Treasury Committee

Oral evidence: The Budget Autumn 2017, HC 600

Thursday 30 November 2017

Ordered by the House of Commons to be published on 30 November 2017

Watch the meeting 

Members present: Nicky Morgan (Chair); Rushanara Ali; Charlie Elphicke; Stewart Hosie; Catherine McKinnell; Kit Malthouse; John Mann.

Questions 145 - 229

Witnesses

I: Robert Chote, Chairman, OBR; Graham Parker, Member of Budget Responsibility Committee, OBR; Professor Sir Charles Bean, Member of Budget Responsibility Committee, OBR.

Written evidence from witnesses:

Examination of witnesses

Witnesses: Robert Chote, Graham Parker and Professor Sir Charles Bean.

Q145       Chair: Good morning.  Thank you very much indeed for being here this morning.  This is our third Budget scrutiny session of the week.  I am hoping that we will be as sharp as we were on the first one.  Please introduce yourselves for the record, because obviously we have not just the audience here but people watching online as well.  Robert, shall we start with you?

Robert Chote: I am Robert Chote and I am Chairman of the OBR.

Sir Charles Bean: Charlie Bean.

Graham Parker: Graham Parker. 

Q146       Chair: Thank you very much indeed.  I want to start with the OBR itself, and its remit and independence.  Please confirm to the Committee—Robert, a question to you—that you have not come under any influence or attempts from the Treasury to alter your economic and fiscal forecasts.

Robert Chote: No, none at all.

Q147       Chair: Thank you.  You stated in your speech last Wednesday that “the Government did alter its planned profile of public spending too late for us to incorporate the change fully into our economy forecast … the direct effect [of which] was sufficient to keep net debt falling as a share of GDP this year”.  Could you just expand on what you meant by that?

Robert Chote: Yes. We have a set of deadlines by which various bits of information have to be communicated to us, so that we can get the product done and finished on time, one of which is that we have to be notified by a particular point if there are measures that are likely to affect the economy forecast.  There is greater latitude to tweak features of decisions that do not have that sort of impact, but there is an earlier deadline on that basis.  Typically, if you have changes in the profile of departmental expenditure, that is the sort of thing, given a certain size, that is likely to have that sort of effect.  We would normally look at the multipliers that we apply these things to in seeing how economic growth would shape out, as a result of that.

On this occasion, the Government, as they are entirely entitled to do, decided after that deadline had passed that they wanted to move £700 million from one year to another, which would not have had a material effect on the economic forecast.  Nonetheless, as it happened after that deadline, we are, and they know that we are, transparent about saying that that would have happened.  It also had the effect of ensuring that the net‑debt‑to‑GDP ratio does fall in 2018‑19, albeit by a very small amount.  Again, we were transparent about that.  There is nothing in the rules that says that the Government cannot do this, but they know that, when they do, we will say so and we will explain it to you and others.

Q148       Chair: Do you think that was done in order to show that the net debt figure did fall, as part of the targets that the Government set themselves?

Robert Chote: It is not for me to put windows into people’s souls, but it seems that might be a plausible explanation.

Q149       Chair: Thank you.  Your forecast also states that “while the number [of policy submissions] submitted just before the deadline was lower than in some previous large Budgets, the sheer number under consideration—as many as we have ever previously received—made the scrutiny process more challenging”.  Again, perhaps it would be helpful if you could explain what is meant by that.

Robert Chote: Obviously we receive, through the course of the period in the run‑up to any fiscal event, a large number of submissions on particular policies for us to scrutinise the costing of those, not all of which will eventually reach the finishing line.  That is inevitable; it happens in every Budget.  Needless to say, from our point of view, a smooth profile of those coming in is ideal.  It gives us time to scrutinise them.  It helps ensure that the inevitable glitches, when you try to pile all these things together at the end, can be minimised.  This was a relatively challenging one.  We have had challenging ones before.  Graham’s superhuman capability of getting through enormous numbers of these things at the last minute has never been tested to destruction.

Chair: I am sure you are volunteering him for that. 

Robert Chote: It is in everybody’s benefit that we not rely on that any more than we have to.

Chair: Thank you very much.  Graham, is there anything you wanted to say on that, given the tribute paid to you by Robert there?

Graham Parker: It was quite difficult this time.  The last week was quite challenging.  We still had about 100 that were not signed off at the beginning of the last week, so there was still a lot to do.  We finally signed off the last one at about seven o’clock on the Friday night. The scorecard is supposed to be all done and dusted by noon that day.  It is the latest we have ever done it, but we just about managed it.

Q150       Chair: These are your printing deadlines—okay.  I want to move on.  Inevitably this morning you are going to get a number of questions from us on Brexit.  We have taken evidence from other witnesses who came before us yesterday.  You wrote in, as you have done in previous forecasts since the referendum.  You wrote, “We have no meaningful basis on which to form a judgment as to the final outcome of the Brexit negotiations and upon which we can then condition our forecast.”  First, do you agree that there are two likely outcomes for the Article 50 negotiations?  In one obviously no deal is struck at the end of the Article 50 process.  In one a transition deal is struck that preserves the status quo for a couple of years, while a longer term trade deal is negotiated.  To the OBR, do they seem like the two most likely scenarios?

Robert Chote: You could characterise it in that way, but you would not want to give the impression that that is essentially suggesting there are two well-specified, well-defined states of the world in which you could end up, and it is simply a question of whether it is one or the other.  Both of those characterisations that you give there encompass a considerable range of potential outcomes.  If there is a deal, we obviously do not know what the deal is going to look like and where, for example, the trade‑off on trade, migration control or other aspects of policy might be.  The no‑deal side, for some people, encapsulates something where the outcome in terms of the trade arrangements with the EU is to fall back to a default, which some people would see as less desirable than a bespoke deal, but that could happen in a relatively orderly way.  The alternative is another sort of no deal that is much more disorderly in terms of its economic impact.  The more relevant distinction from a forecasting point of view might be orderly versus disorderly, rather than deal versus no deal. 

The way we have done the forecast, and we have stuck with this position, is essentially to make some broad‑brush assumptions that are consistent with a variety of what you might describe as orderly outcomes, making assumptions about long‑term impacts on things like trade flow, investment and so on.  There is, though, the possibility that you have something much more disorderly, in the sense that it could have a severe short‑term impact on the supply side, particular supply chains and on the demand side. 

From the forecasting perspective, the point about the disorderly outcome is that there is virtually nothing useful one can say in advance from our perspective about what that might look like, on the specifics of the supply‑side adjustment.  On the demand side, we saw people producing forecasts ahead of the referendum.  Making a judgment about how political developments map into movements in business and consumer confidence, which in turn map into movements in their behaviour, is not straightforward.  Clearly a disorderly outcome looks worse, but being able to say we know exactly what a disorderly outcome would look like, in its economic and fiscal impact, would be going a step further.

Sir Charles Bean: That is a pretty fair comment.  It makes me think of the financial crisis when, at the time, it was very difficult to know how that was going to impact the economy.  You learn as you go along, and that will inevitably be the same here.  The one thing I might add to Robert’s remarks is, even for the orderly no‑deal case, we do not know what that looks like.  He said we do not know what the deal case looks like; we do not know what the terms are.  Even if we revert to WTO rules on an orderly basis, there is still a question about what tariffs are set and things like that.  There are so many moving parts in this and they are moving parts that we are not qualified to take a judgment on.  They are about how the politics unfold.

Q151       Chair: The challenge with that is this is, as I think David Davis has called it, the most significant set of negotiations in this country’s modern history, and trade is obviously going to be a huge part of those negotiations.  As you said, no one knows quite where we are going to get to in the negotiations.  Your forecasts are based at the moment on three studies that consider the effects of different long‑term trade relationships between the UK and the EU, and you also forecast, at the moment, that the trade intensity of the economy will fall gradually over a period of 10 years after the UK leaves the EU.  Really what you are saying, and you can obviously push back against this, is that, because everything is so unknowable, it is almost impossible to forecast it at all.  Would you agree that the importance attached to the OBR’s forecasts is overdone, on the basis that everything is so unknowable at this stage?  Trade relationships and the things that you are relying on to make the forecasts are equally a finger in the wind.

Robert Chote: The precision is clearly unknowable, but it is still worth making the sorts of broad‑brush adjustments that you went through.  If you think about the way in which the economy has performed since the referendum, there are things that one would have predicted at that stage.  We explicitly look at this.  We say, “Here are the sorts of things that we would have anticipated.  How has that come out?”  People would have anticipated that there would have been a fall in the exchange rate and that that would have resulted in a squeeze on real consumer incomes and spending.  That has indeed happened.  You would have expected the fall in the exchange rate to produce some boost to export performance.  That has happened, but not quite as much as we and others would have anticipated, and the impact on net trade has been less of a boost, because it has also been accompanied by stronger import growth. 

People would have said that they would have assumed, because of pull factors, that you would have seen some relative weakness in net inward migration, and that has happened as well.  On investment, partly because of the difficulties around the data, the picture is less clear.  The overall idea that investment seems relatively weak, or weaker than you would have otherwise expected it to be, probably holds up, but it is hard to read too much into the precise quarterly profile of the investment data.  As it is, it suggests there was weakness in the run‑up to the referendum and relative strength afterwards, but I would not lay a great deal on that. 

You certainly do not throw up your hands in the air and say, “There is no point in trying to say anything about this whole process at all.”  You can say useful things and then you track how things perform as a result.  On some of the longer term developments, we are drawing on the available studies that other people have done and our own judgment, making some transparent broad‑brush assumptions, but we are not pretending that we precisely tweak those from each event to each event until we have a basis for saying that we have good reason to think that we ought to have something different here.  You are right in saying this is very important, but we should not allow importance to tempt us into pretending a precision that we do not have.  That is quite important.

Q152       Chair: On the other hand, there are a lot of people relying on the OBR’s forecasts: people outside Whitehall, businesses, consumers large and small, and everything else.  I take your point about the two different scenarios, orderly and disorderly.  If there were to be a disorderly Brexit, there could be a big trade shock in April 2019.  Is the OBR thinking that, at some point, it is going to have to model a forecast on that basis?

Robert Chote: At the moment, we have to produce a median forecast.  On that basis, having the median assuming a relatively orderly outcome is sensible.  It is what most other people are doing.  It is what the Bank is doing; it is what the Minister is doing etc. But that is not to say it is not important, particularly for the Government, to look at the risks around this.  You are very generous to say that lots of people rely on our forecasts.  It has to be said that we produce macroeconomic forecasts because you, Parliament, have asked us to produce a fiscal forecast and that is what it is focused on.  There are many other people’s forecasts.  We are constrained to do it on the basis of a particular definition of current government policy.  There are lots of other people producing unconstrained forecasts, so the idea that the outside world is hugely dependent on what we say on the macro side would be self‑aggrandising for us, and I would not make that claim.

Q153       Chair: Just before I hand over to Catherine, I want to ask you again about the information that you had access to.  In the report, you asked the Government whether there was any more information about negotiations or plans for Brexit, and you referred to the Prime Minister’s Florence speech and also to the White Paper on trade that was published in the course of the summer.  We know that the Government have conducted extensive analysis of the impact of leaving the EU with different sectors.  We know 850 pages were released to Parliament earlier this week.  Do you not think the OBR should have had access to that information?

Robert Chote: We have not seen that.  When we ask that question, which we do routinely as a letter, saying, “Is there anything you want to tell us?” the key point to make is that we are not asking for private information.  If we were told something that was not in the public domain, I would not want to put that into the forecast, come to you and have you say, “Why have you got growth doing this or that?”  “It is because the Government have told us something secret about their plans here.  I cannot tell you what it is, but it is jolly convincing and, therefore, we’ve done X or Y.”  It is right that the forecast is produced on the basis of the information that is in the public domain. 

When we write to the Government asking this question, what we are essentially saying is, “Is there anything that you are planning to put into the public domain, at the fiscal event or prior to the fiscal event, that we should be taking into account, as we are preparing the forecast that we will be publishing on that day?”  We are not asking for information that Parliament and the public do not have, because I would not want to produce a forecast without being able to explain where it came from.  It is as much a process thing, rather than a continuous bleat that we are not being given information that we think we ought to be entitled to.

Q154       Chair: The Chancellor told us that you have statutory access to the Treasury’s own internal modelling on Brexit.  Have you had access to that?

Robert Chote: We have not gone into that.  We have taken the approach we have in terms of where we think the meaningful outcomes are.  As time goes by, we have the ability to go to the Treasury and ask them for their analysis, but we can also talk to other people about that.  You would not rely solely on them.

Q155       Catherine McKinnell: I wanted to ask about productivity.  UK productivity has been weak since the financial crisis.  It has consistently underperformed the OBR’s forecast.  Despite this, until March 2016, you were expecting it to return eventually to pre‑crisis levels of around 2% but, since then, there appears to be an increasingly pessimistic view about the trend and productivity growth.  To what extent has our decision to exit the European Union had an impact on your forecasting of productivity trend?

Robert Chote: It was not the driver of the major adjustment that we have made in this forecast.  What we have done on a number of occasions, and this is probably the third, maybe fourth, time that we have done this since 2010, is look basically at the comparison between the relatively strong period of productivity growth that you had in the run‑up to the crisis or maybe prior to the crisis—it is not clear that that is the absolute break point—and the relatively weak performance since.  If you go back a few years, people would have been pointing to explanations for the weakness of productivity that were quite closely linked to the financial crisis.  Were firms holding on to labour because they thought there was a particularly severe but temporary downturn in demand?  Were the problems in the financial sector getting in the way of a reallocation of capital to potentially more productive, rapidly growing firms? 

For example, in November 2011, we made a big adjustment on the underlying potential GDP growth performance, because some of those temporary explanations were not kicking in, and we have returned to that on a number of occasions, including this one, to revisit that judgment.  In the November 2016 forecast, we did make an explicit Brexit adjustment.  It was the first forecast that we had produced subsequent to the referendum, and one element of that was to say that it seems reasonable to assume weaker trend productivity growth.  We primarily related that to the expectation that investment would be weaker. 

Coming back to your earlier point though, it is not the case that we have constantly assumed that productivity was going to return to its previous level.  We have assumed that, over time, productivity growth would get back to the sort of pre‑crisis rates that you had seen, but that still leaves an enormous amount of lost ground that we assumed would never be made up.  That is different from the way people would make forecasts about a normal economic downturn, where you assume you dig yourself into a hole and you dig yourself back out of it.  That is not the case here.  On this occasion we have basically said that we assume there will still be a recovery in the underlying trend rate of growth and the underlying trend productivity growth rate.  We are not in the techno‑pessimist camp, but we are not going to get back to anything like the historic average rate by the end.

Sir Charles Bean: In a sense, chart 1.2 on page 9 of the report encapsulates the story.  We have come to the point where experience has triumphed over hope.  As Robert has already said, for repeated forecasts we have had this assumption that, essentially, there will be a pretty imminent reversion to something like the pre‑crisis trend.  The failure of that to show up has prompted us to stand back.  We did this alongside the forecast evaluation report in the summer.  Does it really make sense to be continuing to say this recovery is just around the corner?

It is worth stressing that the slowdown in productivity growth is an international phenomenon.  It is not a UK phenomenon.  It may look a bit more pronounced here, and part of the reason for that is simply that a lot of the growth in recent years in the UK has been overly concentrated in relatively low‑productivity sectors, so there is a composition effect.  The generic point that productivity growth has slowed is common across countries, and it is not so obvious that it begins at the time of the financial crisis.  The picture on page 9 makes it look that way but, certainly if you look at US data, people like Robert Gordon, who has studied this in depth, date it at something like 2003.  For the UK, arguably the figures pre‑crisis are flattered in an upwards direction because of the way productivity in financial services is measured.  It uses a technique that imputes value‑added by multiplying the interest rate wedge between loans and deposits by the size of the balance sheet.  Obviously they were expanding like mad pre‑crisis and contracted subsequently.  The underlying story here may be more durable.

Q156       Catherine McKinnell: I want to clarify the impact of the decision to leave the European Union on the productivity forecast.  Your forecast shows that, from March 2019, it will increase more than it has done in the last eight years, which would indicate that, despite an initial downward revision on the basis of a concern about investment, there does not appear to be a big impact from Brexit within your forecasts.

Sir Charles Bean: A reason for expecting some recovery is simply that, as labour markets get tighter, it forces firms to become more efficient in the way they use their labour.  Also, you would expect a bit more upward pressure on pay, although most people have been puzzled by why pay growth has been as weak as it has been.  You would expect, as the labour market gets tighter, there to be pressures that will drive some improvement in productivity.  That is part of the reason why we have not gone all the way to saying we are just going to stay at this post‑crisis average for evermore.  As Robert said, in the long run we are techno‑optimists, but it is a question about how quickly we get back to that.  If you think about things like artificial intelligence, robotics and stuff like that, you would expect them eventually to start showing up in the productivity numbers.

Q157       Catherine McKinnell: Sir Charles, you spent 14 years at the Bank of England as the Chief Economist to the Deputy Governor.  The Bank of England, in its paper on the UK’s EU membership published in October 2015, said, “There is substantial evidence that openness supports economic dynamism through a range of channels, thereby raising economic growth and boosting living standards.”  I would be grateful if you would confirm whether you agree with that, but also the opposite: that a decline in openness reduces economic dynamism, therefore lowering economic growth and cutting living standards.  In the context of Brexit, therefore, are there any concerns in terms of future forecasting?

Sir Charles Bean: There certainly is empirical evidence to suggest that openness is conducive to productivity growth, in particular technology transfer through multinational firms and things like that.  It is worth saying that the evidence here relates to quite big changes in the degree of openness of economies.  Whether you can extrapolate those results across to the case of Brexit might be open to a bit more debate.  Certainly the magnitude of the effect is open to debate, like all these things in empirical economics.

The other thing that is important is that most of the evidence relates to what happens when you make economies more open and more integrated.  Of course, it depends what happens in terms of our relations with other countries and exactly what the relationship with the EU is as well.  Obviously, we may be talking about something that is a reduction in openness, and the question of whether you can apply quantitative estimates that relate to increasing openness and just flip the sign for a reduction in openness is debateable, if you think about the processes that are going on here. I agree with the Bank, but there are qualifications that one needs to bear in mind.

Q158       Catherine McKinnell: What would it take to revisit the assumptions that have been made about trade openness and productivity?  In what circumstances would you revisit those?

Sir Charles Bean: We would certainly revisit the forecast if we get more information about what the post‑Brexit world looks like.

Q159       Catherine McKinnell: What information would significantly alter it?

Sir Charles Bean: Knowing the nature of the deal with the rest of the EU, and the potential relations with other countries, is important.  If we had more information on that, it might well lead us to make some changes to the numbers we have.  Even when we do that—and I want to go back to the point that Robert stressed—there is a huge amount of uncertainty in this space.  Economists, when they try to form estimates about the impact of things, usually rely on looking at similar things that happened in the past and extrapolating from that experience through to the current episodes.  There is no analogue to Brexit from the past.  About the nearest you can get is New Zealand back in the late 1980s.  There are some lessons you can draw from that, but we do not really have an analogous episode, so there is always bound to be a degree of uncertainty.  It will only be as things unfold that we really get a better understanding of what has been involved.

Robert Chote: I would add to that that many PhDs are going to be written on the impact of Brexit over the years to come.  The problem is, if you look back at this chart on page 9, when we are all sitting here in 20 to 30 years’ time looking at this sort of stuff, disentangling what the impact of Brexit was versus how the underlying productivity puzzle resolved itself or did not is going to be very difficult.  You will be able to look at the experiences of the UK versus other countries.  Ex ante it is a very difficult judgment to make but, even when people get to study it ex post, it is going to be very hard for people to say, wherever in this wide band or even outside it you end up, how much of that was down to Brexit and how much of it was down to the pre‑existing puzzle.

From our point of view, at the end of the day, we have to come up with an end-point number that we are reasonably comfortable with, recognising all the uncertainties around it.  Different people, even if they agree with that number, may have got to it by a different combination of their view of the no‑Brexit counterfactual and the Brexit impact.

Q160       Catherine McKinnell: I have one final clarification on the post‑Brexit forecasting, in relation to migration, which you have mentioned a couple of times, and the assumptions that have been incorporated into the forecast.  Are they consistent with the assumption that there will be a tighter migration regime?  What are the implications of that for the forecast?

Robert Chote: What we have done in previous forecasts and this one is basically to use the Office for National Statistics’ principal population projection, which has an assumption in it about net inward migration.  That assumption is not a policy‑driven bottom‑up one.  It is basically saying that, in a few years’ time, you end up with a long‑term average, but you are closer to more recent behaviour since.  Those population projections were updated between the March forecast and this one.  They were updated in October, and so we have incorporated those new projections in it.  They suggest that net inward migration falls to a lower level of 165,000 a year, rather than 185,000, from memory, and we have incorporated that.

The other interesting feature in terms of its fiscal impact is the make‑up of the net inflows as well.  Typically, you point out that net inward migrants are more likely to be of working age than the average of the rest of the population and that has beneficial fiscal consequences over some time horizon.  On this occasion, in the new projections, the mix is somewhat less favourable in that way than it was previously.  We have taken that into account.  It is one of the factors, though not as big as the productivity adjustment, in leaving us with our view of the underlying potential growth of the economy over this period. 

Q161       Catherine McKinnell: What do you mean by “less favourable”?  Could you clarify?

Robert Chote: The ONS does not say, “We’re making a guess on what the new migration regime is going to be.”  It is more mechanistic than that.  As it happens, the outturns for migration appear at the moment to be moving pretty much in line with the projections.  That is not because we have a new migration regime; it would appear to be because pull factors are weaker.  That is partly reflecting exchange rates and what the value of working in the UK is in a potential inward migrant’s home currency etc.  It is consistent with that, but it is not driven by a particular assumption about the end point of the migration regime.

Q162       Charlie Elphicke: I have a very small supplementary on this.  Can I challenge you on some of the assumptions that you are making?  You are assuming that there will be a decline in economic openness in the UK when we leave the EU.  I put it to you that that is simply unjustified, particularly given the comments of the Government that Britain is to be open, and open to not just the EU but the whole world. On your assumption that the mix of net migration into the UK would become older, if we have a work permit system or something like that, that assumption is completely unjustified.  Do you not think you should be revisiting that?

Robert Chote: I am sorry—do we assume that they become older?

Charlie Elphicke: I thought you said that the mix is less favourable.  It sounded like they are become older and less working age.  That sounded like the assumption.  Was that your understanding as well?

Catherine McKinnell: I wanted a clarification on what you meant by less favourable.

Robert Chote: Look at the change from the Office for National Statistics.  We do not do population projections and migration projections ourselves.  We use the one that the Office for National Statistics does, and that is based more on the way in which those flows have been evolving over the recent past than it is on a particular judgment about changes in the policy regime.  It is a more mechanistic approach than that.  That is what we have to go on.  At the moment, obviously we do not know what a migration regime will be.  It is presumably going to be affected by the consequences of the rest of the negotiation.

In terms of openness overall, what we have done here is to look at a variety of studies of the likely trade implications of this and the average across a lot of those.  Most of them tend to the expectation that there will be a net decline in openness, on the grounds that the assumed deal with the EU will be less open to trade in that sense.  That is because of gravity: the size of the EU economy and the closeness of it.  That is likely, until you see other evidence, to outweigh changes in other trade regimes.  We have to make an assumption at this stage, and to go with one that is with the grain of the analysis that trade economists have done at this stage, until we know what the outcome is going to be, is the best we can do.

Sir Charles Bean: Absolutely but, as I hinted earlier, there is a range of possibilities here.  We could well find ourselves in a situation where there is a very good deal with the other EU states that leaves us with access very close to what we have at present—that is presumably what the Prime Minister and the Cabinet will be aspiring to—and where we also manage to cut some good deals with other countries.  Clearly there are possible outcomes where we might have improved openness, but we have to make a central assumption.  As Robert said, we think it is reasonable that that base assumption, that central assumption, in the absence of any other information at the moment, is that it involves slightly less good access, particularly to European markets.

Q163       Rushanara Ali: Good morning.  The OBR has not made any allowance in its forecast for a Brexit divorce bill or ongoing contributions to the EU budget during the transition period.  Maybe, Mr Chote, you can start off by explaining why it has not.

Robert Chote: The approach we have taken is basically to run a counterfactual exercise of what we would be making in terms of contributions if the UK had not voted to leave.  That is set out in the report.  The fiscally neutral assumption that we have made is that, if there is any reduction in that, total or partial, that money would be spent on something else.  That could be spending it directly on something that EU support had been paying for previously.  It could be providing more money for the NHS.  It could be continued contributions to particular EU frameworks or projects after we have left but, in the absence of knowing what the deal is going to be and what those choices are going to be, that seems to be the sensible thing to assume.  The question of whether the divorce bill makes the situation look better or worse will depend on the degree to which the divorce bill simply constitutes making those contributions that we would have made, if we continued to be a member for some time, or whether it is more than that and, in either case, whether the time profile of the payments changes, in which case that would have implications. 

We read in the newspapers that they are coming to some sort of agreement on this.  If that turns out to be the case and this is signed off and in the public domain, we can explore with the Treasury and the Commission what the basis would be for the payments that are intended on this agreement and what their best guesses are for the numbers that will be involved.  Knowing to what extent any given cash number that people are chucking around makes things look worse or better depends on how much of it ends up being the money that is already in the baseline, or whether it is something extra or potentially something less.

Q164       Chair: Can I just clarify what you are saying?  Yesterday, when the Chief Secretary to the Treasury was asked about this in the House of Commons, she said the OBR had included EU payments.  Those are included in the Budget.  But the chart at the back of your forecast or your outlook very clearly states that the contributions stop in 2019‑20, and then, as you say, there are domestic payments.  Can I just be absolutely clear?  Have you included any provision for a withdrawal payment in your forecast published this month?

Robert Chote: As I say, we have assumed that, where you draw the line of saying this is no longer an EU contribution and this is now domestic spending does not make any difference to the numbers.  All we have is a table with a line of numbers.  At some point, you shift it from one line to another, but the numbers are unchanged.  The impact depends on whether you are changing the total amount of the contributions relative to a counterfactual in which we were still a member and continue to be doing it.

Q165       Chair: It could be a situation where we are paying a withdrawal bill, which could be over a number of years, and paying out that domestic spending as well, which is the decision of the Government. 

Robert Chote: At the same time as all this, the Government are making decisions about domestic spending all the time.  You end up with a situation of saying, “Well, we have put some money into the NHS.  This is paid for by X.”  That is the kind of rhetorical hypothecation and, from our point of view, you are looking at what they are saying the domestic spending would be.  We are making this assumption that, if there is complete or partial reduction in the EU contribution, that is spent.  If the money is spent on health or something else like that, it is not creating more of a black hole in the public finances.  It is leaving you where you started.

Chair: The contributions to the EU at the moment stop, according to your outlook, in 2019‑20.

Robert Chote: The forecast in effect assumes that the contributions go on indefinitely. 

Chair: They are characterised as domestic spending.

Kit Malthouse: They are on a different line. 

Robert Chote: At some point, it just appears in a domestic line.

Kit Malthouse: Government spending overall remains constant throughout the period, whether it is in the EU or elsewhere.

Robert Chote: Whenever you leave, exactly.

Chair: It just changes lines.

Graham Parker: It does not stay constant.  It is what it would have been, if we had still been a member.

Robert Chote: Yes, it does not move around.

Q166       Kit Malthouse: Forgive me, Chair.  What the Chair is saying is, if there is a divorce payment, it is effectively absorbed within the overall level of spending.  There is no extra spending. 

Robert Chote: It depends on the extent to which the divorce payment is a commitment to continue to make the contributions that you would have made otherwise, versus something else on top or a re‑profiling of that to make it more upfront or backloaded.

Chair: For example, on contributions to EU officials’ pensions, for example, the assumption is that that stops in 2019‑20.  That is obviously not going to be catered for, one assumes, in your domestic spending line.

Robert Chote: Obviously we have to wait and see what the deal is as regards that particular liability.  That would be an example.  If that comes on top—if you were to make an upfront payment for that on top of the contributions that you would otherwise make—that would alter the number.

Chair: That is not in your numbers at the moment.

Robert Chote: No, we have not made any assumption about that, but it obviously depends on what the deal is. 

Q167       Rushanara Ali: Thank you.  I am going to press on.  In the Florence speech, the Prime Minister stated, “I do not want our partners to fear that they will need to pay more or receive less over the remainder of the current budget plan as a result of our decision to leave.”  Do you agree then that it is now firmly Government policy to contribute at a minimum to the 2014‑2020 Multiannual Financial Framework, assuming that the negotiations proceed amicably?

Robert Chote: I presume that depends on the deal that has or has not just been done, which is presumably the new statement of government policy, as and when it is made.

Q168       Rushanara Ali: More recently, it has been reported that the EU strategy Cabinet sub‑committee has agreed in principle to a divorce bill widely reported of up to £40 billion, including contributions to the EU budget during transition.  Did you seek guidance from the Government over this Cabinet discussion and the likely size of the divorce bill that the UK was prepared to pay?

Robert Chote: No. As I said, we wrote the usual request for any additional information that the Government wanted to point us to that it was likely to bring into the public domain at the time of the fiscal statement.  No: as I say, what we have assumed here is that you have an amount of spending equivalent to the continuation of the EU contributions, whether we have left or not.

Q169       Rushanara Ali: In the discussion of the risks to the fiscal forecast, the OBR lists six categories of fiscal risk but does not identify the Brexit divorce bill among them.  You have referred to some of the points earlier, but can you add more light to why that is not the case?

Robert Chote: We have highlighted it at various points.  We spoke a bit about this in the fiscal risks report that we produced back in July.  Clearly, at one level, we return to the issue we have been discussing as to whether this divorce bill is an extra amount of money or a commitment to pay stuff that would have been paid anyway.

Rushanara Ali: What if it is an extra amount of money?

Robert Chote: If it is an extra amount, you are presumably talking here about a one‑off payment of some billions of pounds that would be dwarfed by the consequences of Brexit, positive or negative, for the long‑term outlook for economic growth.  There is an analogy here with the time of the financial crisis.  It is very easy to get very focused on the one‑off or the period of particular bailout payments, whereas what matters much more is the impact on the economy and the underlying impact that has on receipts flows and spending flows.

Rushanara Ali: Obviously there were consequences with the bailout.

Robert Chote: There were, but the fiscal and economic consequences of what happened to the path of GDP as a result are much larger than the eventual impact of interventions in the financial sector.  There is an analogy with that in this situation.

Q170       Rushanara Ali: Do you agree that another risk arising from Brexit is the possibility the Government will have to set aside more than the £3.7 billion that has already been allocated for the preparations for Brexit?

Robert Chote: Presumably there are risks on the other side as well.  Maybe they feel they will not need to spend all that money.  They have allocated that amount of money.  We have included that in the overall resource spending limits, and then we have to make some modest adjustments to the overall picture for any underspending that we expect against that, but it is basically an addition, like any other addition, to the overall current public expenditure.

Rushanara Ali: It is a massive hike to the £250 million previously allocated.

Robert Chote: It is a lot of money, but it is pretty modest relative to the average forecasting errors in public expenditure and receipts.

Rushanara Ali: Sure, but in the referendum debates about what it would cost, that is not what the public was being told.  Do you have any reflections on that?

Robert Chote: I would leave it the public debates to reflect on that.  I would distinguish between one‑off costs of either the process of this or of the divorce bill, and the longer term impact of whatever consequences this has, which are very hard to predict, for the long‑term path of GDP growth and receipts.  In all probability, whatever the sign is, the latter of those is going to be the more quantitatively significant. 

Q171       Rushanara Ali: If we accept that the divorce bill is around £40 billion and there is an additional cost within that—an extra cost that would not have been borne if we were making our payments while we were within the EU—alongside the contingency planning costs of around £3.7 billion, which could potentially be higher, that is a significant amount of money that could have been spent on public services.

Robert Chote: It is a significant amount of money, but bear in mind that it is a one‑off.  It is a time‑limited payment.  Compare that with the projected uncertainties around the cost of social care.

Q172       Rushanara Ali: We were planning to take £3 billion out of the education system.  These are not small amounts of money when you get into the particular services, just as the cost of dealing with the deficit after the financial crisis has had consequences, in terms of who has to pay.

Robert Chote: That is true.  The Government have chosen to spend this money and they are borrowing more money in the years concerned, as a consequence.

Rushanara Ali: There is a cost to this.

Robert Chote: It has not cut the overall amount of spending on other things commensurate with that.

Q173       Chair: On that £3 billion, the Chancellor said to us that the additional spending on Brexit does not require the reopening of departmental spending plans, because the money is coming from reserves.  Do you agree with that assessment?

Robert Chote: In effect, the overall RDEL limits have been increased.  You can think of this like a reserve on the grounds that, as far as I am aware, this has not been allocated to departments.  You can think of this as having created a £3 billion Brexit reserve that you are then going to spend in the departments that you think need it, for the purposes for which they need it.

Q174       Charlie Elphicke: Can I just return to productivity again, just to push you a bit harder on this issue about the assumption that there is a decline in economic openness of the UK?  It seems to me that this is not justified as an assumption, and I want to understand better how you are justifying this.  What will you say to people who say, “The OBR is basically anti‑Brexit”?

Robert Chote: That would be an accusation without any foundation whatsoever.  We are making a judgment on the basis of the information that is available to us.  We explained in the answer to the question that you asked previously how that was done.  We looked at the range of outside analyses and made our own judgments as well, on the grounds of the most likely direction of openness, making some assumption that, until proven otherwise, you are more likely to see access to EU markets as less easy rather than more easy as a consequence of this.  That is quite important because of what we know about the things that drive trade flows, viz. the importance in gravity modelling terms of the relatively large, relatively close trading blocs.  That seems a reasonable assumption to make until we know what the outcome is but, as we highlight throughout, obviously it depends on what the deal is with the EU.  It depends on what you do in other countries as well.  However, if we came up with a very non‑consensus view that that was going to push you in the other direction, people would probably have concluded that was unrealistically convenient for the Government. 

Charlie Elphicke: You are nevertheless forecasting a reduction on the EU side but not forecasting any increase in Britain finding new opportunities in the wider world.

Robert Chote: It is a net change overall.  Depending on which of the studies we have pulled together on this that you were to look at, some would be more bottom‑up than others. 

Sir Charles Bean: You opened this question by saying you wanted to link it to productivity.  Now, I think I am right in saying that when you put the original effects in, which predate me being on the committee, there is a small productivity effect, but it is tiny compared with the scale of the change that we have made on this occasion.  It is tiny compared with the scale of the shortfall of productivity relative to a continuation of the pre‑crisis trend.  It is very important to realise that the big issue economically for the British economy is productivity; it is not coming from the Brexit assumptions or anything like that.  That is the really big challenge.  If we could do something to address that successfully, it would completely swamp the Brexit consequences, whichever sign they are.  The scale could not be big enough compared with the consequences of the productivity slowdown.

Robert Chote: It is instructive to compare the revisions that we have made over the last few years to both the trend underlying productivity position and potential output with, say, those that the Congressional Budget Office has made for the United States.  It is remarkable.  We have made somewhat larger, but pretty much in the same sort of ballpark, adjustments to the underlying trend productivity growth story until 2020.  For the CBO there has been an even bigger hit to cumulative potential GDP growth, because they have had less favourable news on labour market participation.  It comes back to this point that you can overdo it by focusing on particularly UK‑specific explanations of this.  There is a lot more going on with it than that.

Q175       Charlie Elphicke: Let’s look at that.  Basically for the last decade, from this graph, we have had zero productivity growth.  Has there been any other time in British history when that has happened?

Sir Charles Bean: In the late 19th century, there was a period of very weak productivity growth, which economic historians refer to as the climacteric for some reason.  They still debate its causes.  That period lasted 15, maybe 20, years.  It was a period of very low productivity growth and then it picked up again.  That may have been associated with the advent of electricity, the combustion engine and so forth.  You do see waves of faster and then slower productivity growth but, as I say, you have to go a long way back to find an analogue of the experience we have had recently.

Q176       Charlie Elphicke: Nevertheless, it has lasted for 10 years and it has happened since the crisis.  Normally when you have an economic event like a recession, it acts like brushfire through the economy.  All that is inefficient goes and then the new shoots come up in due course.  Is it possible that economic policy did not enable that process to happen?  That might be part of the issue.

Sir Charles Bean: That was certainly one of the candidate explanations. I can take you through the sequence, because obviously when I was at the Bank of England we were grappling with exactly the same problem.  Initially, the productivity weakness was thought to be temporary labour hoarding and firms hanging on to particularly skilled labour, because they knew it might be difficult to get it back when the economy recovered.  Then when growth picked up, the surprising thing was firms started hiring.  We then started thinking of exactly the sorts of arguments that you make: that the process of what economists refer to as creative destruction had been a slowdown, particularly because banks in those particular circumstances might be slow to recognise non‑performing loans, as they would then have to recognise them on their balance sheets.  At the same time that would impair their ability to extend new loans to potentially more productive firms.  We thought that was a chunk of what was going on but, as the banking system has recovered, that should have ceased to be relevant. 

Another argument that may have some force to it is that the period of unusually low interest rates—normally you would think that low interest rates are good for productivity, as they encourage investment and so forth—may have had the effect at the current juncture of allowing highly indebted low‑productivity firms to survive for longer.  Again, it is similar to your creative destruction argument.  That element may still be around, because interest rates are still low, but the earlier version I gave you about the impairment of banks’ balance sheets is obviously not relevant now.  There is something in these stories, but it is not all the explanation.  If you look closely at what is happening across lots of countries, even countries that did not get seriously affected by the financial crisis have seen this slowdown.  In the US, it looks like the slowdown predates the crisis.

Q177       Charlie Elphicke: Sir Charles, in your own opinion, what do you think drives this and what do you think the possible answers could be?

Sir Charles Bean: The answer is that why it is a slowdown is something of a puzzle.  I should say that there are two puzzles.  People talk about the productivity puzzle, but there are two productivity puzzles.  One puzzle is this slowdown from around the time of the financial crisis: what has caused it and how long it will persist.  It is fair to say that, because we do not properly understand its causes, it is very difficult to know how it will unfold and what we might do to address it. 

The other puzzle is the cross‑section puzzle.  Even if you go back to before this slowdown happened and look at productivity differences across countries, the UK stands out as a relatively weak performer.  People have this factoid that it takes us five days to produce what the Germans and the Americans produce in four.  Now, some of the difference there, anticipating a question that you might ask, can be traced to labour market institutions in Europe, particularly France, that discourage employment.  That is obviously not something that we would want to replicate, but that is only a small part of the story.  We have a better handle on the sources of the international differences.  Some of it is that the UK has had relatively low investment share in both public and private sectors.  Some of it is down to skills and a relatively low level of human capital in the UK, so fixing up the education system is obviously the answer there. 

A final element, which people have only become aware of more recently, because the information has only just become available to get at it but does look quite important, is that the UK seems to have quite a long tail of relatively weakly performing firms.  Our best firms are world‑class but, compared with other countries, and even when you go down quite narrowly into particular industries, you get quite a long tail of poor performers, much longer than in other countries.  The answer there is to improve the management of those companies to get better outturns. Charlie Mayfield chairs the Productivity Leadership Group.  One of its aims is to try to share best practice.  The cross‑section differences are things that are amenable to policy and we know something about those.  Even though we have this time-series puzzle and we do not know why productivity growth has slowed, we can do something to get UK productivity up. 

Q178       Charlie Elphicke: From what you are saying, at the beginning there is an issue with money flooding through the system.  You end up with zombie banks and zombie companies.  That moves over time to a situation where you have the recovery, but you have incredibly low interest rates.  If you are a manager in company, why do you need to bother to provide a return?  Interest rates are so low that any kind of dividends you pay out are going to be better than people get in the bank.  Where is the incentive to our management teams to pull their weight, pull their fingers out, make their companies really successful, do really well and make the effort?  Where is the incentive for skills and where is the incentive for business investment?  Are they possibly the answers that we should be looking at?

Sir Charles Bean: If you move to a world where interest rates are somewhat higher, and we are now moving into that world, exiting slowly from the unusually stimulatory policies post crisis, it will provide pressure on some firms.  Equally, as I said, as the labour market tightens it will also put pressure on businesses to be more efficient, and workers will flow from low‑paid, less‑productive parts of the economy into higher paying areas.  That is how the process of creative destruction that you referred to earlier works.  That is basically the reason we have some pick‑up in the productivity trend in the forecast.

Kit Malthouse: Sorry, can I just ask a question on productivity?

Chair: John was going to ask on productivity too and I was going to bring you in on targets. 

Q179       John Mann: I have lots of questions on productivity, as I have for the last 15 sessions I have attended with you, Sir Charles. I am not going to waste everyone’s time by going through all your answers, save to say that it was repeatedly suggested for many of your reports that you were overestimating growth and, therefore, you were showing that the deficit and debt would come down much quicker.  Chancellors were very happy to quote anything good from you and, as it got slightly worse, not quote you.  Now you have gone in the other direction and I am interested in methodology.  Productivity is a good example and, as you know, I have repeatedly implored you to get some better research on the labour market.  I do not have time to disagree at length with Sir Charles, but that is a pretty one‑sided labour market analysis and does not equate to the question of us being the most flexible labour market in Europe and one of the most flexible, perhaps the most flexible, in the world.  Let’s start with the methodology.  The Bank of England is suggesting 1.4% as opposed to your 0.7%.  What is the difference in methodology, Mr Chote?

Robert Chote: For what?

John Mann: For productivity.

Robert Chote: Over which period?

Q180       John Mann: For the current forecasts, you have 0.7% growth in 2018.  The Bank of England’s forecast figure is 1.25%, so they are forecasting a lot higher than you.  What is the difference in methodology?

Sir Charles Bean: Over the whole of their forecast period, their productivity assumption is not very different from ours.  It is true it is a little higher. 

John Mann: You downgrade it from approximately their 1.4% down to 0.7%.

Sir Charles Bean: We were actually stronger than theirs before.  They also downgraded theirs.

John Mann: What is the difference in methodology?  It is a significant change for you, so what is the difference in your methodology and theirs?

Sir Charles Bean: I would say that we have come to a different judgment.  I want to stress that the particular path here is not a finely tuned outcome of a lot of research, some model or anything like that.  The nature of the uncertainty about what is going on driving productivity, this time-series productivity puzzle that we have been talking about, is such that there is no way you can say, very clearly, that is where it is going to go in the future. 

John Mann: Therefore you are saying it is judgment, as opposed to methodology.

Sir Charles Bean: Yes.

Q181       John Mann: Let me then, Mr Chote, come to your judgment on the reliability of the ONS net migration figures this time, as opposed to the previous times.  I appreciate you have to rely on them.  That is not a criticism, but you are making a judgment.  My instinct would be to say that this is less predictable at the moment; therefore, their statistics are more liable to be wrong than they would have been at any time when you produced a report before.  Is that your judgment?

Robert Chote: If you look at the path of what net migration has done relative to the projections that they produced in the past, there has been a continued period in which it has obviously been higher than they had anticipated.  You are now in a situation of seeing it, at last, coming down in line with the projections.  Somebody might look at that and say, “Actually, this is the first time for ages in which the outturn data appears to be moving consistently with the projection,” but the projection is mechanistic and that is a reflection of the pull factors. 

Your broader point is that there are particular uncertainties over a five‑year or longer horizon from the pull factors and then from the possibility of a policy change at the end of it.  We are forecasting on the basis of current policy and we are not making an assumption about what the new migration regime will be.  Clearly, if you are in an environment in which you are anticipating that there is a possible policy change, if you are somebody else doing an unconstrained forecast of this, that is an extra source of uncertainty.  There is another reason to think that a mechanistic approach only gets you so far but, ironically, we are also in a period when the outturn data is moving more closely with the mechanistic approach than it has done for some time.

Q182       John Mann: We are having to rely on your judgment as well.  The OBR raised zombie firms a lot for some years in deliberations.  Have all these zombie firms gone or are we going to see, with the gradual increase in interest rates now, zombie firms exiting?

Robert Chote: This comes back to the questions Mr Elphicke was asking.  This was an important part of an explanation a while back, but it is still there as a possibility.  The combination of continued tightening in the labour market and the pick‑up and tightening in monetary policy could result in a greater period of creative destruction than you have seen before.

John Mann: That will increase productivity, not decrease it.

Robert Chote: That is one reason we have an increase in productivity growth in the forecast. 

Q183       John Mann: Therefore, you are suggesting that the actual increase, the trend‑line increase, is even worse.  You are downgrading even further, because that is going in the opposite direction.  If zombie firms are going with interest rates, that pushes up productivity.  But you are projecting a downward growth from your estimates, so your projections are worse than they would have been otherwise.

Robert Chote: The forecast for productivity growth is picking up but less quickly than it was previously.  You still have the mechanism that we have just been talking about. 

John Mann: Exactly, but with more zombie firms going, the opposite ought to happen.  Economic theory would say that.

Sir Charles Bean: Absolutely.

John Mann: I am going to ask Mr Parker.  Economic theory would suggest that. 

Graham Parker: Yes, but the point is that, if you look at our forecast, productivity growth is positive in the forecast.  It is just not as positive as it was in our last forecast.  The zombie firms are still having an effect.

Sir Charles Bean: More to the point, productivity growth is picking up from where we are.  It is important to realise that we still have a pickup in productivity growth pencilled in.  It is just not a reversion to pre‑crisis rates just around the corner, which has tended to be the assumption that there has been in the forecast, in the past.

Q184       John Mann: Mr Chote, you give an absolute value of 0.7%.  Do you fit confidence intervals around those values?  If so, why are they not published?

Robert Chote: As you see, we illustrate the uncertainty around the numbers in a variety of ways.  The fan charts that we use are simply a mechanistic view on the basis of the size and distribution of past forecasting errors.  Specifically because the productivity growth judgment is such a key one in the change between the last forecast and this one, in chapter 5 you can see that we have a scenario analysis that is explicitly looking at what happens if productivity growth rebounds relatively swiftly to the pre‑crisis long‑term average rate and what happens if it stays at the rate of the last 10 years or so.  That precisely calibrates and gives some sense of how much impact it would make if the answer is different from the judgments we have made.

Q185       John Mann: With that 2018 figure, what would be the upper and lower bounds of a 95% confidence interval?

Robert Chote: For GDP growth as a whole, page 222 is the alternative scenarios on the fiscal position.  The GDP growth fan chart is somewhere. 

Chair: Is it page 61?

Robert Chote: Very good—it is page 61.  Thank you, Chair.

Chair: I have my uses.

Robert Chote: You can see how wide the fan chart is at that stage and, therefore, the possibilities of growth being high or low.  Coming back to your earlier point on methodology, it is interesting that you elided the performance of the forecast for GDP growth as a whole and the forecast performance for productivity growth.  There is an interesting distinction between the two, because it is striking that we have made the same directional error, as most other people have, on productivity growth, both in the forecasts where we have overestimated GDP growth in total and in the forecasts where we have underestimated it.  It is a separate issue about the pace of the recovery overall and the puzzle, which is that, for a given path of the recovery, firms have just employed a lot more people to deliver it than we and most other people would have anticipated.  There is a distinction in methodology, as you are keen on that, between the two.

Q186       John Mann: I would implore you again: more research on the labour market is absolutely fundamental to solve this puzzle.  I do not see stuff out there that is really getting to grips with the impact of flexible labour markets and what they mean.  Having said that, my final question is about what assessment you have made of the impact of productivity changes on living standards.  Obviously inflation is higher, so what is your assessment of that and how do you see the consequences in terms of your work and the projections you are giving us?

Robert Chote: We do not produce distributional analysis, but the key feed‑through is that weaker productivity growth implies weaker earnings growth.  Earnings growth is averaging about 0.6% a year across the course of the forecast, and that is lower because the productivity path is lower, essentially.  The growth in real household disposable income per head is weaker than that because of other things that are going on, for example with the benefits system and with movements in taxes as well.  It is at that relatively broad level that we get there.  The Resolution Foundation has taken the numbers that we have produced and turned them into more concrete estimates of average earnings at particular points.  As far as I can see, it looks as though they have done that perfectly well.  If you wanted to map this into more concrete living standards numbers, they are the people to go to. 

Q187       John Mann: I would like the Chancellor to do that, but what I am interested in from you is the impact in terms of your report.

Robert Chote: Weaker productivity growth over the medium term implies weaker earnings growth than you would otherwise have had.  That obviously matters to most people, primarily because of what it tells you about living standards.  It also matters crucially to the fiscal outlook, because weaker earnings growth is reducing the amount of income tax receipts in particular that you get in, and that is the largest single hit to the fiscal position from the numbers that we produce.  It matters in the report in that sense, but for other people it matters in the direct effect for living standards.

Q188       John Mann: Can you quantify that?

Robert Chote: Do you mean the income tax effect?

Q189       John Mann: Can you quantify the effect of the slow growth in productivity on income tax?

Robert Chote: I think it is £25 billion by the end.

Sir Charles Bean: It is something like that. 

Robert Chote: If you go to page 18, the productivity revision is making £25.8 billion of difference by 2021‑22.

Graham Parker: The bulk of that will be PAYE.

Robert Chote: That is partially offset by the fact that we think unemployment can be sustained at a lower rate than it previously could be, but that is the dominant effect. 

Chair: Kit, do you have a supplementary before you go on to targets?

Q190       Kit Malthouse: Yes. I want to talk about fiscal targets but, before that, obviously every country has a different cocktail of cultural, economic and historical influences on its productivity. How much of a drag on our productivity is the huge ramping up of compliance over the last 20 years?  The compliance drag now seems to be enormous.  Particularly over the last few years we have had some quite significant new impositions on business, which have ramped up the number of lawyers, accountants and compliance officers that they all maintain, which effectively is just more employment dividing the output and, therefore, reducing productivity.  Do you identify that as a drag?  We have such a byzantine tax system now that even the Revenue does not know how to calculate tax in some areas. 

Robert Chote: Charlie, are you aware of any literature on this? 

Sir Charles Bean: I have to say I am not aware of any literature, but it is certainly a legitimate question to raise and there may well be literature out there of a type that I am not familiar with.

Q191       Kit Malthouse: We are just putting through the second‑largest finance bill of all time, and it looks like we are going to have another big one, with 300 or 400 measures this time. Is it a possibility that the complexity of the tax system could itself be a drag on productivity?

Sir Charles Bean: The sign is right.  I certainly would not want to go from that statement, though, to saying that is the explanation for the puzzle. 

Kit Malthouse: No.  I said it was part of the cocktail.

Sir Charles Bean: I am sure the sign is correct—the basic argument that it is a potential drag. 

Q192       Kit Malthouse: On fiscal targets, you use some loaded language in your report about “illusions”, particularly with regard to housing association debt.  Do you want to explain what you meant by “illusion”?  Are we fooling ourselves?

Robert Chote: Fiscal illusion is an International Monetary Fund term of art that basically means being aware of changes that can affect a particular official measure of debt or deficits, or the thing that you are targeting, but does not affect the underlying health of the public finances.  An example would be that selling an asset for roughly what it is worth does not leave the public finances in a greater or better position.  You have simply swapped one asset for another, shares in a bank for cash, but there are some measures of net debt that will include past assets.  If you sell the assets, you reduce net debt, but you are not fundamentally changing the underlying health of the public finances.

Housing associations specifically are an unusual example.  We obviously produce the forecast on the basis of the definitions in the national accounts, and the Office for National Statistics has to decide what is in the public sector and what is not.  They decided in 2015 that housing associations should be treated as part of the public sector and that they should have been treated as part of the public sector since 2008.  They are applying international statistical standards, which say that you decide whether something is in the public sector on the basis of not who owns it or who finances it but who effectively exerts control over it.  They felt that the degree of control exerted over housing associations by central and local government was sufficient that they should be treated in that way.

The consequence of that is that, obviously, the borrowing that housing associations do and the stock of debt that they have then comes on to the public sector balance sheet and clearly makes the headline numbers look worse.  The Government were very clear from the outset that they wanted that to be reversed, and they announced various changes of a deregulatory nature that would be sufficient to get the ONS to reverse that decision and then have the numbers taken out.  What you have seen is a period in which the public finances looked worse than they were, because the housing associations were categorised as part of the public sector.  From our point of view, we think it is very important, when those sorts of things happen at the border between the public and the private sector, to be as transparent as we can about their impact and how it makes things look worse or better.

Q193       Kit Malthouse: The use of the word “illusion”, you are saying, is an international technical term, rather than a comment upon the Chancellor’s abilities as Paul Daniels or otherwise to create headroom for himself.

Robert Chote: No, it is not a phrase that we can claim credit for ourselves.  We discussed this; there is a good chunk of it in the fiscal risks report that we produced back in July.  As I say, the IMF has written a lot in this area.

Q194       Kit Malthouse: On this reclassification that effectively created headroom, would you then expect a change in the fiscal target?  Has it effectively created a loosening by not adjusting the targets?

Robert Chote: It contributes to giving greater room.  It helps achieve the targets as they are written.  It is not decisive in the sense that, if the housing association change had not happened, they would have been breached, but it is helpful at the margin.  There is, and you can explore this with the Chancellor, a philosophical question of whether, given that the Government always felt that housing associations should be off the balance sheet, they set the fiscal targets in that knowledge and in that expectation.  Having returned to the situation that they think is appropriate now, were they pre‑empting that?

Kit Malthouse: They are saying the previous period was distorting things, not the current period.

Robert Chote: I do not know whether they will make that argument to you, but you could try asking.

Q195       Kit Malthouse: Are there any other illusions that we should know about?  What about Bank of England term lending?

Robert Chote: I would not describe it as an illusion.  It has a particular impact on the profile of the debt‑to‑GDP ratio. Because these loans are basically on a four‑year term, most of them end that four‑year term in the target year for having the debt‑to‑GDP ratio falling.  The debt‑to‑GDP ratio in the target year falls by about 3% of GDP and the term funding scheme accounts for more than 2 percentage points of that so, again, it is helpful but not decisive.  The supplementary target would not have been breached in the absence of that, but it is one reason why it is worth looking at both the profile of net debt as a share of GDP and net debt excluding the Bank of England, just to highlight where these sorts of effects can show up.

Q196       Kit Malthouse: When we look at the national debt or deficit, I know that technically from an ONS point of view they say it is correct but, in your view, is it a true reflection of what the country owes?

Robert Chote: The difficulty is that there is no single measure that answers all the questions that you would want to ask.  This is why we went into this at some length with the fiscal risks report.  You should look at a variety of things.  There is public sector net debt and public sector net financial liabilities, which is a relatively recent arrival at the party, which basically contains all financial assets and liabilities, and is not just making a distinction between liquid and illiquid that, in some cases, people might think is rather counterintuitive in the way that is applied in particular cases.  The broadest measure, something like public sector net worth, brings into account non‑financial assets and liabilities as well.  People have tended not to place a great deal of weight on those numbers, because the valuation of physical assets in the public sector is not particularly sensitive; it is hard to do with great accuracy and timeliness.  Graham, is there anything you want to add on that?

Graham Parker: No.  We do look at all these different metrics on debt, so there are figures in the book about them.

Kit Malthouse: We could tot it all up and see what the eye‑watering worst‑case scenario is.

Robert Chote: If you go to the fiscal risks report, we look at a variety of different balance sheet summary measures.  The housing association question is basically a problem for all of them, in the sense that it is defining the boundary of the public sector at which you are looking.  The differences between the metrics are based more on things like whether you count illiquid assets or not, as part of the snapshot.

Q197       Kit Malthouse: I have not looked recently, but what would we see on the whole‑government accounts?  Would it show a significant difference from this trajectory?

Robert Chote: The whole‑of‑Government accounts are partial but in a different way.  One big change is that they would include the present value of public service pension liability, but only for past employment.  Again, there are other elements that are not included in the whole‑of‑government accounts either, which is why it is important to look both at flow projections and at balance sheet measures. 

There is another problem with a lot of balance sheet measures. You can see this has shown up in the way in which the whole‑of‑government accounts report contingent liabilities and provisions, which are a useful source of information for risks to the public finances in things like nuclear decommissioning and clinical negligence.  If you change the discount rate that you apply to the future flow of spending, receipts or whatever it might be, you can get the balance sheet number moving around quite dramatically, but in a way that is not telling you anything terribly useful about how the underlying health of the public finances has changed since the last time you looked.

Q198       Kit Malthouse: I understand.  On this desire to return the public finances to balance by the middle of the next Parliament, as I think was the pledge, are we on track?  It is hard to tell because the chart cuts off and it is hard to see the glide path thereafter.

Robert Chote: Formally speaking, we do not forecast far enough into the future.  This is on the basis that, when that target was set, the end of the next Parliament was presumed to be 2025.  Obviously fate has intervened since then but, if you stick with that as the meaningful horizon, all we can say at this stage is that it is not clear, on current policy and judging from where you are at the end of our forecast, that we are on course for that yet.

Q199       Kit Malthouse:  Have you had discussions with the Treasury about what your interpretation should be of policy now?

Robert Chote: They have to give us a definition of policy as far as the end of our forecast.  They do not define it beyond that.  What we have is a situation in which the budget is still in deficit by 1% of GDP or so at the end of the forecast period.  If you were simply to make the assumption that the deficit continues to shrink at the sorts of rates that we have assumed beyond the end of the Spending Review, it is 2030 rather than 2025 when you get to that point.  You are also dealing over that timeframe with a period in which there is likely to be age‑related upward pressure on expenditure.  Although I cannot point you to a number and say, “No, it is not going to be balanced; it is going to be X.X,” it is safe to say that, on the basis of the information and the policy state at the moment, we cannot say they are on course to achieve that.

Kit Malthouse: 2025 is out.  It is looking more like 2030, on a best‑guess basis.

Robert Chote: As the forecast extends, the Government will have to decide what they want to assume, as they have done with this forecast.  They have assumed a cut in public expenditure as a share of GDP in the new final year of the forecast.  Obviously they would have to make those decisions.

Q200       Kit Malthouse: Are you expecting a formal change of target or policy, or will we just drift along, effectively ignoring the targets?

Robert Chote: It is the Government’s choice what policy it is.  We have to judge against it as best we can.

Q201       Kit Malthouse: Would you say that the target is of no use now?  Is anybody paying attention?

Robert Chote: In general, I would not go that far.  It is not for us to say what the targets should be; it is to judge against them.  The fact is there is something that the Government have committed themselves to.  Admittedly, targets change and they commit themselves to new things at different stages.  We transparently say how close they are to that, and you can see that there have been occasions in the past when the Government have changed policy in order to bring themselves back on course to achieve a particular target.  The key difference between us and the Bank of England is that we do not set policy.  We do not even say, “We believe that the Government should do what is necessary to hit the target for a balanced budget in 2025.”  That is for you and for them, not for us.

Q202       Kit Malthouse: Given that the target and the mandate are moving around, are we effectively doing things on a discretionary basis now, rather than having a rules‑based target system?

Robert Chote: The rules are never airtight on that basis and that is not unique to fiscal policy.

Kit Malthouse: Ten of the last 12 rules have been completely ignored.

Robert Chote: Rules inevitably come to the end of their natural life and are replaced by other ones.

Chair: That is very diplomatic.

Q203       Kit Malthouse: Are they more like a fiscal equivalent of forward guidance?  They are almost certainly likely to be wrong, but they give you a bit of comfort for the moment.

Robert Chote: You raise an interesting question.  There will be some people who would say: if you have an independent forecast, why do you need rules at all?  The Government are showing by the policy they have declared that the numbers in here are what they want them to be.  Arguably, the rule is of greater value if it is telling you about the reaction function: if you move away from that path, what would they do in response to it?  It has never been articulated in that full way.  Governments, this or previous ones, have never accompanied their fiscal rules by saying, “This is what we want to achieve and, if it falls short of that for reason A, B or C, we will respond with policy 1, 2 or 3.”  That would be a different role for these things.  Most countries have them and they never work perfectly.  They all die a death at some point, but they have periods of usefulness while they go along.  I think that would be the consensus view. 

Q204       Stewart Hosie: Sir Charles, you are forecasting that, after falling sharply to near record lows in early 2017, the household saving ratio will continue falling and stabilise near zero from 2020.  The outlook states, “The saving ratio cannot continue falling indefinitely and we expect it to stabilise over the medium term.”  Does that mean that you think there will be a cash saving ratio of effectively zero?  Is that inevitable and is that sustainable?

Sir Charles Bean: No.  First, there are some wiggles in the saving ratio in the short term, to do with dividend income and the way the taxes on that are levied and so forth.  The forecast had some further modest decline in the saving ratio, but the broad underlying picture that we have is of it pretty much stabilising around where it is at the moment. 

The point that the saving ratio cannot keep on falling is an important one.  The only way you can do that is by building up more and more debt and, ultimately, that would be unsustainable for households, so that is the reason for thinking that there is a limit to how far down it can go.  Obviously if you have temporary fluctuations in income, you might have short‑term movements in the saving ratio, but you would not expect it to settle at a very low rate.

It is fair to say that there are clearly risks around the saving rate.  In my time following the British economy, there has been the perpetual power of the British consumer to surprise me on the upside.  We may well find more strength in consumption coming from some further falls in the saving ratio.  To the extent that is associated with a build‑up of household debt, one might start getting worried about some of the consequences of that.  Of course, there is also the possibility that, at some point, particularly if the environment starts becoming more uncertain as households see it—say unemployment started rising sharply—you might see a reversal in the decline of the saving ratio that we have seen.

In the last economic and fiscal outlook that we produced in March, the scenario analysis that we had at the end, which was the analogue of the productivity scenario that we had this time, was exactly an analysis of two alternative scenarios of the path of the saving rate.  As the consumer is such a big part of the economy, even relatively modest variations in the saving rate can end up having very big effects on activity in public finances. 

Q205       Stewart Hosie: I will come back to the issue of consumption versus debt and what happens in the case of a rapid or medium‑term reversal in the debt ratios.  You mentioned that this is unsustainable, and I agree, but you also spoke about the risks.  What are the key risks in terms of the saving ratio staying low?  How might they be crystallised?  Would they be symmetric across the population or would certain people suffer far more from a loss of savings than others?  How would those risks crystallise in economic terms?

Sir Charles Bean: This is a case where you want to go beyond the aggregate picture.  I cannot remember which page the balances are on.  I will try to find them.  Page 87 has the household net lending position, which is broadly in balance over the forecast.  The household sector as a whole looks okay in our forecast, but that is perfectly compatible with there being some households that may be building up a lot of debt and exposing themselves to risks in the future.  On the other side, there are going to be households that are saving more. 

Q206       Stewart Hosie: On that first group, are there enough of those households and are they close enough to the edge for you to be looking at some kind of systemic risk issue?

Sir Charles Bean: I have to say that it would not really be us who would do that.  It would be the Bank.

Q207       Stewart Hosie: You would look at the economic consequences of a systemic risk.

Sir Charles Bean: Yes, and I would certainly expect the Bank to be monitoring that.  There have been one or two comments in recent financial stability reports, as I recollect, implying that this is something they have a close eye on.  Clearly if this was a big enough group of households involved, it could have macroeconomic implications for our forecast.  If you get that subset retrenching very sharply, it would impact on overall spending.

Q208       Stewart Hosie: In terms of the bearing that the whole population or a subset would have on your growth forecasts, to what extent do the forecasts rely on consumption growth and a further reduction in savings, rather than investment and trade?  How much of the growth you are forecasting is effectively based on people spending and reducing their savings?

Sir Charles Bean: The share of consumer spending in GDP is pretty constant over the forecast period.  Consumption is growing pretty much in line with GDP.  You would not say that this is growth driven by an impulse primarily coming from the consumer sector.  It is reasonably balanced growth.  It is weak growth, and that reflects the downgrade in productivity and the slow rate of growth of household incomes but, in terms of the consumption, investment and net exports balance, they are all contributing roughly equally.

Q209       Stewart Hosie: Robert, the forecast also notes that “real consumption growth did not slow as sharply as real income growth, perhaps because households were slow to take on board the consequences of the depreciation for their future living standards”.  Given that you expect consumption to continue outpacing income, are you implicitly assuming, or even explicitly assuming, that households continue not to take that depreciation of their living standards on board?

Robert Chote: We are certainly continuing to assume that, for some while at least, consumer spending grows more quickly than consumers’ incomes.  That is implicit in the slight further fall in the pension‑adjusted saving ratio, and then it stabilises.  To what extent that is reflecting that particular question about whether they have taken on board the likely impact of that, versus a desire to smooth their consumption for other reasons or what their expectations might be about future prospects for pay increases for other reasons, is up in the air.  That can certainly be part of the explanation. 

What partly motivated that comment was, in the period immediately after the referendum, you were expecting to get a squeeze on consumers and the contribution that consumer spending would make to growth, as a result of higher inflation.  That was offset, to a degree, by the fact that people were willing to borrow more to shield themselves from that squeeze for a while.  We would not expect that to go on, because that squeeze comes out once the exchange rate effect has dropped out of people’s annual accounts.

Q210       Stewart Hosie: It is that period of correction that I am looking to.  We saw with small businesses during the banking crisis a huge degree of forbearance, and now we have incredibly low interest rates, including for individual personal consumers.  If the real interest rates rise, and not even necessarily by that much, there could be a very rapid correction to the saving ratio.  In essence, people stop spending. 

Robert Chote: It is true.  It obviously depends on what is driving the increase in interest rates.  If you are in a world in which interest rates are rising because people have had positive surprises on income growth, that is less of a concern than if interest rates are rising because of risk‑premium‑related reasons that are not matched by that.  It depends in part on what is driving that change, but that is certainly a vulnerability.

Q211       Stewart Hosie: As a final part of this, Sir Charles, the sectoral net lending forecasts show continued falls in Government net borrowing, being accommodated in part by household net saving being stuck at historically low levels, as we have been discussing.  To what extent has the reduction in public borrowing been swapped for increases in household borrowing, and is there any direct link or correlation between the two?

Sir Charles Bean: There certainly is a correlation. 

Q212       Stewart Hosie: Is it causal?

Sir Charles Bean: That is the much harder thing to tease out.  My story for the last few years would be that consumers retrenched at the time of the financial crisis.  We know that the saving ratio shot up then for perfectly understandable reasons, with the rise in unemployment and the slowdown in income growth.

Stewart Hosie: That was for a very short period.  It shot up, but it did not stay high for very long.

Sir Charles Bean: On the contrary, it stayed high for quite a while and then has been coming down gradually.  That is the sort of mirror, at least partially, of what has been happening to government borrowing.  As the Government have been borrowing less, households have been saving less.  That is the flip.

Q213       Stewart Hosie: Let me ask you the final question on this, just to crystallise that.  Do you believe that the Government can bring public sector net borrowing effectively to zero while household saving rates go up and remain healthy?

Sir Charles Bean: That depends on what is happening with the corporate sector and, for that matter, the rest of the world. 

Stewart Hosie: We could see both high government borrowing and low personal savings, which would be a terrible combination.

Sir Charles Bean: Possibly, but suppose there is an autonomous recovery in investment.  Now companies start dis‑saving.  You could have a world where households might be saving quite happily and also government borrowing has come down to relatively low levels to balance or whatever.  One of the puzzles at the current juncture is why corporate surpluses are as large as they have been.

Q214       Stewart Hosie: I have never heard capital investment in the corporate sense described as “dis‑saving” before.  Thank you for that. 

Sir Charles Bean: It is effectively.  It is using your funds to spend productively.

Chair: We like new words in the Committee.  Charlie has a very brief question, and then I will bring Catherine in.

Q215       Charlie Elphicke: On page 49, chart 3.8 on productivity growth, as you will see, the successive forecasts made over the last seven years have all been hopelessly wrong.  If I can take you to page 61, there is a fan chart that basically forecasts that economic growth may either be as high as 4%, which I do not recall it ever being in the last 50 years, or as low as ‑1.5%.  Is there any point in doing forecasting?

Robert Chote: Yes, there is.  You have to plan.  You, the Government and Parliament, are spending people’s money or taxing people’s money, managing the economy at that level.  There is a role for forecasting in deciding, with all the recognition of the uncertainties and the risks that lie about this, what you think the implications of the decisions you are taking with other people’s money are.  That is worthwhile.

There is also a sense of value in this from accountability—being able to explain what the narrative is and what you think you are doing.  This is not true just of the economy.  Virtually every policy decision and every law you are passing is presumably based on: “This is what we think the world will look like in the absence of this decision; this is the decision we are taking and this is what we think will result as a consequence,” whether you are investing in education, changing welfare policy, defence of whatever it may be.  In the economic sense, it is a detailed, quantitative setting down of this, and we go into considerable detail not because we are triumphantly more confident than other forecasters that we can do everything to the first decimal place but because we are showing our working.  We are setting down, as you have asked us to do, a clear analysis of the public finances, but we spend an awful lot of time explaining the uncertainties around this. 

The interesting thing is, at one level, you could just say, “Well, of course this could all turn out differently.”  One of the key things we always try to do in chapter 5 is to say which bits, if they are wrong, matter to what the Government have decided they are trying to achieve in terms of fiscal targets.  There may be some things that we could be quite wrong about but that do not matter a great deal to the underlying fiscal position at the end.  For reasons of accountability and for good policy, this is a sensible thing to do. 

On the economy side, having a forecast is a disciplining exercise of setting down what you think your understanding of the world is so that, as new information comes in, you have a benchmark to say why it is interesting and what it is telling you that is new and different.  A clear and explicit recognition of the uncertainty around all that is appropriate and necessary.

Q216       Charlie Elphicke: Let’s take paragraph 3.16 and the words in brackets.  It says of this fan chart, “The chart implies a roughly one-in‑four chance that the economy will shrink in calendar year 2018 or 2019.”  Effectively you are saying there is a one‑in‑four chance of a recession.  Do you really think that is the case?

Robert Chote: On the basis of the average of past forecast errors, if you were a Martian, you landed and you saw just what we have said now and the size and distribution of the errors when we and our predecessors have said that before, that is the conclusion you would reach.  It is not a subjective probability distribution of us saying this is what the particular set of risks and uncertainties looks like today.  We try to explore that in other ways as well, but it is an important reminder that recessions do come along.  As we said in the fiscal risks report, there is a roughly even chance that you get a recession in a given five‑year period looking ahead.  Obviously Governments and central banks do all they can to avoid those sorts of circumstances, but it would be unwise to plan on the basis that, to coin somebody’s phrase, we have abolished boom and bust.

Sir Charles Bean: On the fan chart here, the set of forecast errors that it is calibrated on includes periods of unusual stability, being the period before the financial crisis—the so‑called Great Moderation—and financial crisis.  Both of them are in there.  If we had only calibrated it on the Great Moderation period, we would have nice narrow fan charts.  We think that would give you too much comfort.  As Robert said, we know bad shocks come along now and again.  It is valuable to be reminded that nasty things can happen and good things can happen.

Chair: That goes back to where we started about trade shocks.  Mr Parker, was there anything you wanted to say on any of this?  No?  Okay.  Kit.

Q217       Kit Malthouse: Basically, what you are saying is, by producing a fan chart, your middle‑range forecast is almost certainly wrong.

Sir Charles Bean: Absolutely. The one thing about forecasting we are forever trying to get across is that you should not expect economists to be able to make point forecasts that are going to be fulfilled.

Q218       Kit Malthouse: What is that joke?  If you get five economists in the room, you will get five views or six if one of them has gone to Cambridge.

Sir Charles Bean: You are right—absolutely.  If you have a pair of dice, the central expectation of what is going to happen is seven.  If something other than seven comes up, you do not say the forecast was wrong.

Q219       Kit Malthouse: You are basically saying that we are playing tennis.  We are trying to serve the ball.  We have to get it in this bit, but we are more likely to get it somewhere in the court.

Sir Charles Bean: Absolutely.  The value of the exercise, exactly as Robert said, is to expose the risks.  Where things can go wrong is the value in the exercise.  It is not in the central forecast.

Chair: I do not know if you are playing your tennis any time soon with Mr Malthouse.  Rather than move to the esoteric, let us just move back to a couple of specific final questions from Catherine.

Q220       Catherine McKinnell: As an Opposition MP, I have found your analysis of the uncertainties, but also the parameters of the forecast, incredibly helpful.  I wanted to ask particularly about local authorities.  This year, you included a particular box on local authority funding.  In previous years, the boxes have been allocated to the results of self‑incorporation and trends in consumer credit, so it would seem to indicate that there are some particular concerns in relation to local authority funding, although I do not want to put words in your mouth.  I would like you to explain to the Committee what your analysis has been of the use of reserves and their drawdown by local authorities previously and obviously your forecast for that as well.

Robert Chote: Let me ask Graham to say a bit about the detail.  The reason for highlighting it at this stage is our experience has been, for some years, assuming the squeeze on the central government grant income that local authorities receive, they were likely to start using up some of their reserves earlier.  We have been surprised, year after year, by the fact that local authorities, despite facing this pressure, were putting more money into their reserves rather than less.  There are a variety of reasons why you might think that is the case, not least their uncertainty about the future state of their finances and, therefore, wanting to put some away.

We have, however, seen it move in the other direction.  That is one of the reasons for highlighting this.  It has been a trend that has been there for a while that has suddenly moved in the other way, so we highlight that.  There is also an interesting contrast with the contrary story on the central government, where you have bigger underspends than anticipated. Here you have a different story, but let me hand over to Graham.

Graham Parker: The first thing is the reserve story is key to our forecast for a local authority, because we basically forecast by looking at major sources of income and then assuming they spend that.  The only difference, and hence the contribution to the overall deficit, on the current side anyway, is the use of reserves.  You just forecast the council tax, business rates and grants from central government, and assume they spend it.  Then you get these swings from year to year when they were adding to reserves.  It was easily £2 billion or £3 billion going into reserves, so we could be wrong by quite a bit.  We have been forecasting that they would start using reserves at some time and it has finally happened. 

The box is really trying to look at that a bit more and illustrate the fact that these were different types of authorities.  The larger authorities, having social care responsibilities, were drawing down from reserves, whereas a lot of the district councils that had no such responsibility were still adding to their reserves.  There are particular things about the GLA as well to do with TfL, Crossrail and things like that, which have complicated things, so they are a special case.  The box is meant to illustrate why this is important and the uncertainties in the forecast.  Really there are still quite a lot of reserves out there and we do not really know what is going to happen over the next few years.

Q221       Catherine McKinnell: I guess the key question is: are local authorities using reserves to fund day-to-day services?  The chart would seem to indicate that social care is a major cause of the drawdown of reserves.  We know that there is a social care precept, but there are huge questions about whether that is sufficient to fund the day-to-day social care needs as well as the other obligations of local authorities.  Do you share the Local Government Association’s concern that the use of reserves is a one‑off solution that merely buys time?  It does not address the systemic underfunding that authorities face.

Graham Parker: It is not for us to say what is the right thing for local authorities to do in this, but I do not think continued drawdown of reserves is sustainable.  It could happen for a bit longer.  There are still quite a lot of reserves out there.  There is still something like £23 billion for English local authorities in their reserves, but a lot of that is earmarked for certain areas of spending, so they cannot spend it on general day-to-day pressures.  Clearly there is an imbalance.  Some authorities have no reserves at all, whereas some authorities could carry on drawing down reserves for a few years.  Others just cannot, as the LGA says. 

Q222       Catherine McKinnell: Do you have any concerns about the sustainability of local authority funding in some areas?

Graham Parker: We do not look at individual local authorities.

Catherine McKinnell: What about overall?

Graham Parker: Overall, we are assuming that we are only going to have modest drawdowns of reserves in the next few years and a return to a no‑change‑to‑reserves scenario at the end of the forecast period.

Robert Chote: That is basically saying that there may be a pressure here that is greater than the local authorities anticipated and that may be permanent, but it takes time to get into a position where you can deal with that in a sustainable fashion.  In the intervening period, you need to take the reserves down somewhat to do that.  To have the capacity to do that may of course be part of the reason that they were putting the money into the reserves in the first place, so that they gave themselves some greater freedom at that time.  Your underlying question of whether you can get into the position that they are content with where they are on a sustainable basis is a live one, but it is not one that we comment on directly.

Q223       Catherine McKinnell: The Chancellor announced, in my view, a few tweaks around the edges to the Universal Credit policy.  What key messages would you like the Committee to take away in terms of the impact of the announcements and the forecasting?  Are there any?

Graham Parker: We have big problems with the universal credit forecast.  You would expect this, because it is such a major change.  In the pre‑UC world, with all the legacy benefits it is replacing, there are quite detailed models for each and they rely on quite good and up‑to‑date data on caseloads, average amounts, etc.  There are lots of data out there and now you switch to something completely different. 

At the moment, we are having to forecast on a marginal cost basis, so we look at the legacy benefits as if Universal Credit did not exist; then we try to estimate the marginal effect of Universal Credit on top of that.  That is getting more and more difficult.  We would obviously like to switch to a proper full‑cost basis, but it is very difficult to do that, particularly at this stage of Universal Credit.  As more and more people come on to Universal Credit, it makes the data on the legacy benefits less reliable.  We have to try to add the cases that are now on Universal Credit back on to the legacy benefit, so we have the proper pre‑Universal Credit forecasting base.  This is quite difficult and it is probably going to get worse as more and more people get on to Universal Credit.  The data that are currently available to us from Universal Credit are not really as good as we would like them to be to enable this putting it back on to a legacy benefit basis, so that is our major concern with the forecast. 

As we say in the book, there are some other concerns as well.  It currently takes too long to run the Universal Credit forecast.  Most bits of the fiscal forecast we can turn around in a couple of days, but to do the full marginal cost forecasting for Universal Credit takes a couple of weeks, because it is so complicated, because they use the whole process for the operational needs as well.  Those are the main concerns on the forecast.

Q224       Catherine McKinnell: It may or may not reassure you that you are not the only people having difficulty with Universal Credit.  As an MP representing an area that has had the roll-out and one of the first pilots, there has been difficulty in terms of the complexity of the system and rolling it out.  In terms of how that impacts upon your forecasting, what do you see as the future of the assumptions that you will base your forecasting on as the system becomes clearer and more efficient?  Will you be doing any work on the impact of Universal Credit on the labour market and your forecasts in that regard as well?

Graham Parker: Yes, we would like to.  Again, it is a bit early yet.  We have not put anything in the forecast yet about the effect on Universal Credit, but we will be working with DWP to see if there is any evidence for it.  If there is any evidence, we will bring it into the forecast.

Robert Chote: One of the difficulties with that is that you start off with small areas where this may be being piloted etc., but can you scale the answers you get from those relatively small studies up to a much larger scale and assume that it applies in the same way?  There is always that degree of challenge there.  We are going to be producing a welfare trends report in the new year—I am not quite sure exactly when yet—which will go into the universal credit forecast in a good deal more detail.  As Graham said, the underlying issue for us is this is a very large chunk of public spending, around which there is considerable uncertainty about the ability to forecast as we go through this transition period. 

Q225       Catherine McKinnell: The changes made in this Budget to Universal Credit have not particularly impacted upon your forecast.

Graham Parker: We obviously signed off the measures and incorporated them into our forecast but, as you say, they are not big compared with the actual size of the overall spend on universal credit.  One consequence of the measures is a further delay to the roll-out, when we have had to make several delays already.  It is now probably up to five years behind where it was originally thought to be.  That is a major consequence.

Catherine McKinnell: It is absolutely clear that we are not very clear.

Q226       Chair: We would welcome hearing from you when that welfare trends forecast or work is done, so we will put a marker there to invite you back.  I just want to finish on the stamp duty changes for first‑time buyers that were announced, because one of the lines in your forecast is about the potential increase in house prices as a result of that change—the cutting of stamp duty for first‑time buyers of properties under £300,000.  In this case, you thought it was around 0.3%.  That is pretty marginal, is it not?  Is that within the margin of error?

Robert Chote: Yes.  You are talking there about the average effect across all houses.  The effect on the prices of those that are transacted by first‑time buyers is greater than that, but obviously the vast majority of housing transactions do not involve first‑time buyers.

Q227       Chair: The 0.3% is across the whole market.  There was also the suggestion that, in putting that line in and coming up with that 0.3% figure, the OBR had not taken account of all the other changes in the comprehensive housing package that had been announced that day.  What was your view on that?

Robert Chote: That is an explicit number for this measure.  In terms of what you can incorporate from the other things, there are issues around the speed and certainty, both of implementation and of impact.  With a change like stamp duty, it is in the Government’s power to do this.  They can do it quickly, and it is an unusual feature of property taxation like this that you crystallise the long‑term impact of the lower tax bill.  It shows up crystallised upfront relatively quickly, not instantaneously, in the price, and therefore comes out there.

A lot of the other policy areas, so planning reform and areas where you are attempting to catalyse private sector activity, have the question of it not just being the Government who is the player here.  What do local authorities do, etc.?  There is a greater uncertainty about the implementation and the impact, at the end of the day.  It is also more likely to be later in the forecast period.  The Chancellor is perfectly reasonable in saying he would expect this to have some sort of benefit and it moves in the right direction from the perspective of increasing housing supply, but it is the sort of thing where you necessarily have to wait to see this show up in the data, alongside what else is going on in the housing market.  With the stamp duty change, it is an unusual tax in the sense that it crystallises as an upfront impact.

Q228       Chair: As the OBR, did you warn the Treasury of your concerns about the house price impact?

Robert Chote: They are not really concerns.  That is the impact that we said it had.

Q229       Chair: Did you discuss it with the Treasury before?  Was there anything more than, “This is a line in our outlook”?

Robert Chote: With every policy that comes along, you are doing a costing and we are probing and challenging the analysis that the Government come to us with.  All these sorts of issues arise then, and we similarly look at the indirect effect of all the measures on the economic forecast as a whole.  As I say, the fact that the implications for affordability look as they do rather mirrors HMRC’s conclusions when they did their evaluation of the temporary first‑time buyer window so, in that sense, I do not think it would have come as a surprise.  The elasticity that we use, which is what determines the 0.3% as 0.3%, and not larger or less, is a number that we had published beforehand, so I do not think there would have been anything to startle the horses in any of that.

Chair: Lovely.  Thank you all very much indeed for coming in and giving evidence this morning. I particularly thank Graham, who I think is due to step down next year.  I am not sure whether we are going to see you.

Robert Chote: It is not his swansong yet. 

Chair: In that case, we shall look forward to seeing you whenever the next opportunity is.  We may have trends in welfare, and there will no doubt be spring reports as well but, for now, thank you all very much indeed for your evidence this morning.