thing?
Robert Chote: Yes. From our point of view, you will have a much clearer idea than I do about what the Brexit parliamentary timetable is. Clearly, you have an interest in that as does the Treasury. It would obviously be helpful to us to know, for example, whether that points to an early or late Budget. These decisions are not always easy to take well in advance, but greater clarity would be welcome for everybody.
Q10 Chair: Before I bring in Charlie, I just want to turn to Graham, because you have served as the fiscal expert on the Budget Responsibility Committee since the OBR’s inception, but this is the last time that we are going to see you in that role—I can see you are slightly demob happy already.
The relationship between this Committee and the OBR is an extremely important one, so I have been invited to ask you to turn the tables and perhaps give you an opportunity to reflect on how the Committee could do a better job in scrutinising the work of the OBR. Of course, you are welcome to say that it is all perfect and no improvements are necessary, but is there anything in particular from your experience in the way that the Committee has scrutinised the work of the OBR?
Graham Parker: The only thing I might say is that I’ve always been a bit disappointed that there has not been more concentration on the actual public finances, which is what we are basically set up to do—to produce the public finance forecast. We spend at least three quarters of the time at these sessions talking about the economic forecast, which we do only as a means to produce the public finance forecast. That’s my only comment, but it’s a question of what matters for you that clearly dictates that. But sometimes we could do with a bit more questioning on the public finance forecasts themselves.
Chair: Thank you. We will try to live up to that.
Charlie Elphicke: Turning to the public finances—[Laughter.]
Robert Chote: I think Graham will be taking this question!
Q11 Charlie Elphicke: In evidence to the Committee in November, you noted that the fiscal impact of the Brexit financial settlement “would be dwarfed by the consequences of Brexit, positive or negative,” which obviously would have a significant impact on the public finances. You didn’t make a judgment about whether it would be positive or negative for growth, so would you say that there are opportunities that could lead to higher growth after we leave the European Union?
Robert Chote: On the growth context, Charlie may have more views on this, but yes. We have made, as you know, a set of broadbrush assumptions that had impacts on the growth performance before. Those included things like impact on business investment and therefore on productivity growth via that route, and migration, with a possible change there. There is obviously a broader discussion, which we had last time, about potential links between openness and productivity growth.
There are then other issues about whether there are particular policy changes that the Government—any Government—may feel that they were constrained from making by EU membership and that they could make different choices. Clearly, opportunities to change policy could move in a more or a less growth-friendly direction—
Q12 Charlie Elphicke: In other words, we could deregulate and become more competitive, and have faster-trend growth, rather than disappearing into the economic slow lane with the rest of the European Union.
Robert Chote: Or you could have greater use of the absence of state-aid rules and in the same way it might lead you in the other direction. From our point of view, it is a question of seeing what concretely emerges in terms of those policies, and it could take things in either direction.
Q13 Charlie Elphicke: Just to be clear, let’s say we go for a relaxation of state-aid rules and go for nationalisation. Do you think that would increase or reduce economic growth?
Robert Chote: It would depend on what you were using it for in specific circumstances. As I say, you would have to look at the particular sectors. If you had a situation where you were simply doing a lot of intervention in order to keep companies in the UK, which wouldn’t hitherto have been possible, that wouldn’t be a nationalisation but it would be something that might happen.
Q14 Charlie Elphicke: But if one went down the route of nationalisation, does the history of economics show that that increases or reduces competition, productivity and growth?
Sir Charles Bean: I hesitate to generalise, but it is probably fair to say, on balance, that economies with large nationalised sectors have performed less well. But that’s not to say that selective intervention may not be effective in some areas.
Q15 Charlie Elphicke: Moving on, the OBR has to produce a fiscal sustainability report that looks at the public finances over a 50-year time horizon. Are we likely to see a more detailed analysis of the long-term fiscal consequences of Brexit under different assumptions in the forthcoming FSR, which is due in June?
Robert Chote: Not based on different potential policies, for the same reason that we haven’t been doing that on that basis before. One interesting question with the long-term projections is that, obviously, a key focus in terms of the impact of Brexit, or indeed any other policy, on long-term growth performance is what it implies for underlying trend productivity growth. That, oddly and perhaps counter-intuitively, does not show up as much as a fiscal issue in our long-term projections, because of the general assumption that we make that the parameters of the tax system and the welfare system over the long term are linked to earnings growth rather than to inflation. If you have a more pessimistic long-term assumption about productivity growth and potential GDP growth, we are all poorer, in terms of both private incomes and the amount of money that can go into public services, but it does not show up as a fiscal problem in terms of a wider gap. What we did in our fiscal risks report was illustratively to say, “Let’s allow a weaker outlook for productivity growth and potential GDP growth to hit revenues but, for the sake of argument, leave spending where it is,” which obviously you wouldn’t do over that period of time. Then it does manifest itself as a fiscal problem.
Q16 Charlie Elphicke: Turning to the kind of relationship we will have with Europe after we leave, would you say that, conceptually, what the Government are seeking from the future relationship with the EU lies somewhere between an EEA-type scenario, with single market membership, and a full free trade-type scenario of access to certain sectors and barriers in others? Is that where you are on that?
Robert Chote: We will wait and see where we are when there is an agreement at the end of the day. The Government, I am sure, have their negotiation strategy designed to deliver a preferred outcome, and the outcome that comes out of that may or may not be the same. It’s for them to judge what they are aiming at, and it’s for us to see where we all end up and to take that into account accordingly.
Q17 Charlie Elphicke: Yet when it comes to trade, you have been making certain assumptions—in particular, a decline in openness. On what do you base the idea that there will be a decline in openness when the Government have said that we want to be open to the entire world and increase our global trade?
Robert Chote: That was based on a series of analyses that were done prior to the referendum—I think specifically by the LSE, the OECD and NIESR. Each of them was basically taking the view that you were more likely to see a decline in trade intensity rather than an increase, because the likely impact of any reduction, or rather of increased frictions in trade with the EU, was likely to outweigh in its economic impact what would happen elsewhere, which is consistent—not necessarily, of course, gospel truth—with the cross-Whitehall study.
Q18 Charlie Elphicke: But these are pre-referendum studies, aren’t they?
Robert Chote: Yes.
Q19 Charlie Elphicke: And also cross-Whitehall studies and all the rest of it. The pre-referendum economic forecasting was heavily criticised and described in certain quarters as “Project Fear”. Do you think it is important that we have more recent studies that look at what the Government are actually saying in terms of the policy they want to pursue, which is not to reduce openness?
Robert Chote: I would distinguish between the pre-referendum analysis and the broad directions. Although the numbers will be slightly different, the cross-Whitehall study, which is cross-departmental, has come out with the same provisional conclusions on direction as the Treasury study—it is not the immediately pre-referendum “We’re going to have an emergency Budget” study, which history has not smiled kindly on, but the earlier one, which Charlie was obviously involved in at the time. While there were lots of things that reasonable people might disagree about, the central conclusion, based in part on the widely accepted gravity approach, was that until you see evidence to the contrary, your best assumption at the outset is that you are more likely to see an aggregate reduction in openness rather than an aggregate increase. It was striking, and this is something that we would want to look at. I was struck by how little offsetting impact trade deals outside the EU had in the cross-Whitehall study—that is not to say that I am disagreeing with it, but I was struck.
Sir Charles Bean: Absolutely. The likelihood is that leaving the EU will lead to increased barriers there and a reduction in trade intensity. The question is how much. Presumably, we want to minimise any associated costs and at the same time maximise the advantages from being outside. Where we end up obviously depends on how the negotiations go not just with our partners, but with other countries. We would not want to put ourselves in the position of trying to take a view about what the end point is going to be—there are too many moving parts that we are not qualified to talk about.
Robert Chote: You raise an important point about the issue of opportunity. It is not as though once we have this withdrawal agreement the position is then abundantly clear. There are then opportunities, which could be positive. But they could also be negative, both in terms of the changes in economic policy that might be made as a consequence of this and the potential trade agreements. In years to come when hapless PhD students are trying to unravel the impact of this in 20 to 30 years’ time, you have to say, “Well, okay, would this particular intervention have happened in the absence of the Brexit decision?” In disentangling this, people are going to come up with hugely different answers, even after the event, to what impact any of this has had.
Q20 Charlie Elphicke: Let’s have a look at trade intensity. There is an assumption out there that there will be a reduction in trade intensity. I want to challenge that on two grounds. Limb A: the European Union has £100 billion trade surplus with the United Kingdom which they would not necessarily want to lose. They would not necessarily want to see a reduction in trade intensity from their side, because they are the winners.
Limb B: they nevertheless act in a manner that one might describe as economic self-harm to reduce trade intensity. What is the opportunity in the UK economy for the generation of increased internal trade to compensate for the reduction of goods coming from the European Union?
Sir Charles Bean: The key point here is if you put barriers to trade between countries, you reduce the extent to which they can specialise in areas where they have comparative advantage. It may well be the case that post-Brexit we have exactly the same net trading position with the EU but at the same time we export less to them and they export less to us. That would, in principle, be likely to involve a less efficient allocation of resources to production.
Q21 Charlie Elphicke: Although the alternative is that the converse could be true.
Sir Charles Bean: It is not obvious why the converse should be true. You are putting an obstacle in the way of the efficient regional allocation of production.
Q22 Charlie Elphicke: Let’s further examine this obstacle. We have heard the Government talk about frictionless borders. I myself represent Dover. I produced a blueprint as to how we could have electronic borders for the digital age, rather than analogue borders for the 1950s, and look to the future. We can do fast turnaround. There is no reason why trade should be impeded if both sides are on that side.
Secondly, the trend in the world is for global standards rather than regional standards. If we end up moving towards global standards, if we are able to work positively with our European partners and have frictionless borders, would it not be the case that a reduction of trade intensity as an assumption may well turn out to be false?
Sir Charles Bean: No. This absolutely goes back to a point I made earlier. Presumably the aim when we leave is to minimise the cost, and precisely if we have a good trading relationship that approximates as closely as possible to what we have at the moment, that will do least damage to the efficient regional allocation of production.
You are quite correct that that would imply a smaller if not negligible reduction in trade intensity. The key is obviously getting a very good deal that enables good access to our existing markets, and then taking advantage of new opportunities in trading with other countries.
Q23 Charlie Elphicke: Let me focus on that in my last question. Is it not the case that, if we went through a disruption of supply chains, which are quite intricate across Europe, it would cause harm not just to the British peoples, but to all European peoples, and the logic would militate against either side going down that route? It also enables the United Kingdom to have a focus, as you put it, on expanding our trade elsewhere, which could leave to a higher trend growth over the medium to long term.
Sir Charles Bean: It is certainly true that putting any obstacle in the way of trade is mutually damaging. When we talk about supply chains, it is like putting a brick wall in a factory that people have to step over each time they transfer work from one side to the other. You would want to make that wall as small as possible. It will certainly damage our trading partners as well, and that is why there is every incentive to have a sensible, good trading relationship, but it is quite possible that we will have something that falls short of what we have at the moment.
Q24 Chair: I promised that we were going to move on to public finances, but I have one final Brexit question. The Treasury recently disclosed that the European Commission is taking infraction proceedings against the UK in relation to exchange-traded commodity derivatives. I wondered whether the OBR had been made aware of those proceedings.
Graham Parker: Is that the one about customs duties?
Q25 Chair: No, I don’t think so.
Graham Parker: Then the answer is no.
Q26 Chair: Okay. Presumably that constitutes an unquantified risk to your forecast, and it sounds to me like that ought to be discussed.
Robert Chote: I am not sure it is the same, because—if you have the book in front of you—4.207 was a legal proceeding that we were notified of relatively late. That is the one Graham is talking about.
Q27 Chair: Is that the shoes?
Robert Chote: That is a different issue, is it?
Chair: Yes. So the answer is no. Apparently Alison has one Brexit follow-up, and then we will get off it.
Q28 Alison McGovern: It is very brief. Robert, you said at the beginning of your remarks on the issue of trade intensity that both your forecasts and the Treasury forecasts relied on a widely accepted gravity approach to the model. Could you just say what that is and why it is widely accepted?
Robert Chote: I will hand over to Charlie, who will give you a much more sensible answer, but it is basically the idea that countries tend to trade more with larger, richer countries that are closer to them. That is my knowledge of gravity.
Sir Charles Bean: That is a pretty good description, but it works extremely well as a model describing the empirical allocation of trade between countries. The gravity model fits the data very well in explaining how much country A trades with country B.
Q29 Alison McGovern: Members of our House who are not in this Committee have criticised the gravity approach to modelling trade. What would you say to them?
Sir Charles Bean: As with any model, there are always shortcomings. I think the view in the academic world among trade theorists is that the gravity model is still a pretty good description. Obviously, how you may choose to model interventions in that such as trade barriers can affect what comes out, but I certainly would not accept the view that it is a discredited model that one should not take any notice of. I don’t think that is true. The thing about models is that you should never think of them as being suitable for all possible questions that you may be interested in. The gravity model may be good for looking at some questions and less suited to looking at others.
Robert Chote: The contra-response to some of the critiques of those models is that they assume that the pro-trade position is less costly, and therefore you would just get an awful lot of trade moving, so you would have a dramatic restructuring rather than things happening at the margin, which the gravity model would be. For some of the big effects that you get as a positive impact out of this would imply very big shifts in resource allocation and trade destinations, whereas this model suggests that “big, close and rich” has quite a powerful impact, which you would presumably combine with inertia as well to suggest that things would not move very quickly.
Q30 Stephen Hammond: Gentlemen, good morning. Perhaps we can now genuinely move on to public finances, rather than a back door to something else. In your 2017 fiscal risk report, you highlighted the numerous examples of fiscal illusion, which on the surface improve public finances but in reality do not. That is not new, is it? I co-authored a paper on that over 10 years ago called “The real balance sheet”, looking at the Gordon Brown fiscal illusions. From that come a number of questions. First, can you say how long fiscal illusion or on/off balance sheet financing has affected the UK public finances? To what extent do you think that scale has changed over the last decade or two decades? Why did you include the concept so heavily in this report? I suppose I am asking about the scale now. Has the scale changed over the historic period, or currently looking forward?
Robert Chote: That is an interesting set of questions. As you say, one reason for including it is that these issues have been around a long time. It is not unique to the UK—I think the IMF first used the phrase “fiscal illusion”, and it points to that in a variety of different contexts. The underpinning idea concerns the way that even official international-standard measures of the public finances are constructed. There are sometimes policy decisions that you can take that will make those numbers look better, but with a take-one-step-backwards view of the underlying health of the public finances, that is not the case.
Judging whether the scale overall is increasing is pretty difficult because there is such a mix of different sorts of illusion, and the mix of those things has changed over time. People did get a lot more excited and went back into the period you were talking about with off-balance-sheet PFI-type approaches. We wrote various reports on that, and when I was back at the IFS, it was generally reckoned that you would see perhaps 3% of GDP addition to net debt if those various things had been done on balance sheet. At the time that seemed enormous and something to be hugely worried about, but obviously in the wake of the financial crisis, 3% of GDP didn’t seem as big as it was.
Another point that comes and goes is asset sales. It can never be repeated too often that, if the Government sell an asset for roughly what it is worth, they are swapping one asset for another—they swap the flow of income that they would have received from the asset in the future for an up-front cash sum and a measure of public sector net debt. They get the benefits of the sale but do not see the loss of the receipts. The liquid assets will be counted and illiquid ones will not be—that is important.
The issue that you would point to, and that we noted here as growing, is student loans. For example, the contribution of interest on student loans to the Government’s measured interest and dividend receipts is a little under a third in 2016-17, and will be in excess of 50% by the end of the forecast horizon. Not all of that interest will actually be received, because some of it will be written off under the terms of the proceedings. There are even some of what the cognoscenti would regard as being better balance sheet measures than public sector net debt—for example public sector net financial liabilities, or PSNFL as it is occasionally known, encapsulates a wider variety of financial assets and liabilities and therefore gives a bigger picture. However, that also has the flaw that student loan debt is basically taken at face value and not the amount that will be repaid.
If you had to point to one thing that is on an upwards trajectory, that is probably it. Saying whether overall fiscal illusions are a bigger or smaller problem is harder.
Graham Parker: If you are talking about going back in history, there is scope for this. Before probably 1997, the main deficit aggregate was the PSBR, which was about cash. Student loans would have scored against that because it was to cover the whole borrowing requirement—how much cash did the public sector require? There was not this distinction between spending that scores against net borrowing and loans, as there is now under current aggregates. Clearly, it has mushroomed since then.
That of course coincided with the introduction of fiscal rules. Whatever fiscal target you have, if it is expressed in a deficit that is measured in one way, you may find the temptation to look at ways to do other things.
Stephen Hammond: Sure. That leads neatly to some of the questions that follow on from that answer. I take the point about the IFS saying 3%. I think there were a range of numbers, some of them considerably larger than 3% in terms of if you took the whole local authority debt back on.
Robert Chote: Yes, that was basically just saying that if you had borrowed through conventional terms to do this and spent in that way, that is what you would get.
Q31 Stephen Hammond: The point that people were making at the time, of course, was that PFI schemes appeared to be local recourse when in fact they were national recourse and the number was a lot bigger.
Three questions follow from your answer, Robert. First, do you think there is any suggestion or evidence that Governments have taken fiscal decisions for presentational reasons? Secondly—
Robert Chote: Do you want to take these in order?
Q32 Stephen Hammond: Okay, let’s take them in order. That is question one. It clearly follows on from what Graham has just said about cash and the fiscal rules.
Robert Chote: On the issue of whether policy has turned out differently, you need only look as far back as the November report, when we looked at the decisions that were taken to get housing associations back out of the public sector again. The ONS put them in on the basis of a judgment about the degree of control that Government had. In a written answer in evidence to the House of Lords, the Government was explicit in saying, “We did just enough, and exactly what needed to be done to get this reclassified”. It was not a broader discussion about whether this was the appropriate regulatory regime for the housing association sector. With a fiscal institution, it is a process of taking a step back and asking, “Has the big picture really changed? Do we think that Government policy is less important in the housing association sector that the Government would be less likely to intervene if the players in this sector got into trouble?” That would be a good example. You could also point to earlier things like Network Rail, where there are fairly clear footprints in the sand to show that decisions have been—
Q33 Stephen Hammond: Of course, Network Rail has come back on the balance sheet.
My second question was about whether you thought student loans actually had a more risky profile for the balance sheet or the debt borrowing position than PFI did 10 years ago. Do we think the scale of what we are doing now in terms of fiscal illusion is of greater risk than it has been historically? Is there any way you can assess that?
Robert Chote: On student loans, the magnitude of numbers is larger. We are looking at student loans adding something like 11% of GDP to net debt at peak. Then there would be a question mark over whether or not that is an appropriate number. In that sense, it is larger than the 3% of GDP, but you are comparing an apple with an orange.
Sir Charles Bean: The risks on some of the PFI projects would be much harder to quantify than the risks associated with the student loan book.
Graham Parker: The other thing is that at least the Government resource accounts do actually quantify the effect of write-downs. Even if they are not scored in the public finances yet, the Government’s accounts do show the estimated level of write-downs and loans in future, so there is some recognition of the risk there.
Q34 Stephen Hammond: My last two questions are more technical. You make the point about the asset sale, and clearly there will be an interest cost or an interest receipt change position. How does that show in the accounts? Is there any way of showing that, if we sell that asset, there may be a declining—or accelerating—set of costs or receipts against it?
Secondly, what would you recommend the Committee to say about rule changes to national accounting so that fiscal illusion is more apparent and less driven out from the numbers?
Robert Chote: Maybe I will get Graham to answer the first question. There is a value, obviously, in observing the international standards. You get into deep water in moving away from internationally accepted standards, even if with some of those internationally accepted standards you say, “Well, hold on, this isn’t putting the fuller picture”. We see it as our responsibility to point these things out when we see them and to be as clear as we can when policy changes take place that may or may not be the result of the Government gaming these things at the border of what counts in a particular variable that they care about, as we did with housing associations. We point it out, for example, if the Government takes a whole series of decisions that just ensure that debt falls as a share of GDP in 2018-19. We show that. That is a key second strand of our role.
Our role is not merely our formal function of telling you, on the basis of our forecasts, whether the Government is going to hit its targets, but explaining how it is hitting those targets, or how it is hitting an informal target that someone boasted about in a speech last time and is desperately keen still to be able to boast about in the future. Where are they stamping on the lumps in the carpet, and where is the one remaining lump in the carpet that just gets larger and larger that they would probably rather you didn’t look at it?
Sir Charles Bean: Can I add something to that? It would be a mistake to think that there is some perfect fiscal measure out there that will be robust to all these sorts of shenanigans. As soon as you home in on one measure that is robust to a particular problem, you will almost certainly open it up to abuse in other respects. I do not think the solution is thinking about national accounts definitions, changing targets or anything like that. The key thing, precisely as Robert says, is that watchdogs like ourselves are active in calling out when an action is being taken essentially to exploit the way a measure is defined rather than for its own justification.
Graham Parker: On the other question, we do change our forecasts. There is a line in the receipts forecast called, “Interest and dividend receipts”, so when there is an asset sale we will—
Q35 Stephen Hammond: Is there usually a note to the accounts to explain that number?
Graham Parker: If there is anything going on in that number, we will explain it in the bulletin. There is certainly a line that says, “Interest and dividend receipts.” Obviously, when Lloyds bank shares were sold, because Lloyds had started paying dividends, we took those out of the forecast. But you don’t get it all, because we only have a five-year forecast and, clearly, you are selling an asset that would have given you a much longer-lasting stream of interest or dividends.
Q36 Stewart Hosie: Robert, Government policy can only reduce risks; it cannot eliminate them entirely. In your view, are there any areas in which the Government are not taking adequate steps to reduce long-term fiscal risks?
Robert Chote: As I say, at the end of the day, what the appropriate balance is in a particular area is a decision for the Government. As you say, part of the purpose of the Government is to take on risk and to transfer it in that sort of way, and it is rarely optimal for the Government to eliminate risk entirely even if it can do so, because there are consequences that flow from that.
If you think, for example, about the sorts of things we have been looking at here that the Government have started to respond to, in the fiscal risks report I was most struck by the greater sensitivity of the public finances to interest rate and inflation shocks as a consequence of the change in the size, structure and composition of the debt stock. It is interesting that one of the few eventful bits of the fiscal non-event was a decision by the Government to reduce the proportion of debt being issued in index-linked form, which I presume is partly a response in that sort of direction. The other way the Government would say that it has improved—but only time will tell—is by having a more effective gateway and process of thinking carefully before you take on new contingent liabilities. That is an important area. There are obviously some legacy ones, like nuclear decommissioning costs. There is a wide variety.
In terms of the long-term outlook, clearly a major risk is what you will end up having to spend on health and social care over the long term. If you are looking for the single largest fiscal number that would come out of that, it is not what anybody would describe as a fiscal risk. It is essentially a risk in the form of a policy decision that Governments may or may not feel the need to take. If you are in a world where Governments in future feel that they need to see health spending increasing, consistent with accommodating the effects both of an ageing population and, more importantly, of non-demographic cost pressures, that has some serious implications for the decisions that you need to take across the rest of public expenditure, across tax raising, and across your fiscal targets.
It would be inappropriate to say that this is something that they have not addressed, as though they should have addressed this now. It is something that you address over time as you are making fresh spending decisions. However, it is the largest number that is out there.
Q37 Stewart Hosie: You said in the fiscal risk report that “there seems to be a very high risk that health and adult social care spending will be on an upward trajectory as a share of GDP due to demographic and other cost pressures.” What do you think would be required in the 2019 spending review to address those pressures?
Robert Chote: You can address it in two ways: either by addressing the pressure directly, or by adapting other policies to the decision to accommodate that pressure. I think the Government would say that there will be scope to increase productivity performance in healthcare, which will allow you to get a greater output for less money, and that that would allow you to put in place spending settlements for health that would not rise as quickly as the numbers that we have sketched in the risks report.
The alternative response, though, is to say, “Okay, we accept that you will have to spend that additional amount of money.” Hence, you have seen, for example, some discussion about whether you should have some sort of dedicated or, at least rhetorically, hypothecated change in national insurance to provide some additional resource. Then there are, of course, the range of other decisions that you would make in other areas of public spending and taxation. When you get to the spending review, it is bringing those things together into a coherent story that is the challenge.
Q38 Stewart Hosie: All other things being equal, and in the absence of hypothecated NI or further cuts to other spending Departments, is the projection for spending, given the increased demand in costs in this area, consistent with eliminating the deficit by the mid-2020s?
Robert Chote: It would make it more difficult rather than less so. By the end of the current forecast horizon in 2022-23, we still have a deficit of 0.9% of GDP. As you know, our forecast does not extend as far as the 2025 or thereabouts that is implied by the vote on the fiscal objective in the charter on fiscal responsibility. As in chapter 5 of the report, if you were to accommodate those sorts of pressures, you are more likely to see upward pressure on that deficit, rather than it making further progress towards that balance objective. In that sense, it would make things harder.
Clearly, even up to 2025 is a relatively short time horizon, and when we are thinking in the risks report and in the fiscal sustainability report about non-demographic pressures, you are talking about big, stylised judgments over a number of decades on how much of that would or would not manifest itself in any particular five-year period. That would be very hard to say—this is something where you are looking more over a decade.
Q39 Stewart Hosie: Let’s roll those last couple of things together. You mentioned that they could increase productivity within that sector, and that that would ease some of the pressures. That makes sense, but let me ask a question in relation to delivering increased productivity in health and social care, and the Government’s long-run fiscal sustainability targets. How can they be met if there is a massive squeeze on net migration? If those who work for modest wages in the sector no longer come in, and if the contribution from migration is no longer there at all, is that not an example of how the targeted reduction in migration will impact not only on the health and social care sector and its future productivity, but on fiscal sustainability more generally?
Robert Chote: There is clearly an issue there. This would not just be for net migration, but if you did something that increased the cost of providing a given unit of healthcare output—it could be the fact that the workers you need to employ to do that are more expensive than they otherwise would be, but it could be for other reasons as well—then that is sharpening the trade-off you have to make about whether you want to spend the money to accommodate this or whether you want to provide less output or take care of it elsewhere. That would be an issue. As you say, overall our assumptions are that lower net migration is a net negative for the public finances over the sorts of horizons we look at primarily, because of the fact that net inward migrants are more likely to be of working age. That is a separate issue to the specific pressures you point to in health.
Q40 Stewart Hosie: That is an interesting point that I was going to ask about. You have confirmed that it is still a net positive. Can you envisage a situation at all where a more restrictive migration regime being applied to EU migrants could ever be positive for the public finances?
Robert Chote: There is a possibility that it could be a trigger to greater productivity. If labour is more expensive or particular categories of labour are harder to get hold of, that could encourage firms to innovate in ways that increase their underlying productivity. That could have longer-lasting positive effects. It is conceivable.
Q41 Stewart Hosie: It isn’t likely, is it?
Robert Chote: Well—
Sir Charles Bean: It’s possible, like a lot of things.
Robert Chote: You can see the assumptions that we have made historically. If you look back at the scenario analysis that we did in March 2016, that looked at the impact. We picked one of the ONS’s available scenarios for population dynamics based on particular migration assumptions. If you moved from the central one to the high or low migration variant—it makes a difference of about 65,000 a year by the end—it made about £6 billion of difference to borrowing, if I remember correctly, at the end of that forecast horizon.
Needless to say, there are huge uncertainties around this. The underpinning assumption we make is that in things like productivity and their employment rate for gender and age, net inward migrants are like the rest of the population. If you were to have changes in migration policy that changed the nature of the net inward migrants relative to the characteristics of the rest of the population, that could turn out differently. Once again, we bring out the broad brush.
Q42 Stewart Hosie: I have a final question on an unrelated subject. To what extent will demographic change affect the public finances in the medium term—the next five, 10, 15 or 20 years?
Robert Chote: With pension expenditure, the link to policy means you are seeing more upward pressure there because you have had a period in which you have had the pension age rising. The demographic effect is having more upward effect on pension spending over the next five years than it did over the past five because it was offset by policy. Is that correct?
Graham Parker: Yes, there will be another pension age rise in the next five years.
Robert Chote: But it is more fiscally unhelpful—the combination of the demographic change and the associated policy change over the coming five years.
Q43 Stewart Hosie: I understand from what you have said previously that moving away from the triple lock would have no discernible impact on the fiscal numbers, certainly in the short term.
Robert Chote: Moving away would have no impact?
Q44 Stewart Hosie: A modest impact. I think that is what was said.
Robert Chote: Modest would be fair over the medium term, yes. Having the triple lock over the longer term essentially ratchets pension spending per recipient higher as a share of GDP over time. The time profile with which it does that depends on when, over quite a long time horizon, you have periods when real wages are falling and when it is more often the inflation rather than the 2.5% that is the biting increase, while earnings growth, which is more likely to reflect what is going on with GDP growth, is weaker. That tends to ratchet up.
Whether it happens in any particular five-year period depends on what the particular outlook is for real wages over that period, basically. The way we model it in the long term is by making a simplifying assumption of pension rise by earnings plus X every year. The one thing that you know is that it will not be that smooth a pattern. That may be the average pattern over time, and that is what we are trying to get at, but actually it would take place in chunks of time in which you had real wages falling.
Q45 Alison McGovern: I have some further questions on public spending. In estimating the impact of changes in fiscal policy on growth, the OBR uses broadly the same multipliers as when it was established in June 2010. Do you think that that is right? Do you think that changes in the fiscal position still have a consistent impact on the economy?
Robert Chote: I will ask Charlie to say more, but when we made that judgment at the outset, it was based on a review of the literature at the time. Again, as with Brexit, there is a range of views and we have to make a judgment on them. We made the expected judgment that there are some things for which, whatever the multipliers and the size of the impact, you could have a reasonable expectation that it is probably higher for capital spending than for bog-standard public services spending or tax changes, simply because of the likelihood of the money leaking out into imports or saving. The other judgment that you have to make is how quickly these things decay. Over the long term, you assume that the effect comes out to zero.
The other issue, when we are looking at it in a particular set of circumstances, is that if you announce a change in spending that is some way in the future, it is less likely to have an effect because we assume that monetary policy will have taken it into account by the time we get there. We look at all those considerations afresh each time, but the numbers have not changed. We have not felt the need to change them.
Sir Charles Bean: No. Clearly if there were compelling empirical evidence to suggest we ought to change them, we would. I think it is worth emphasising the point that Robert finished on, though: multipliers depend on a range of other supporting assumptions, like the nature of monetary policy and whether it is in a position to offset any inflationary impacts—or not, as the case may be—as well as things like the exchange rate environment. We should not think of the multiplier as a fixed number that applies in all circumstances, but for a given set of assumptions about surrounding policies and so forth, I think the answer is that we have not been confronted with evidence that suggests that the numbers we have at the moment are inappropriate.
Robert Chote: In recent years, there has been particular concern about whether they are right, at a time when you are at the zero lower bound and there is not much scope for monetary policy to respond. We are now in a world where monetary policy is tightening again, so I suspect that some of the people who would have had a particular concern over some of those years have now gone back to thinking that maybe the multipliers will be relatively small because the Bank of England is in an environment where it is likely to offset a medium-term fiscal loosening.
Q46 Alison McGovern: We will come back to monetary policy in a second, but I just want to pick up on what Robert said about your assumption that capital expenditure has a particular impact and bog-standard public spending has another. I imagine that the answer is probably no, but do you think that you should be involved in the review that the Chancellor announced of the place of education spending and whether it is bog-standard public spending, or whether it might be investment in human capital?
Robert Chote: That’s a very interesting set of issues that has been around, for example, the discussions of the merit of golden fiscal rules. It is a slightly different question from the multiplier impact. The discussion you referred to basically asks whether you should have a system that says—simplifying dramatically—“Investment: good. Day-to-day spending: good up to a point.” When you have day-to-day spending on things such as teaching, training and those sorts of areas, you are investing in human capital, and of course, people would debate whether spending on physical capital actually yields long-term benefits. That has always been a reason why some people say that you do not want to place too much weight on the capital/current distinction for fiscal rules in a fairness sense, on the grounds that you need to be wary of saying that capital benefits future generations but day-to-day public services spending is consumed by the current generation and does not have future benefits. But there are wiggles in both directions. The multiplier issue is separate. To come back to the other questions, I do not think that either of them are for us to pronounce on. We police the rules as the Government of the day decides to define them.
Q47 Alison McGovern: But presumably it might be something that you would take a view on at some point.
Robert Chote: Well, not on the—we are aware of the—
Alison McGovern: If we got to the point where the Government were going to change the rules, at some point you might have a look around it.
Robert Chote: Saying whether a particular rule is a good or a bad idea, or indeed whether it ought to be hit at a particular moment, is not the responsibility that you have given us in the legislation. Certainly, my former colleagues at the IFS could cheerfully burble on for hours about it. Indeed, you could dig out—
Alison McGovern: I am sure we will give them every opportunity, but—
Robert Chote: I spent eight years burbling to your predecessors on exactly these issues, not so long ago.
Q48 Alison McGovern: I will ask a couple of extra questions and then wrap up. At risk of drawing you into something that you currently do not have powers over, we have seen a lot of shift on the disaggregation of inflation and on the relationship between the fiscal position and growth. We are yet to see any of that sort of disaggregation, so we look at those numbers on a UK-wide basis, yet, clearly, the fiscal consolidation had a bigger impact in some locations in the UK than others. Have you any thoughts on whether the literature is developed, or needs developing, on what fiscal consolidation has done to growth at a local, regional or sub-national basis?
Robert Chote: Certainly, it is not our responsibility, because we are tasked with forecasting and projecting the public finances at a UK level. We have some additional responsibilities over the devolved Administrations for particular areas of devolved tax, but we are not supposed to be producing a Scottish budget deficit forecast, although we look at the impact of particular measures there.
I am sure work has been done on the impact of the fiscal consolidation on a regional basis. If you were looking at it in a macro but regional sense, one would say that there is an issue about the quality of the disaggregated GDP by region numbers. You may also be thinking about a further level of disaggregation to particular public services and the pressures in particular areas, which I am sure people have done that work on, but it is not our bailiwick.
Q49 Alison McGovern: Just one final question. In your public spending forecasts, you revised the profile of tax litigation since November. Why?
Graham Parker: We start off by looking at what is in the HMRC accounts for their visions, mainly, against their accounts—for where they expect to lose money in court cases. Since last time, they told us that they expect one case to not actually be settled as quickly as possible, so we have pushed that back. There are other complications as well. We are back to national accounts scoring again, I am afraid. These things only score as spending in the national accounts when there is the very final settlement. HMRC will pay out well in advance of that in most cases if they lose at one stage of the appeal process, but it will not appear as spending until the last stage of the appeal is completed and ONS have paid out the whole of the money. There is a big time lag for that to happen and it is very uncertain, so apart from for the first year or two we have just taken the total and spread it evenly over the last three years of the forecast.
Q50 Alison McGovern: Is that the right way? Does it have any broader impact?
Graham Parker: I don’t think so. I can’t think of any broader impact. Again, we are back to this: when they pay out any money it will score as a financial transaction, so it affects net debt when it is paid out but does not affect the deficit until the final settlement. But these are fairly early days of this kind of arrangement. All we have done is re-profile things.
Robert Chote: It is an area we looked at a bit in the fiscal risks report. It is the sort of thing where there might be a whole series of cases that HMRC thinks there is a less than 50% chance of losing, but it makes sense, looking over a long period, that they are not going to win every one of them. Therefore, having some profile of an expected amount that comes out as a result is a sensible thing. What can move it about though is that this can be quite sensitive, in particular if they win or lose a leader case that has consequences for a whole lot of follower cases. It is then not simply the case in question but the fact that you assume that it will result in either a whole string of other actions being taken or actions that you had anticipated being taken not being taken.
On Graham’s point, working out when all of this would be completed to the point at which you would score it is difficult. The underlying significance is, I think, that we assume that there will be some flow of spending throughout the forecast, and it is sensible to assume that even if HMRC were to say, “Well, we’ve got a whole lot of cases on the go. Our best guess is that there is a less than 50% probability we’ll lose any of them”, you would still have a number in there. The second thing is that it is not a colossal amount of money but it is an important source of medium to longer-term risk in the sense that if you have a big case that, for example, has a lot of follower implications, it can change the profile significantly.
Q51 Rushanara Ali: Moving on to local government finance, in your November forecast you assumed that local authorities would draw down about £1 billion of reserves and now you are assuming that it will be £1.5 billion. That is a significant increase. Can you say something about what you think is driving that? To what extent do you think the statutory duty to provide housing for homeless people is part of it? What other factors are driving it?
Robert Chote: I will hand over to Graham in a moment. In looking at what we have done on the local authorities spending and receipts thing, there is an important distinction, but we got local authority spending higher. That is partly because local authorities are now being allowed to raise council tax by more and so the increased spending includes the receipts component of the forecast change, which washes out in terms of its impact on the deficit. Then, as you say, there is the judgment we have made that local authorities are likely to underspend their budgets by less than we had anticipated, that they will draw down more from their reserves in the current year, that they will continue to draw down those reserves for one further year than we anticipated back in November and that there is a bit more of that happening. Let me hand over to Graham.
Graham Parker: There is not much more I can say. The thing about local authority data is that there is not that much available in-year. We have tried to look at what is going on, and the data we have seen on non-education spending suggests that local authorities are spending more than we thought they were going to. There is a very interesting NAO report on this that came out just before the spring statement, which we have not seen but which confirms the kind of analysis we were looking at: a lot of councils are now overspending on their service expenditure, particularly the counties and the single-tier authorities that have social care responsibilities. So an awful lot of the overspends appear to be on social care, both adult and children, and it seems to be continuing. That is the basis for why we increased our spending this year.
Q52 Rushanara Ali: How concerned are you about the increase in spending and the use of reserves in a way that is quite different from what has gone on in the past, and the ongoing reliance of some authorities on the use of reserves?
Graham Parker: From an overall public finances perspective, we are cushioned by the fact that local authorities cannot actually go bust. In the case of Northamptonshire, they are just restricting what they can spend to a certain level. In a sense there is not a big risk to the overall public finances here. Clearly there are risks, but the risks are on the services that local authorities can provide. This will obviously change. They vary a lot across the country. Councils with big reserves can exist quite happily, you would think, by drawing down reserves year after year after year, but of course there are councils without reserves who will not be able to do that.
Q53 Rushanara Ali: Did you look at any specific data area? Is there a local authority by local authority breakdown or is that something that DCLG or MHCLG as it is called now does?
Robert Chote: We have a new logo.
Graham Parker: The only thing we do on this particular issue is to look at the GLA and its TfL subsidiary in particular separately. We look at the TfL business plan, because that has a big impact, and all the different investment programmes involved in it. They tend to move very differently from the normal local authorities. We try and isolate those separately. There is not much we can do with the rest.
Q54 Rushanara Ali: Would the Department be able to provide that?
Graham Parker: The Department has statistical returns by authority and lots of other information that they collect. We do not tend to look at that as a general rule. Sometimes on the capital spending side there are some very lumpy capital investments that we look at.
Robert Chote: The key thing is that it is not a bottom-up forecast. It is an overall judgment on the use of reserves, but informed by certain things. There are clearly pressures for those with social care responsibilities. We talk to experts who have an overview of the sector in trying to get a feel for that as well, but it is not bottom-up.
Q55 Rushanara Ali: Do you think it should be?
Robert Chote: From our point of view I don’t think you would get a sufficiently better answer. You might not even get a better answer at all from doing it that way.
Q56 Rushanara Ali: How many other authorities do you think could go down the Northamptonshire County Council route? Would it be helpful to have a bottom-up way of accounting to be able to answer that question?
Robert Chote: Frankly, given the amounts of money involved, that would not be a particularly useful question for us. It is clearly something where the Department will have an interest in whatever methods it uses to keep an eye on where things are going. The amounts of money are important for the people of Northamptonshire, but from a national fiscal perspective the fiscal risk at the national level is not the consequence of x number of additional authorities getting into a Northamptonshire position. It is when the Government next make a big spending decision and decide they need to put more money into the sector, so it is a policy fiscal risk as distinct from anything else.
Q57 Rushanara Ali: Sure, but the reason I am asking this question is to get a better understand of how many other authorities could be at risk. Obviously, we hope that would inform Government thinking and responses, to mitigate against others falling into similar kinds of circumstances, where there are consequences. They may not be for your organisation, but there are consequences to that happening.
Robert Chote: Indeed there are, and I am sure MHCLG and the Treasury are looking closely at the data that is available and the broader intelligence that they get, but it is not part of our core responsibility.
Q58 Rushanara Ali: A number of local authorities are starting to get into investing in commercial ventures. How much have you looked at the opportunities that that poses but also whether there are any risks posed as a result of those sorts of ventures?
Robert Chote: That is something that we have looked at in recent forecasts and it is something where CIPFA and the Government have responded.
Graham Parker: I think there are some new guidelines that come into effect for the next financial year, which probably will restrict opportunities for this kind of commercial investment. We have actually reduced our forecast for this kind of capital. You are right that in the last couple of years there have been some very big investments, which have an impact on our capital spending forecast.
Q59 Chair: I just wanted to go back to capital investments, to ask about the H in the Department’s name—housing. I think the OBR’s forecast is that the Government will fall well short of the 300,000 target for house building. The Committee recommended in our Budget report that the borrowing cap for local authority investment in housing stock should be removed. I wonder whether the OBR had a view on, if that were to happen, how far it might assist in meeting that target, or whether you have any other comment on the impact on local authorities’ finances that you wanted to put on the record.
Robert Chote: It is not. If that became a Government policy and they decided to proceed with that, we would look at that in detail, but it is not something that we have looked at ex ante.
Q60 Stephen Hammond: Gentlemen, I am sure that, like me, you have been re-reading your economics textbooks written by A. A. Milne. In his opening remarks, the Chancellor talked about being Tigger-like. He based that on some of the underlying strengths of the UK economy that he pointed out. Give that you have said that the outlook for the economy and for public finances remains broadly the same, is there anything to justify the Chancellor’s current change of character?
Robert Chote: We aim to be Owl-like, as distinct from Tigger-like or Eeyore-like, as best we can.
Stephen Hammond: I have lots more torturous analogies.
Robert Chote: It is always best to get them out.
Clearly, the judgments that we made in terms of the underlying growth projection were, importantly, changed when we did the November forecast. The biggest uncertainty over the medium-term outlook remains the underlying trend of the productivity position, where the decision that we took back in November was a reflection of both the continued weakness of actual productivity well beyond the financial crisis and the immediate impact of that, and the fact that it is a noticeably international rather than a purely domestic phenomenon. It is, however, a central forecast; there is a 50% chance in our view that it is higher or lower than that. It would not surprise me if some people take a more or less optimistic view. In terms of the next couple of years, we are not a million miles away from where the rest of the forecasting community is.
Sir Charles Bean: We are pretty similar in our judgment on potential outputs—say, to the bank—and those outsiders who have similar concepts of potential output. I would really stress that there is a lot of uncertainty, because we do not understand why productivity growth has been as weak as it has been over the past decade or a bit longer if you look at the US data. In the US, the slowdown seems to predate the financial crisis.
Q61 Stephen Hammond: There is some evidence to suggest that it did in the UK, too—there was certainly a movement around 2007.
Sir Charles Bean: Yes. The thing is that the UK picture is clouded by some measurement issues connected with how you measure value added in the financial sector, which makes it appear that productivity growth was quite rapid in the pre-crisis era, but arguably the way the contribution of the financial sector is measured does not properly control for the build-up of risk. The way the contribution of the financial sector is measured is—crudely put—to look at the interest rate times the size of the banks’ balance sheets, and the contribution of the financial sector to value added was at its peak in the quarter when Lehman collapsed. That tells you there is something funny with the—
Q62 Stephen Hammond: Well, you can just double your consultancy fee for doing exactly the same work, which is the point you are making and which is why you have to be very careful about that in terms of measurement.
Sir Charles Bean: The generic point is that this is an international phenomenon. It’s one that is not well understood; there is a lot of academic work and a lot of academic debate about it. The judgment that we have made recognises that we could see an early bounce back to pre-crisis rates or we could be in this world of low productivity growth for a lot longer.
Q63 Stephen Hammond: At the moment, you are calling the two quarters of productivity growth a blip.
Sir Charles Bean: Yes.
Q64 Stephen Hammond: What factors are you looking for so that you could call it a sustained increase?
Sir Charles Bean: First, I’ll explain why we think it is a blip. If you look at chart 3.19, the thing that stares out at you is this. Basically, GDP came in line with our expectations, pretty much, in the second half of last year. Employment has continued to grow. The pick-up in productivity—that’s only when you look at output per hour—is all down to a decline in average hours worked. Average hours is normally the thing that businesses flex when demand changes, so a decline in hours like this is normally something you see when output has slowed, and normally output slows more than hours, and productivity actually declines in those circumstances, so this looks odd. The question then is: what’s going on here? I will give you four and a half hypotheses.
The first hypothesis is simply that this is measurement error to do with the hours numbers in the back half of last year. The hours information comes from the labour force survey. It’s a sample; inevitably, there is sampling error. Sometimes you get cohorts of households coming in that are unusual—so-called cohort effects. That is our favoured hypothesis at the moment.
There is an alternative hypothesis, which is that the measurement error—actually, it’s not so much at the back end of last year; maybe it’s that hours were artificially inflated in the first half of last year, when productivity growth was weak. That is an alternative, which we will monitor. If that turned out to be right—you can look at chart 3.18 on the previous page—that would lead us to shift up our productivity profile. It would be a level effect on the future productivity profile.
The next hypothesis, which I am going to describe as a half-hypothesis, because you can regard it as a combination of the two, says that it’s the seasonal adjustment. The seasonal pattern of hours has changed, and the ONS seasonal adjustment has not yet caught up with that; it will take them time to do that.
My next substantive hypothesis would be that maybe the output numbers are wrong. Output growth was weaker in the second half of last year than we thought. There is quite a lot of discussion in the report about the conflicting signals at the moment from different measures of GDP—income, expenditure and output—and that is a salutary reminder that you should not treat the GDP numbers as gospel. We know they get revised for a long time afterwards and so forth. It might well turn out that output growth is actually weaker, which would mean obviously, correspondingly, that these productivity numbers would look weaker.
The final hypothesis is that something real has happened and that productivity growth suddenly picked up in the back half of last year. In terms of the relative likelihood of those four and a half hypotheses, I have ranked them according to my view of descending likelihood. As we go through this year, we will obviously get more information. By the time we come to our next forecast, we will have a couple more quarters’ observations. If productivity growth remains very strong, that will start making you think, “Ah, maybe something real has happened.”
Q65 Stephen Hammond: If I listened to you right, you ranked those hypotheses in descending order of probability.
Sir Charles Bean: Yes, those are my priorities.
Q66 Stephen Hammond: I was going to ask a question about the difference between your forecast and the Bank of England’s, for instance. I take it that the descending order of the hypotheses explains that. If I listened correctly to what you just said, mis-measurement is a book cost. That is something Professor Chadha at the NIESR has talked about, and he gave evidence to that effect. If you go back two years ago, that was dismissed, but it is obviously higher up your thinking. I would be interested to know why it is higher up your thinking.
Secondly, you talked about a sampling error. Can you say a bit about where you think that might go in the future? Is it a discrete event? Do you think there is any evidence to suggest it is going to continue?
Sir Charles Bean: On the sampling, the labour force survey is a survey of 40,000 households with a revolving panel—households come in and stay in for five quarters, and a fifth of them drop out each quarter. The natural thing, if you’ve had an unusual cohort come in, is that it would stay there for five quarters. There are other things that could be driving sampling error on top of the cohort effects.
The reason I put mis-measurement of the hours as the most likely is that it stands out as not being what you would expect, given what is happening to output and employment. Basically, those indicators suggest that the economy has been chugging along at a moderate rate. We know the world economy has been growing pretty strongly. It is not obvious that something happened that led to a sharp slowdown in the UK at the back end of last year. The natural thing, when you are trying to put these conflicting signals together, is to think that the hours data is the thing that is most likely to be giving a distorting signal.
I should say that this is the norm. When I was on the MPC, every month we would debate some set of indicators that appear to give conflicting signals, and you come up with all sorts of clever hypotheses for why there was some real phenomenon going on, and then three, six or nine months later the ONS would revise the data, or something like that, and the problem would go away.
Q67 Stephen Hammond: This is my last question. In your answer a moment ago, you referred to revisions to GDP forecasts. There has been an upward revision to your forecasts from 1.5% to 1.7%, which leads me to three very quick points. First, how much of that is due to historic revisions to ONS data? Secondly, how much of that is due to upward forecasts for European economies? Finally, in terms of the trend of forecasts you have made and revisions to those forecasts, is there any correlation upwards or downwards that we should be looking at?
Sir Charles Bean: The starting point for the forecast—we have output a bit high relative to the potential we had in previous forecasts—is a joint consequence of growth in the past that was higher than we expected, coupled with looking at a bunch of surveys and things like that to judge the amount of spare capacity in the economy. We have taken a view that things are a little bit tighter than, for example, the Bank views them: it has about 0.25% GDP spare capacity and we are about 0.25% above potential output. That difference is well within the margin of error for these things. The growth profile over the next few quarters is almost entirely down to a stronger world.
Q68 Catherine McKinnell: Let me return to Brexit, following on from some of the questions that Stewart was asking you earlier about migration. You have stated that one key Brexit forecasting assumption is that the UK will adopt a tighter migration regime following departure from the EU than that currently in place. Can you describe how that assumption is incorporated into the forecast? You also said, when you responded to Stewart, that you use ONS figures, but the ONS has said that it does not attempt to predict the impact of future political and economic changes in terms of migration. How do you marry those two things together in your report?
Robert Chote: That is right. The ONS’s approach is basically a more mechanistic one of assuming that the near future will be quite like the near past, and that further on you return to some more average flow. As I said, the choices we have to make are between different population projections, some of which will assume mechanistically a higher or lower flow of net inward migration than the ONS’s central projection. As you say, none of that is based at its end on a judgment about the impact of particular policies; it is a more mechanistic approach than that.
From our point of view, the judgment we made back in November 2016 was that, in the light of what at the time were continued, very strong net inward migration flows—higher than we had anticipated—we would have moved to a higher net inward migration population projection than we were operating on in the previous March. However, we decided on the basis of the Brexit vote that we would stick with where we were. In effect, we didn’t actually change the population projection we were using, but we had a lower one than we would have had in the absence of the vote.
That partly reflected the judgment that, if anything, the policy regime was likely to move in a more restrictive direction, but more important in the nearer term was the expectation that the pull factors would also be weaker. You would be likely to see less net inward migration even in anticipation of that new policy regime coming into place. That is partly because of the consequences of the exchange rate for the incomes of those individuals who are net inward migrants in their home currencies, for example, and broader issues about the degree of attractiveness of the UK in people’s perceptions. We therefore stuck with the ONS’s central projection. In fact, net inward migration numbers have moved pretty closely in line with the central projection that we have been using. The principal projection assumes net inward migration of 246,000 in the year to June 2017. The latest data we have are for the year to September, and it was 244,000—so very close, albeit it a quarter apart. On that basis, at the moment that judgment seems to have been okay. Net inward migration is declining; the decline is in EU migration, not non-EU migration, and it is broadly of the profile that our chosen population projection delivers.
Q69 Catherine McKinnell: Okay. Going back to the public finances, how do you expect that reduction in net migration to affect employment rates, wages, wage growth and therefore the public finances? What are your predictions?
Robert Chote: I might ask Charles to say a bit more about wage growth. The increase in employment that we have in the forecast is a consequence of population growth. The employment rate is not delivering that; it is a consequence of the population growth. The outlook for wages is more down to our productivity judgment.
Sir Charles Bean: The assumption is that the migrants have the same characteristics as the UK population, so it is as though you have a domestically generated increase in the population. It blows everything up, but it does not change things like wages, the natural rate of unemployment, and so forth. It is effectively neutral on pay in the long run.
Robert Chote: We have a weak path for real wage growth over the course of the forecast, but that is fundamentally down to the judgment about productivity growth.
Q70 Catherine McKinnell: If the Government were to be successful in bringing migration down to the tens of thousands, which is the stated aim, what impact would that have on your economic forecast?
Robert Chote: As I mentioned earlier, if you go back to the study, or the exercise, that we did in, I think, March 2016, the difference between the high principle and low migration assumptions was about 65,000 a year by the end of those forecasts. In the scenario analysis that we did at the time, it made about £6 billion a year difference by the end of the forecast horizon. That is to say, if the net inward migration was higher, borrowing would be £6 billion or thereabouts lower—towards 0.2% or 0.3% of GDP.
Again, coming back to what I said earlier, and to what Charlie has just said, clearly the working assumption that we are using is that the characteristics of the net inward migrant are the same in terms of productivity, gender and age specific employment rate and so on. If an incoming migration regime looked to us to be likely to change that mix, the answer might be different, because you would have more or less of an impact on wage growth for a given change in migration than we have pencilled in so far. So far, we have just assumed that the composition—
Q71 Catherine McKinnell: So your projections at the moment do not assume that the Government will assume their stated aim? They assume that we will remain at similar levels of net migration.
Robert Chote: I think we get down to 165,000 by the end, or thereabouts.
Q72 Catherine McKinnell: But if the Government were to achieve their stated aim, would that have a negative impact on our productivity and GDP?
Sir Charles Bean: Probably on GDP, because the economy would be smaller in terms of the number of people.
Robert Chote: It would be less so on GDP per head.
Sir Charles Bean: It is much less clear on productivity, because it depends crucially on the composition of the workers who are coming in relative to the UK workforce. If these workers are exclusively skilled workers with very high wages—
Q73 Catherine McKinnell: But you have already said that your assumption is that net inward migrants are the same as UK—
Sir Charles Bean: No, that is the assumption that we make at the moment, because there is no stated Government policy about what the regime is. If you are going to say that the Government are successful in getting it down to fewer than 100,000, you had better tell me how they achieve that, and what the policy is. Depending on how the policy works, that could certainly have implications for productivity and earnings.
Q74 Catherine McKinnell: Okay. That is helpful. I also wanted to ask about the labour market more generally, and living standards. Do you think that the relationship between unemployment, wages and inflation has changed in the last 12 to 18 months? If so, what are the factors behind that change?
Sir Charles Bean: I am not sure why you would particularly want to focus on the last 12 to 18 months. There has been a general puzzle over the quiescence of earnings growth for almost the past decade. When I was still at the Bank, we were expecting pay growth to start picking up, as unemployment was falling quite rapidly in 2013 and 2014.
Q75 Catherine McKinnell: I guess I was looking more specifically at the equilibrium rate, which seems to have fallen below its level in the last 18 months.
Sir Charles Bean: We have taken a judgment to reduce our estimates of the underlying equilibrium unemployment rate, basically because actual unemployment has fallen below our previous estimate, with little sign yet of pay growth picking up, which is an indication that our previous judgment perhaps was not in the right place, so we have nudged it down. It is worth saying that the Bank of England’s assessment is even lower than ours. We may well take a view in future forecasts that we need to lower it further.
However, it is worth saying that there is a limit to how far this can go. We already have low unemployment rates by comparison, going back to the early ’70s. They are not as low as they were in the ’50s and early ’60s. The key thing that connects the whole question about what the rate of growth potential output is in the UK is that we have been able to grow at a reasonable rate since the financial crisis—I think it is about 2% on average since 2010—despite the fact that productivity has virtually stagnated because we have had a lot of growth of the labour input. That has come about partly through population expansion with migration, partly through increased participation, particularly of older workers, and partly because the natural rate of unemployment was falling.
But we are now approaching an environment where you cannot count on the labour input continuing to grow as rapidly if migration is going to be cut back or dropped back autonomously. We are reaching low unemployment rates and it is debatable whether you can push the participation rate up very much more. This is why it becomes crucial to get productivity growth up—to go back to the pre-crisis growth model—where growth was much more driven by productivity improvement than by increases in the labour input.
Q76 Catherine McKinnell: Where do wages come in to that in terms of future production? I know the Resolution Foundation has said that the forecasts for living standards are very gloomy. Households are in the midst of the longest squeeze endured since the financial crisis. Would you agree with those statements? Do you think there is more of that to come?
Sir Charles Bean: Certainly, as we have it, it is a projection that has continued pretty modest real income gains over the forecast period. That is our central view. That essentially reflects the weakness of productivity growth. There is some pick-up there, because we have some pick-up in productivity. Indeed, you would expect that, as the labour market gets tighter, businesses will be forced to utilise their labour more efficiently. They will also probably have to pay more to get the workers they need and so forth. The normal labour market pressures may help to put some upward pressure on pay going forward, but it is true that the outlook we have, unless we get a sustained pick-up in productivity growth, is one where inevitably living standards remain not exactly stagnating—
Catherine McKinnell: “Gloomy”, as the Resolution Foundation has put it.
Robert Chote: One wrinkle that is worth noticing here is that, for all of the reasons that Charlie has explained, we have a weak projection for real average earnings growth. We think it will be about 0.7% a year over the course of the forecast. Net household disposable income is even weaker. That is partly a consequence of the fact that you have a continued freeze for working-age welfare and there is also some impact from dividend payments. Depending on which of those two metrics you look at, you have a slightly weaker picture for net household disposable income than you do even for earnings growth. As Charlie says, at the end of the day, the big question is whether you get the underlying improvement in productivity growth, because that is where living standard improvements come from in the end.
Q77 Catherine McKinnell: To what extent have you managed to look at and take into account what is known as the gig economy—the increase in what some call flexibility and others call insecurity within the labour market—and the potential forecast, ultimately in terms of public finances and tax receipts?
Sir Charles Bean: There are certainly issues to do with the gig economy and about whether they are being fully picked up in the statistics, and whether they are affecting tax take. The gig economy in particular is people working relatively short jobs, and lots of them, hired through internet platforms and things like that. Some of those may get picked up in the statistics. Some may not. Whether the people earning are always declaring them in their tax returns, I would leave as an open question. One might think that some of that is escaping the tax net as well. It is possible that the measurement issues around the digital economy have become more important. The statistics review that I completed for the last Chancellor has a considerable discussion of those sorts of issues.
Q78 Catherine McKinnell: If those statistics don’t improve, there could be some unknown factors—it is a rising phenomenon—which will affect your ability to predict future tax receipts.
Sir Charles Bean: Yes. Certainly the view that I came to when I wrote the statistics report was that the gig economy was probably not large enough at the time of writing to be a significant factor behind things like the productivity slowdown and stuff like that, but it could become more important in the future, with Airbnb and Uber and so forth. On the other hand, there are other aspects of the digital economy to do with unusual new business models for charging for activities delivered over the internet, the disintermediation of activity from the market economy and home production, such as people organising their own holidays through the internet. That can have a material impact on the GDP numbers for instance.
Q79 Chair: Finally, I want to ask about spare capacity. I think the forecast is that the economy has been above or close to capacity since mid-2014, but until a year ago, it was thought the economy still had some spare capacity. Perhaps you could talk us through the revision. I think one of the forecasts includes a scenario in which the economy is already overheating, with output around 3% above potential. I wonder if you see that as a major risk.
Sir Charles Bean: In terms of the changes to judgment to where we are, the way we do that is by looking at a range of indicators—about nine different models. Some of these are surveys; some of these are econometric models. We look at them against the background of what has been happening to inflation and pay and those sorts of things. Looking at all those things together, we thought you could tell a better story about the past few years if the profile was a bit higher. Where we were in the previous forecast was right to the bottom of that swathe of indicators of the margin of spare capacity. We are now a little bit more within that.
Sorry, I didn’t get the second point about 3%.
Q80 Chair: The latest forecast includes a scenario in which the economy is already overheating. Don’t worry—you haven’t forecast at 3%.
Sir Charles Bean: Okay, yes. That was merely to illustrate. Suppose we have got it wrong and the economy is actually in a much tighter position. There have been plenty of occasions in the past. If you go back to the late 1980s or the 1970s, they are similar cases, where policy makers at the time misjudged the amount of spare capacity. We thought about the current juncture where central banks are starting to think, “Should we be starting to tighten policy?” Obviously, in the US they have started that. Unemployment is getting back to historical lows. It made sense to look at an environment where we might have underestimated the inflationary pressures that are there in the economy. One way to get that would be through our misjudging the margin of spare capacity that we are operating with.
Q81 Chair: But that is a scenario at the moment, rather than anything pointing in that direction?
Sir Charles Bean: Yes, and it was quite deliberately to expose a misjudgement about the economy and the amount of spare capacity. The other scenario we had was where we misjudged the pressure on resources, but for good reasons, if you like, because the world economy turned out to be much stronger and so forth. The first scenario, where we were actually further above the potential than we think, is, if you like, a bad scenario.
Chair: Thank you very much indeed for your evidence this morning. It was fascinating, as ever. I can certainly promise Sir Charles and Robert a return visit before the Committee later this year, all things being equal. Graham, thank you very much. I think this is meant to be your last session. Thank you for all the evidence you have given to the Committee over the years. It has been much been appreciated. I wish you the very best for life beyond the OBR.
Graham Parker: Thank you.
Chair: Thank you all.