Official Report

  • Finance Committee 28 March 2012
    • Decision on Taking Business in Private
    • Economic and Fiscal Outlook (United Kingdom)
      • The Convener: Under item 2, we will take evidence from Robert Chote, the chairman of the Office for Budget Responsibility, on the United Kingdom economic and fiscal outlook. We will hear from Mr Chote on the OBR’s latest economic and fiscal outlook, which was published alongside the budget last week, and the OBR’s recent methodology note and forecast on Scottish taxes that might be devolved to the Scottish Parliament. I welcome Robert Chote and invite him to make a short opening statement.

      • Robert Chote (Office for Budget Responsibility): Good morning, everybody—it is a great pleasure to be here. Thank you for inviting me. By way of advance apologies, if my voice goes and I become unresponsive later in the proceedings, that is through the flesh being weak rather than the spirit being unwilling. I will do my best. I am beginning to think that I should be more worried about my eyesight than my throat, but there we go.

        I will begin by saying a little about the economic and fiscal outlook, and I will then turn to the Scottish tax forecasts. A comparison between the current set of economic and fiscal forecasts and the ones that we produced at the time of the chancellor’s autumn statement back in November last year shows that, fundamentally, there has not been a great deal of change. The budget was largely neutral in its impact on the economy and public finances. The balance between the giveaways and takeaways in tax and spending measures does not exceed £2 billion in any year of the five-year forecast, so we are talking about an addition or subtraction to spending power in the economy of perhaps 0.1 per cent of gross domestic product. Therefore, it is not a significant budget judgment in the sense of being something that will have an enormous impact on the path of the economy in that period. However, we have made a couple of adjustments to elements of the forecast to reflect specific measures, such as excise duty changes that have an impact on inflation.

        In addition to the neutral budget, economic and fiscal data have been performing broadly as we expected in the period between November and now. The economy has performed much as expected. As we anticipated, it shrank by a little in the last quarter of last year, but it seems to have come into 2012 with a little more momentum than we had previously anticipated. On the fiscal side, we have not changed our forecast for Government borrowing this year by a great deal, although that masks some slightly more significant changes in spending and tax. Government departments appear to have spent significantly less than they planned to and than we anticipated they would in November. In a sense, we have had to revise down departmental spending by about £6 billion this year. However, counter-weighing that is the fact that it looks as though tax revenues will come in about £5 billion weaker than we anticipated in November. A large part of that is to do with the fact that receipts from self-assessed income tax have been particularly weak in the past couple of months. Therefore, those two effects largely offset each other.

        The starting point is fairly unchanged from the situation in November. The big picture remains that we anticipate a steady and unspectacular economic recovery in the next five years, by the standards of recoveries from past recessions. That is reasonable given that we are coming out of a financial crisis; credit conditions are not yet back to normal; fiscal consolidation is providing headwinds; and consumers and businesses are dealing with their balance sheets. We have a steady recovery.

        The OBR is slightly more optimistic than the average of independent outside forecasters and slightly more pessimistic than the mean forecast from the Bank of England’s inflation reports—we are somewhere between the two. We have a forecast for the economy to grow by about 0.8 per cent this year. That is fractionally up on the November forecast simply because of the greater momentum coming into the first quarter, but fundamentally there is not a great deal of change. We have fractionally nudged down the expected growth rate in 2013, but other than that the picture is pretty much unchanged.

        There has been a little bit more change in the composition of GDP, but nothing dramatic. We have taken a slightly more pessimistic view of the outlook for business investment. In the short term, that reflects the fact that business investment numbers are very volatile on a quarter-by-quarter basis, and there was quite a steep fall in the fourth quarter of 2011 that has a sort of arithmetic, automatic knock-on effect into the weakness of calendar year 2012 growth.

        Looking further out, we have got slightly less dramatic investment growth in the final years of the forecast. That reflects a further examination of corporate balance sheets and our belief that official data may be somewhat overstating the amount of cash piled up in corporate balance sheets that is just sitting there waiting to be invested. There is a fair amount of cash out there, but it is not clear that it is in the hands of the sort of companies that might go out and engage in capital investment to quite the extent that the official figures suggest.

        We have looked back at the performance of investment growth coming out of the previous recession. Again, it seems to be sensible to see a picture that shows a robust recovery but one that is not quite as strong as the previous recovery, so we have business investment picking up by about 40 per cent over the next five years, compared with the 50 per cent increase in the equivalent five years of the previous recovery.

        Offsetting that weaker investment, we have slightly stronger growth in consumer spending. In the short term, that is partly because of payment protection insurance pay-outs, which are a kind of mini-equivalent of building society demutualisations in providing some households with windfalls that they might go out and spend. In addition, asset prices are slightly stronger in this forecast than in the previous one, so that edges things up a bit. Within a broadly unchanged growth picture, there is a little bit of reallocation between consumer spending and investment, but I would not overstate the size of that.

        We have stuck with our view of the evolution of the potential output of the economy, which is very important for discussing the structural health of the public finances. Again, we assume that the potential growth of the economy takes a couple of years to return to its long-term trend growth rate. We assume that that fits with the idea of credit conditions still needing to normalise for a little bit. We end up with a path for potential GDP over the next five years that is broadly in line with the estimates of, for example, the Organisation of Economic Co-operation and Development and the International Monetary Fund and which is a bit more optimistic than that of the European Commission.

        In terms of the amount of spare capacity in the economy, we estimate that the economy was running at about 2.7 per cent below full capacity in 2011, which is bang in line with the average of outside forecasts, although there is a fairly large range from about -0.5 to -4 per cent, which has quite a big impact on the assessment that is made of the size of the structural deficit versus the cyclical deficit at the moment. Overall, the picture is little changed from that of November.

        If we look at the path for the budget deficit, again that is not dramatically different from the picture that we painted in November. We expect to see the deficit continuing to shrink as a share of GDP over the coming five years at roughly the rate that it has done over the past couple of years. The one dramatic change since the last time is that we have a big one-off fall in public sector net borrowing in the coming financial year, which simply reflects the fact that the UK Government has decided to take the Royal Mail historic pension deficit and some associated assets on to the public sector balance sheet. In the mysterious world of public accounting statistics, that has a variety of counterintuitive effects, one of which is a one-off reduction in Government borrowing for a year. We will also see, as the assets are then sold, a step reduction in public sector net debt of £23 billion or so after a couple of years.

        However, it is important to look at the long-term picture. Although the transaction looks very favourable to the public finances over the time horizon that we focus on in the report, we need to bear in mind that the assets that the Government is getting from the transaction, which amount to £28 billion or so, are outweighed by £37 billion or £38 billion of liabilities—if you think of that as the estimated up-front one-off equivalent number to the flow of payments to Royal Mail pensioners that the Government will have to make in future. In that sense, the transaction looks negative for the public finances, because the liabilities outweigh the assets.

        On the Scottish tax forecasts, as you know, in the context of the Scotland Bill we were tasked with the role of producing forecasts for the four taxes that are proposed for devolution—or, in the case of the aggregates levy, pending devolution, legal obstacles being overcome. We have done that for the first time. As I think that I said when we met informally, I was keen that we should produce the forecasts in as transparent a way as possible, so we put out a methodology paper, which explained how we intended to do it, ahead of the budget. We have had useful discussions with Her Majesty’s Revenue and Customs, which is doing most of the actual number crunching, as it has expertise in the tax areas concerned, and with representatives of the Scottish Government.

        As you have seen in the paper on the Scottish tax forecasts, we have ended up in a situation in which we are assuming that we try to identify some share of the relevant UK tax base and apply that in future. In some cases, such as income tax—or the share of the particular definition of income tax that we are talking about in this context—that looks quite attractive, on the ground that the relationship has been very stable. Income tax is the largest element in the Scottish forecasts. In other cases, where the relationship is less stable, we have ended up assuming a fixed proportion that is carried forward—in large part because there is no alternative, more sophisticated approach that ends up with a better answer—rather than saying that it does not move around much.

        We then had discussions as the pre-measures forecast was coming through—again, we ensured that the Scottish Government was represented at the discussions—to talk about its application to particular sets of numbers that we had. There was an issue to do with taking into account measures that had been announced in previous budgets, which might have impacts on the particular tax ratios, so the discussion was useful.

        Of course, we were not able to talk to the Scottish Government about the implications of the measures that were to be announced in the most recent budget—they were held very tightly between the UK Government and national newspapers, so it was not possible to have such discussions. Therefore, we have applied HMRC’s estimates of the direct impacts in that regard. We might want to go back to that and think about whether we want to look in a more sophisticated way at whether the measures are likely to change the long-term share of, for example, the Calman wedge to the UK tax base.

        As you will see from the report, we produced a set of figures. The numbers for 2012-13 range from about £4.5 billion for income tax right down to a healthy £43 million for the aggregates levy. We very much view the forecasts as work in progress for the time being. This is a learning experience, for us and for everybody. In particular, the budget that we have just had gives the opportunity for quite a nice natural experiment of going back to look at the forecasts, because it contains precisely the sort of measures that might give rise to worry about an asymmetric effect between the UK and Scotland—I am thinking about the high stamp duty for extremely high-level properties and the 50p tax rate moving to 45p.

        Therefore, there is a deal more work to be done. However, we hope that we have set out as clearly as possible what we have been trying to do. We will be interested to hear members’ and other people’s views on how to do it better or differently in future.

      • The Convener: Thank you for your informative opening statement, which has no doubt kiboshed many of the questions that members were preparing to ask.

        I want to focus on Scottish taxes. There is an issue to do with the accuracy of forecasting when you are depending on a fairly two-dimensional measure. I am thinking about income tax, for example. If we consider the June 2010 budget forecasting errors for UK receipts in 2010-11, the forecast was 2.2 per cent below the outturn for income tax and 4.9 per cent above the outturn for landfill tax. Even within the limit of the forecasting tools that are available to you, it is clear that there is margin for error, as there always is in forecasting.

        Over time, will it be possible to develop further tools to look at the matter in more detail? It is clear that, because of the lag and so on, looking at income tax receipts will not help with accurate long-term forecasting.

        09:45
      • Robert Chote: Yes. Each year, we undertake the forecast evaluation report exercise, which involves our going back to look at the performance of the previous year’s forecast to see what we can learn from how things have turned out.

        However, we must recognise that forecasting such revenues is partly an art as well as a science. One has only to look at the sorts of changes that have been made in the budget—which, admittedly, have been relatively dramatic—because of the way in which we had information coming in on self-assessment. Along with HMRC, we had the particular difficulty of working out how much forestalling was going on with the 50p income tax rate and to what extent people had been shifting income. We now have the problem of the rate going back to 45p. Will that encourage people to move income from one year to the next? Then there is the underlying issue of what behavioural response we get to some of those policies and what impact they will have on genuine labour-supply decisions, people’s decisions on how much to work, when to retire and—in extreme cases—whether to leave the country. Other responses include tax planning, tax avoidance and tax evasion. The recent budget represents a particularly dramatic case.

        We must bear in mind that there are two elements to forecasting accurately. The first is whether one can forecast the underlying economic determinants of these things—in other words, what the economy is doing—accurately. The second is whether, given one’s knowledge about the state of the economy, one can forecast the level of receipts that will come in from that or, on the spending side, how much spending there will be.

        Generally speaking, in normal times, the economic error tends to be relatively small and the fiscal error tends to be relatively large. It is on the fiscal side that the errors show up. When there are dramatic developments in the economy or there is a turning point between recession and recovery, it tends to be the other way round—much of the error is made as a result of a failure to realise by how much incomes were going to go up or down and the fiscal error is relatively small by comparison. On this occasion, there have also been some policy changes that have made the fiscal issue more difficult.

        We will certainly look at how we can improve our forecasting in the future. One might hope that the fact that Scottish taxpayers will be flagged by HMRC will give a better guide on income tax, although, as we have said, the share of Scottish income tax—or the Calman wedge—as a proportion of the total UK base has been very stable, so I do not know how much of an improvement that will result in. One might hope that there will be an improvement.

        The broader issue for the longer term—which depends on structures of Government in the future—is whether you end up with independent policy, with the result that you need a Scottish macro forecast to underpin it. We are not in a position to provide that, of course. Finding out how much difference that would make to long-term forecast accuracy would be a suck-it-and-see exercise. A reasonable judgment could not be made on that in the short term. We will continue to look at that and to see how we can do things better.

      • The Convener: Paragraph 2.15 of “Forecasting Scottish taxes” states:

        “The Scottish share of non-savings, non-dividend income has been very stable”,

        yet paragraph 2.19 states:

        “Employment grew more strongly in Scotland than in the UK as a whole in the years prior to the recession”.

        Why did the increased employment rate not appear to lead to an increase in the overall share of income tax?

      • Robert Chote: That may be because of changes in the distribution of incomes—in other words, changes in the way in which income per person was changing or changes in the relative distribution of incomes at different tax bands. As you say, it is striking how stable that has turned out to be. We have included in the methodology note an estimate of sensitivity to relative differentials in terms of differences in employment and earnings rates. Those differences are not enormous, so one would not expect them to be driving things too much. If employment and earnings are going in opposite directions at the same time, the effects can cancel each other out.

      • The Convener: Will the OBR look at the issue of borrowing? As you know, the resource limits will be only £500 million, with an annual cash limit of £200 million, which is quite small given the overall budgets that will be in play. Professor David Bell states in his report to the committee that

        “the possibility of larger worst-case forecast error due to longer forecast horizons, poor OBR forecasts”—

        which we hope will not happen—

        “and the uncertainty of the inflation adjustment to the borrowing limit suggest that the proposed borrowing limits are too low to adequately protect the Scottish Government against the risk of revenue shortfalls. And this does not allow for the cumulative effect of successive years of over-optimistic forecasts.”

        Inflation—although it is not mentioned—is not allowed for, either. What is your view on the issue of the £500 million? Is that limit too tight? Should there be more flexibility?

      • Robert Chote: That matter is beyond our formal remit. The task that we have been given is to produce the forecast for particular taxes, and it is for those well above our pay grade to judge how that forecast should be used in terms of longer-term adjustments and how it should be applied to the borrowing issue.

        The borrowing guidelines were, I think, set out before we had begun to think about the forecasting methodology, and it would not be unreasonable if people were to look back at those and ask, given the methodologies that are being applied, what sort of room for manoeuvre might be needed and whether that which has been provided is adequate.

        I guess one issue—I am getting out of my area of expertise—is whether the adoption of the Holtham method of having some indexation to the differential movement in the Scottish income tax base versus the rest of the UK income tax base might insulate the need for borrowing against some part of the error if there is a difference moving on that side. There is also the possibility of building up cash balances.

        We have not been asked to reach a judgment on whether the number is adequate, but that number was out there before we had had much chance to discuss the methodology. I would not be surprised if people wanted to return to that, to look at what the methodology would have yielded in the past, and to think about whether the numbers are appropriate.

      • Elaine Murray (Dumfriesshire) (Lab): I seek clarification on the fiscal forecast. You said that there was an upward revision to the GDP forecast of 0.1 percentage points. What is the range of uncertainty on that figure?

      • Robert Chote: Around the GDP forecast?

      • Robert Chote: It is significant. The range of forecasts for 2010 GDP growth—I do not know whether you have that to hand—is a good example. In the June 2010 emergency budget, we had a forecast of around 1.2 or 1.3 per cent. At the time of the November 2010 report, that was up to 1.8 per cent. At the beginning of 2010, the range of independent outside forecasts for growth in that year was between about 0.7 and 2.2 per cent. That gives you some sense of the range that people were operating in.

      • Elaine Murray: There is not really much difference between 2.1 and 2 per cent, and the difference is insignificant in terms of the range of uncertainty on the figures.

      • Robert Chote: Exactly.

      • Elaine Murray: So there should not really be any joy in UK Government circles that the figure has risen by 0.1 per cent, because that is within the range of the uncertainty.

      • Robert Chote: As there should not be despair that the 2013 figure has gone down by 0.1 per cent. An important point to bear in mind is that once the year is over, the Office for National Statistics will retrospectively change its mind about how much the economy has grown. The first estimate that it produced—early in 2011—for how much the economy had grown in 2010 was 1.4 per cent. The current 2010 growth estimate is 2.1 per cent. The rewriting of history on its own shifts the figures.

      • Elaine Murray: That begs the question whether the numbers should be quoted to that degree of accuracy if there is such a range of uncertainty, although perhaps that is a nerdy point.

      • Robert Chote: No, that is a perfectly legitimate issue. That is one of the reasons why the Bank of England tends to draw fan charts and does not publish precise numerical estimates of how much different components of spending will grow by. However, we do, because the fundamental task that we have been given is to increase transparency and trust and our job is to show our working. The fact that we publish many more numbers than the Bank of England includes in its inflation report does not indicate that we think that we are smarter than the Bank of England; it just indicates that we need to show our working because it is about whether people trust the numbers as being a professional judgment rather than politically motivated wishful thinking. It is not the same concern that somebody would go to the inflation report with.

      • Elaine Murray: I have another question on the fiscal forecast. You say that the self-assessment tax return is down by £3.6 billion. Do you have any idea why that should be?

      • Robert Chote: It is partly down to the fabled 50p tax rate.

      • Elaine Murray: Were people not coughing up for their taxes?

      • Robert Chote: There are two effects to disentangle. First, when the 50p rate was introduced, the Government announced that it was going to introduce it but the implementation took place some time later. A consequence of that is that people appear to have shifted about £16 billion of income that they would have been taxed on in future years into 2009-10 in order to be taxed at the lower rate. For most of us who are on fixed salaries, that would not have been an easy thing to do, but for a company director who is able to decide with their colleagues when they are going to be paid dividends, it would have been easier. That is a dramatic number and, as that effect unwound, there was more taxable income in 2009-10 and less of it than would have been expected in subsequent years.

        Secondly, there is the related issue of whether the underlying behavioural response to the 50p rate was as big or as small as was originally estimated. According to the HMRC analysis, which we have looked at carefully and drawn lessons from for the forecast as well as for the costing of the move to the 45p rate, the evidence suggests that there has been a bigger underlying behavioural response—not just shifting across time—than the previous Government expected when it introduced the 50p rate. The best estimate is that, instead of raising the £2.7 billion a year that was originally assumed in the 2009-10 forecast, it looks likely to raise between £0.7 billion and a little over £1 billion, depending on how the calculations are done. That effect also shows up.

        Any judgment of the self-assessment return for the first year in which the 50p rate was implemented that is based on not even a full year will be pretty heroic, but we have to make that judgment. In particular, we have to make a judgment because the Government has decided to change the rate again and we need to be able to say whether we believe that its estimate of what that will cost or raise is reasonable and sensible. Those two elements help to explain the weakness in the self-assessment return forecast. Some of that showed up in PAYE in previous forecasts, but we did not have the evidence at the time to say that it might be down to the 50p rate.

      • Elaine Murray: My final question is about the tax forecasts and the landfill tax return, which is expected to grow over time. That worries me a bit because the Government’s zero waste policies are intended to reduce the amount of waste that goes to landfill. Why do we expect the landfill tax return to increase over time? Surely, if our waste reduction policies are successful, that tax take should decrease over time.

      • Robert Chote: That figure is based on Scotland having a constant share of the UK figure. No view is taken on the relative effectiveness of policies in different parts of the UK and how they are likely to change the figures. The higher tax rates in themselves will generate higher revenues even if the underlying tax base does not change. We have taken the best advice that we can from the relevant departments about their expectations regarding the quantity of landfill.

      • Elaine Murray: I am not sure what the UK Government is doing on waste these days, but there is a strong desire here, across all parties, to see the amount of waste that goes to landfill decrease significantly in order that we can achieve our climate change targets. I would be quite worried if we were assuming a financial picture that depends on something that is rather at odds with our policies.

      • Robert Chote: Again, you are back to the difficulties of coming up with behavioural responses. As the rate is raised, more revenue is received, offset by the impact of the behavioural change. An element of judgment is therefore involved.

        10:00
      • Paul Wheelhouse (South Scotland) (SNP): Welcome again, Mr Chote. I have been very interested in your evidence. I would like to look at some of the detail in the UK forecasts, if I may, so that I understand the implications for Scotland. Table 1.1 on page 11 of the economic and fiscal outlook sets out assumptions about the economic forecasts for key indicators. I was drawn to the general Government investment figures, which show the impact of the capital reductions in 2011, 2012 and 2013. I am intrigued by the drop in 2016. Paul Johnson of the Institute for Fiscal Studies suggested that another spending review might be needed and that there might be other decisions. Do the assumptions about further cuts in capital funding reflect that?

      • Robert Chote: Basically, we have current and capital spending plans laid down fairly clearly for the current spending review period to 2014-15, and we assume that they will be delivered. In contrast to demand-led elements of expenditure, such as debt interest and social security payments—which we forecast—for the spending of Government departments on public services, public administration and associated capital spending, we take the spending review numbers and include them with some adjustment in the short term if it looks like there is underspending or overspending.

        In the two years subsequent to the spending review, we have basically gone to the Government and asked what its policy will be over that period. We then applied that, because there is no spending review. The Government has set down a policy in respect of the total managed expenditure, which is the broadest aggregate of total spending. I will simplify that somewhat; it has said that it assumes that spending will continue to grow at the rate that it grew in the spending review period with a few twiddles to the baseline. We then take away our forecasts for the demand-led bits—social security and so on. That leaves an amount of money for the departmental expenditure limits, so there is that component, which kind of falls out in the last couple of years.

        Another issue that must be borne in mind and which helps to explain why the contribution of general Government consumption and investment changes in a national accounts sense, is that we must distinguish total public expenditure, which we always think about as a share of GDP. That is a good metric to measure it, but not all public expenditure contributes to GDP: the bits that are counted are, essentially, production and consumption of goods and services, not the taking of money from one citizen and giving it to another. There, in terms of its impact on economic growth, we are looking at the real change in Government spending or Government consumption of goods and services.

        One thing that we have noted in the data so far is that it appears that the cuts are showing up more in the price than in the volume of Government expenditure in that area. The fact that it looks now as though we have a greater contribution from, for example, general Government consumption and investment in the bottom half of table 1.1 than we had in November is not because the Government has loosened the spending plans; rather, it is because nominal spending is turning out pretty much as expected, but more of it is showing up in prices than in output, in a national accounting sense. Therefore, that nudges the figures up. That is an additional confusion—as if things were not complicated enough.

      • Paul Wheelhouse: Thank you for that.

        I want to confirm that my understanding is correct about another thing that I was drawn to. In table 1.2, which is a fiscal forecast overview, you show that public sector net debt—to reduce that is obviously one of the UK Government’s targets—is forecast to decline from 76.3 per cent in 2014-15 to 74.3 per cent. Obviously, there are a number of underlying factors, but what is the main driver for that?

      • Robert Chote: The main driver is the shrinking of the deficit over the period. If you are considering whether net debt will fall in 2014-15, you need to consider what the level of the debt will be in 2014-15 and what will be going on in terms of interest rates and, therefore, what will be the stock of the Government’s debt and what will be the change in the size of the deficit.

        The Government has a supplementary target to ensure that the ratio of public sector net debt to GDP falls from 2014-15 to 2015-16, so back in November the Government took some measures to help to ensure that it continues to be on course to achieve that. Our best view is that the changes and the underlying changes in the forecast—the underlying budget measures—have not moved the position very much; we will still see, as in November, the debt to GDP ratio falling by about 0.3 per cent of GDP in that year. The notable difference is that the whole level of public sector net debt is lower than it was in November, which is because of the contribution of the assets from the Royal Mail pension transfer.

      • Paul Wheelhouse: Thank you for that. I was also drawn to ask a question on your assumptions about unemployment. In table 1.1 of the executive summary, you show International Labour Organization unemployment figures peaking in 2012 at 8.7 per cent, and you forecast a steady decline thereafter, with the rate of decline picking up particularly towards the end of the forecast period. Can you share with the committee what you have assumed about regional differences in unemployment? What would be the assumption in terms of the Scottish aspect if there is, if you like, policy off—that is, if there are no changes in policy from now?

      • Robert Chote: I am afraid that we do not produce regional—or, indeed, industry—breakdowns. That is beyond our remit. Our remit partly reflects the fact that the forecasts are used fundamentally to drive fiscal forecasts for which such regional dimensions are less important—although they are obviously very important for other reasons.

        We do have an implicit breakdown of the figures for the public and private sectors. The relative importance of the two sectors obviously varies regionally, so it might be possible to feed those figures through. We come up with a projection for the total change in general Government employment over the period of the forecast, which we do by examining the components of the public spending forecast that can be used to employ public sector workers. That includes the bulk of Government departmental expenditure limits plus things like the budget for the BBC, but it does not include debt interest, social security and so on.

        That projection represents a decline, from the beginning of 2011 to the beginning of 2017, of about 730,000 across the UK as a whole. That projection is about 20,000 higher than we had in November and it reflects the fact that the pool of money available to pay people will be slightly smaller.

        Further, public sector earnings growth has been slightly less quick than we had expected from recent data. So, people will be slightly cheaper, but the first effect outweighs the second effect. You can then compare that number with the overall change in employment. Assuming that private or market sector employment rises by about 1.7 million and general Government employment falls by about 0.7 million, that leaves you with a 1 million change.

        People may want to draw some conclusions for regional differences from that if they wish to. However, as I said, it is beyond our remit to do regional breakdowns.

      • Paul Wheelhouse: Thank you for that helpful answer. My last question is about the two scenarios that you looked at in the executive summary. In the first you consider potential changes in the euro zone area, which everyone is probably quite familiar with, and in the second you consider a short-term oil price shock. You say that that would have a relatively modest impact on the projections. However, can you expand on what you assumed in your central forecast about oil prices and production levels? Can you also give some guidance on the impact on the projections of a longer-term rise in oil prices?

      • Robert Chote: We take a futures curve and assume that it is the best view of oil prices. In quite a lot of our previous forecasts, the futures curve has tended to be flat from whatever level it started. If we compare the forecasts in November and now, we see a much bigger increase in the spot price than in the longer-term price. We exaggerated it as a purely illustrative way of presenting a scenario. We assume that the recent increase in oil prices is not locked in permanently and that prices will move back down in line with the futures curve.

        We consulted the experts at the Department of Energy and Climate Change who have detailed knowledge of production, capital expenditure and the other elements that affect the receipts picture—it is not just production that drives the situation. We accept that they are the experts on that; they also talk to the industry in coming up with estimates.

        As Paul Wheelhouse suggests, the scenario that we show is that another short-term spike followed by a reduction would have a relatively modest impact. Receipts from the North Sea would have a direct boost, but there are countervailing effects from higher inflation, for example, which pushes up debt interest costs, and there is an impact from weakening in the economy more generally. Overall, the effect would not be dramatic. If the price remained higher for longer, we might expect some of the negative effects to become more pronounced. I have no doubt that we will come back to the issue as the price evolves further.

        For many years, the received wisdom on the fiscal impact has been that everybody notices that receipts from the North Sea go up when oil prices go up, but costs go up for everybody else, which depresses revenues. The idea that higher oil prices are a clear gain to the UK—albeit a shrinking gain, as production shrinks—is not as straightforward as it appears at first to be.

      • John Mason (Glasgow Shettleston) (SNP): My questions follow on a little, but not entirely, from that answer. Where does the exchange rate come into inflation? Is it assumed to be part of the inflation picture? Is it assumed that, if the pound goes down, inflation will probably go up? Do you look at the exchange rate separately?

      • Robert Chote: We do not project the exchange rate—that falls out of relative interest rates in the UK and the rest of the world. There is no assumption about a great change in the exchange rate. However, you are right: if we look backwards, one reason why inflation was relatively high and was depressing economic activity was the impact of imported costs. Some of those costs—for example, for energy and food—were driven by market-specific factors, but they had an import element, too.

        Evolution in the forecast as time has gone has occurred because we have had a period in which relatively high inflation has squeezed real disposable income, which has in turn squeezed consumer spending. We assume that inflation will continue to fall, partly because past increases will fall out. For example, domestic energy prices went up. They have come down a bit but, more important, the years after they went up by a lot will fall out of the annual comparison.

        Over time, inflation will come down and the labour market will continue to pick up. However, not until 2013-14 will we get back to a situation that we might think of as being more normal, in which wage increases outpace price increases by a significant margin, consumer spending starts to act as a driver for the recovery, rather than—as it is at the moment and will be for some time—being a relative drag on it, and a bigger contribution comes from investments and consumer spending.

        If inflation does not drop as rapidly as expected and the labour market performs in the same way, we might expect more of a squeeze on disposable incomes and we might expect there to be consequences for consumer spending. The pace and sustainability of the decline in inflation need to be looked at closely in thinking about potential risks to the central forecast.

        10:15
      • John Mason: I am thinking in particular of oil prices. In recent years, the pound has been quite weak against the euro—certainly compared to what it used to be—and oil is largely marketed in dollars. I presume that if the pound falls against the dollar, our fuel prices immediately go up, even though we are an exporter.

      • Robert Chote: That is right. We have been focusing on the implications of the sterling oil price for inflation.

      • John Mason: Did you wish to ask a supplementary, convener?

      • The Convener: Yes. With regard to table 1.1 in the executive summary of the economic and fiscal outlook, you rightly say that it will be 2014 before there is any significant above-inflation increase in incomes, and you have forecast inflation at 1.9 per cent in 2013 and 2014 and 2 per cent in 2015 and 2016. Why have you chosen those figures? Is it because you are unable to predict the level of inflation beyond that and are simply sticking to targets?

      • Robert Chote: That is down to a combination of things. In the short term, certain elements of the recent increases will drop out and the margin of spare capacity that we believe is in the economy will, as you might expect, put downward pressure on domestically generated inflation: it will push it down towards the target. In the longer term, although there will still be spare capacity in the economy, there will be a little bit of offsetting upward pressure from unit import costs. Moreover, you might argue that, on that kind of horizon, there will be some anchoring of inflation on the assumption that the bank will go for the inflation target. That seems to be consistent with the IMF’s analysis of what it has called PLOGs—a rather ugly abbreviation that stands for persistently large output gaps—and their consequences for inflation, which are that it is possible to get it down and stabilise it at a relatively low level. For the time being, although the output gap has not been closed entirely, it is not unreasonable to assume that inflation will be broadly consistent with the target.

      • The Convener: Does your forecast for average earnings differentiate between the private sector and the public sector?

      • Robert Chote: For the purposes of the macro-forecast, we have taken a broad overview of the whole labour market. With regard to the projection for general Government employment, we have to look a bit more at what is going on with the pay bill per head in the public sector, and worry a bit more about the picture of overall settlements, pay drift and so on and how such factors affect the number of people you get for any given £1 billion chunk of money that you spend on public sector workers. For those purposes, there is a bit more focus on the minutiae but, as I have said, for the macro-forecast we have taken an overarching view of how you might expect wages to perform, given the amount of spare capacity in the economy and the underlying performance of productivity growth, predominantly in the private sector.

      • The Convener: Obviously the 4.5 per cent forecast growth is an amalgam of the public and private sector figures but, given the prospect of regional pay for public sector workers, do you predict a differential in growth between the sectors over the next few years? Will the bulk of growth in average earnings happen in the private sector?

      • Robert Chote: We certainly take into account the announced policies on public sector pay in the short term—including, for example, the pay freeze in many areas and the 1 per cent increases that will come after that—and a return to a more normal historical relationship between public and private sector wage increases in the longer term. I do not have the precise numbers on me, but I can certainly dig them out for you. Basically, in the longer term, we assume a return to a historically normal relationship between the two, instead of trying to guess what public sector pay policy will be in the years subsequent to one having been announced.

      • The Convener: Indeed. I call Gavin Brown, to be followed by Mark McDonald.

      • John Mason: Have I finished my questions, then, convener?

      • The Convener: I must apologise to the deputy convener.

      • John Mason: He interrupts me and then he ignores me. [Laughter.]

      • The Convener: Mr Mason is very sensitive.

      • John Mason: You mentioned the Royal Mail, but I have to say that I find that particular scenario bizarre. We understand that you are limited to how far forward you can go; those particular assets come in, which will affect public debt. However, are we saying that the liabilities lie so far ahead that we can just forget about them for the purposes of the forecast?

      • Robert Chote: We do not forget about the liabilities. Our forecast over the five-year horizon looks only at a limited period of the liabilities’ crystallising, in terms of the additions to spending that they imply. The amount of pension payments is roughly £1.3 billion to £1.6 billion a year, in the years of the forecast. That is a small proportion of the total. We did, though, feel that it was important, on the ground that looking over the five-year horizon gives a skewed picture of the overall gain or loss of that to the public sector, to point out that if you look at an estimate of the present value of the future payments, stretching out some way into the future—not into the indefinite future, as the set is particular—the present value of the liabilities exceeds the present value of the assets.

        One thing that should be borne in mind is that turning the flow of year-by-year pension payments into a one-off cost crucially depends on a discount rate that is used to turn 10 quid in 15 years’ time into X quid today. That is also true when you look at the burden of public sector/public service pension payments as a whole, which we do in more detail in our fiscal sustainability report that looks at whole-of-Government accounts.

        A lesson of the first set of whole-of-Government accounts, on which we commented last summer, is that that rate can change, depending on the accounting convention that is used to come up with it. There can be quite a big change, which can make the burden of future pension payments look dramatically bigger or smaller than it would otherwise seem. In the whole-of-Government accounts, the interest rates that high-quality corporate borrowers are able to borrow at are used for a discount rate.

        In the latest set of whole-of-Government accounts, the burden of public service pensions appeared to have increased significantly over the previous years. That had nothing to do with the expected future year-by-year payments; it had everything to do with the fact that the rate of interest at which high-quality corporate borrowers could borrow had changed, and therefore the interest rate, which was used to crunch all that back into a single number, had changed as well.

        We have to put a health warning around the estimate of the future value of the liabilities. Nonetheless, the public finances situation looks more negative if it is looked at holistically than if you just focus on a two-year to five-year horizon.

      • John Mason: It is good that that has been highlighted, because the value of future pensions and so on has been a huge issue for the private sector. We will not spend any longer on that issue.

        From reading your papers and others I get the message that it is easier to forecast trends than it is to forecast step changes. Obviously, the recession was a big problem and caught everybody by surprise. Similarly, you have touched on the point that people’s behavioural responses—for example, to a 50p rate—are difficult to forecast, especially when comparing one year to another. Are we going to face the same problems if the Scottish Government does something different from the UK on income tax? Is that hard to forecast because we do not know how people would behave?

      • Robert Chote: Yes. It would bring in a different and additional set of uncertainties, depending on the tax; the fact that there may be a differential between the rest of UK rate versus the Scottish rate clearly could be significant. It is very hard to predict exactly how important that will be as an additional source of forecast error. If you were to decide not to have any of the 10p additional rate, or a much more dramatic one, you would have a big differential and you might judge that the behavioural response would be particularly important. If the difference, in the end, ends up being not particularly great, that may not be a huge source of additional error. The more different the policy choices in Scotland are from those in the rest of the UK, the more financial incentive there is for people to change their behaviour for good or ill, so the task of forecasting the outcomes ends up being more complicated.

      • John Mason: Moving money from one year to another is, as you said, quite difficult to do with salaries, so perhaps it will not be so much of an issue. You already mentioned the possibility that people might leave the UK because of a 50p top rate of income tax, so if our rate was 1p less than England’s, people might start coming here.

      • Robert Chote: As you can imagine, the size of the differential would be quite important in determining how strong that effect would be.

        The issue of people shifting income arises when a change is announced but not implemented immediately. That is what happened with the 50p rate. Leaving aside any long-term judgements that people will make about tax avoidance, retiring or fleeing the country, if the Government announces a tax change that makes it worth while for people to have or spend income in year X rather than in year Y, it creates a whole new set of behaviours. However, it is a one-off change. It washes out and comes on in the future.

        For us, the task of forecasting what will go on with income tax revenues over the next few years is complicated by the fact that we had one of those time-shifting policy changes when the 50p rate was introduced and will have another one going in the opposite direction with the 45p rate, because that is not being implemented immediately either. However, there is a qualitative difference in the nature of the change. When the 50p rate was introduced, there were many years in the future out of which somebody could take income to stick it into the one year in which it needed to be to beat the 50p rate, but with the lag in the introduction of the 45p rate, it is really about getting the income out of 2012-13 and sticking it into subsequent years.

      • John Mason: People will still hope that the rate might even come down further

      • Robert Chote: Exactly—so we would expect the time-shifting response to be considerably smaller in the case of the 45p rate than with the 50p rate, partly because the differential in the tax rates is half the size but also because the time window from which there is an incentive to shift money is smaller.

      • John Mason: My overall point is that there is uncertainty about how people’s behaviour will change because we have not had the powers before. Perhaps it will not change.

      • Robert Chote: Yes—and there is no experience from the existing tax-varying powers from which to draw conclusions.

      • John Mason: It is often said that Scotland and the UK have been mirroring each other over the past few years and that the percentage share of the tax take—be it income tax or the other taxes—has not varied much. If that started to change, how quickly would we pick it up?

      • Robert Chote: At the moment, we are reliant—particularly for income tax—on the information that we get from the survey of personal incomes, which, as we explain, has provided the information that justifies the view that the share of the tax take does not change much. There is a lag in that data; it takes some time to come in and we must think not only about how measures that were announced in the most recent budget will affect the relative share, but about how measures that have been announced since the previous survey of personal incomes, but before today, will affect it. We discuss that in the paper that we produce at the time of the budget.

        When we get to the point at which HMRC has flagged individual taxpayers as being Scottish or not, we will be less dependent on the information from that survey and we might get a more exact picture. I hope that that will mean that it is easier to determine the share in real time and that we will be less reliant on trying to interpret what a survey that is 18 months out of date might tell us.

      • John Mason: As we move forward, can we expect increasingly accurate data and forecasts?

      • Robert Chote: I will not pledge that to you. One effect may move in that direction, but lots of uncertainties will come from other directions.

      • John Mason: That is fair enough.

        You have mentioned the Calman and Holtham commissions and the different ways in which the block grant might be adjusted. Will that have a knock-on effect on your forecasting work?

      • Robert Chote: It will not really, because we provide the forecasts and it is not for us to say how that information should be used to determine the one-off block grant adjustment or the future adjustments. That is a matter for the Scottish Parliament and Westminster. We just provide some numbers that you can use for those purposes.

      • Gavin Brown (Lothian) (Con): The Economist recently had a lengthy article about how billions upon billions of pounds are being hoarded by companies just waiting to invest. You suggested earlier, though, that the official data is overstated in that regard, which certainly ties in with what I hear on the ground from companies that I speak to. Could you expand on how you reached that conclusion?

        10:30
      • Robert Chote: Yes. The main thing that we looked at in trying to interpret the situation was what the Office for National Statistics had. The ONS has the difficult task of deciding how much UK companies have on deposit with overseas financial institutions and then identifying what sort of companies those are. Basically, the ONS reached a judgment on the basis of a BIS survey that showed that about 80 per cent of what BIS identifies as being UK claims on overseas financial institutions is in the hands of companies, so it shows up in the national accounts estimates of the cash piles.

        Our concern is that that 80:20 relationship was derived some time ago. If we consider the growth of non-bank financial institutions—hedge funds and so on—the proportion of the money that may be in the hands of the sort of companies that you and I might think will potentially go out and invest in new machinery or capital equipment may be lower than 80 per cent, and rather more of it may be in the hands of, for example, hedge funds. Our concern is that the large pot of money that is out there just waiting to be invested may not be in the hands of the sorts of companies that would go out and invest in a way that would show up as gross domestic fixed capital formation in the national accounts.

      • Gavin Brown: That is helpful.

        You have a central forecast of 0.8 per cent growth for 2012. You forecast that we will avoid a technical recession but state that we should be wary of a negative quarter. Are there any quarters in 2012 for which it would not be a surprise if they turned out to be negative?

      • Robert Chote: I have a deal with my economic colleague at the OBR that we will never use the sentence, “Now is a uniquely difficult time to do a forecast,” on the ground that that is always an easy get-out. However, this is not an easy year for interpreting the quarter-on-quarter changes in GDP, primarily because of the impacts of the diamond jubilee and the associated bank holiday and the Olympics.

        In common with the Bank of England, we expect there to be positive growth in the first quarter of this year, but one should never rule out the possibility that it will be negative. We only have to look at how people were surprised by the unexpected decline in the last quarter of 2010 to see how these things can come out of left field. However, the first quarter looks relatively positive. We forecast weaker and basically flat growth in the second quarter, which implies a 50 per cent chance of it being positive and a 50 per cent chance of it being negative. We expect a rebound from some of the bank holiday effect into the third quarter and then for that to go back a bit.

        We try to come up with a best guess at the headline quarter-on-quarter changes but also an underlying measure that strips out those sorts of effect. If we look at the underlying measure, we have a positive, steady, unspectacular growth rate through the first three quarters of this year that will pick up towards the end, but there is a rather more zigzag pattern if we look at the headline changes. If we were looking for a particular quarter in which the danger of growth being negative is greatest, it would probably be the second quarter of this year.

      • Gavin Brown: Thank you. I will move on to the Scottish tax forecasts. I have a thinking-out-loud type of question. You described in your report what you forecast for 2010-11. I briefly compared some of those figures with the figures in “Government Expenditure & Revenue Scotland 2010-2011”—the Scottish Government produces an expenditure and revenue report annually. For landfill tax, the Scottish Government’s figure is identical to your figure of £99 million for that year. However, you said that the aggregates levy was £46 million, but the Scottish Government said that it was £54 million. Now, in the grand scheme of Government spending, an £8 million difference is quite small but, as a percentage difference, it is quite high. If it were translated to income tax in particular, it would be an issue. To what extent have you considered the methodology of the GERS report? Do you have any plans to consider where the differences might lie?

      • Robert Chote: We do. In our discussions with Scottish Government colleagues we have tried to ensure that there is consistency where possible in the broad methodology between what we are doing and GERS. In the case of landfill, a different data source is used here because of a difference in the way in which HM Revenue and Customs is picking that up. I will go back and check. I think that GERS came out relatively recently—

      • Gavin Brown: Yes, just last month.

      • Robert Chote: So it was subsequent to our report. I am not sure whether the difference is because the same data source has moved or because we are using slightly different ones, but that is exactly the sort of thing that we want to keep a watching brief on as we go ahead, just to check that there is consistency between the two. However, it may be that there is a slightly different data source.

      • Paul Wheelhouse: On a point of clarification, my colleague Gavin Brown indicated that it was the Scottish Government that produced the GERS figures. My understanding is different. Will you confirm that GERS is an Office of National Statistics publication and that ministers do not see it until it is produced?

      • Robert Chote: I think that that is correct—it is a national statistic, and the United Kingdom Statistics Authority did a report saying that it shared that view. It is no part of our job to say whether GERS is correct but, as you say, it is something that the Statistics Authority has confirmed it regards as meeting the requirements of national statistics.

      • Mark McDonald (North East Scotland) (SNP): Not to labour the point about the 50p rate, but you said that it is not just the initial year’s income that could be transferred back but future years’ incomes. Will you comment on that? How far forward could an individual transfer their income to avoid paying the 50p top rate or another top rate?

      • Robert Chote: If you are talking about company directors who are able to decide the timing of dividend payments, you could certainly think of that as being a multiyear exercise. If memory serves me correctly—I would have to go back to the specific HMRC analysis—there was an assumption of roughly two to three years of a diminishing effect but with most of the action happening in the year closest to the change, in which it is easiest to transfer the money. There was some significant and measurable estimate for a couple of years prior to that year. I do not think that HMRC made any adjustments beyond that. In principle it could go a lot further than that but, obviously, if you were trying to model how such changes were working, it would be a relatively rough or difficult exercise.

      • Mark McDonald: You said, that with a new tax rate coming in, people might move their income into the following year. Could that be done cumulatively so that their income is pushed much further forward to avoid a top rate?

      • Robert Chote: It could, in principle, but it is not clear that the incentive is as great to do that. As I say, there is the one year you want to get income out of, whereas, when the rate is going up, there is the whole future period in which you might want to bring income forward. The nature of the incentive is somewhat different there.

        Clearly, you might want to shift income out of the coming year, not just into the following year but into subsequent ones, but the incentive is not quite as dramatic to do that. The assumption that is implicit in the forecast is that the shift of income will be significantly smaller than the shift of £16 billion to £18 billion that HMRC estimates took place around the 50p rate. We are talking about something closer to £6 billion.

      • Mark McDonald: I am not having a dig—I realise that you are not responsible for the policy—but it strikes me as odd that the top rate of tax can effectively be dictated by clever accounting tricks rather than by anything else.

        I assume that most of the Scottish assumptions are based on maintaining the status quo with policy, rates and so on. With regard to the powers that are being devolved to Scotland through the Scotland Bill, have you carried out any scenario planning for the impact of different policy angles that might be taken?

      • Robert Chote: We have not done any scenario planning on that basis. However, if policy changes, HMRC will take that into account in its forecast.

        In the UK context, it is relatively straightforward to decide when a policy is a policy; it is announced as a measure in the budget red book. For example, when the coalition came in, there were noises early on about moving the tax allowance threshold to £10,000. However, we include in our costings and forecast only the progress towards that in announcements in individual budgets about the setting of the allowance. In one paragraph of the document, we highlighted the coalition’s statement about the desire to put this policy in place; when we asked the Treasury about it, it said that there was no commitment to achieve it by a particular time, so we flagged it as an area in which a policy announcement had been made but the announcement was not firm or precise enough about size and timing to be included in the central forecast. An issue that we have not explored and which I will probably have to look at is what, in the Scottish context, constitutes a policy declaration that is firm enough to be taken as unchanged policy and therefore considered in the forecast.

      • Mark McDonald: I guess that, as far as stamp duty and the property market in particular are concerned, it is important to have the most up-to-date figures available, because you want to be able to react to market changes. Given that powers over stamp duty will be coming to Scotland under the Scotland Bill, how are you seeking to improve things in that respect and get the most up-to-date information available for Scotland?

      • Robert Chote: With stamp duty, the fact that you are dealing with immobile properties and that you know where those properties are gives you some sort of advantage over the slightly more difficult and complicated task of defining whether certain individuals are Scottish or non-Scottish taxpayers. I presume, of course, that it is slightly easier to locate a physical structure.

        We are relying on HMRC having the data available and certainly improving it as best it can over time. Of course, an uncertainty that has arisen with the changes to the 50p rate—and this affects the UK as a whole—is the number of properties out there that are worth £2 million or more. We have cited evidence that suggests that only 1 per cent of Scottish properties fall into that category; given how small that element is, it is a potential source of asymmetry. Nevertheless, in all of these cases, HMRC is ensuring that it has the best available information and we will certainly be very receptive to any suggestions for doing things better and will pursue with the body how best to do that.

      • Mark McDonald: If the Scottish Government decided to be creative and do something different with the slate of taxes coming to Scotland under the Scotland Bill, that would have implications for the block grant. Would it be worth while to plan out scenario A, B or C for the implications for the block grant if, say, a different set of reliefs were offered for stamp duty or a different stamp duty policy were pursued?

        10:45
      • Robert Chote: That would depend on knowing how dramatic the change in policy was going to be. We would have to look at whatever the system was. It is clear that the income tax slice will be different, but the broad structure of the system will not change dramatically. When Paul Johnson gave evidence, I do not know whether you asked him whether he thinks that stamp duty is a well-designed tax. If you did, I can probably predict what his answer was. It is not my job to reach judgments on this any more, so perhaps the committee could come up with something that Paul Johnson might regard as a better-designed way of taxing property transactions. You would have to look at that substantively and ask what it would yield; you would need to take the whole structure of the policy seriously.

        The same would be true of a power for a new tax. That would have to be looked at in isolation to see what it would generate, because we need to continue to produce a UK-wide forecast for the fiscal aggregates. We need to continue to be concerned about that for our broader remit, not just in relation to whatever input we can provide for the purposes of changes in Scotland.

      • Mark McDonald: The other day, Reform Scotland issued a press release that called on the OBR to produce much more detailed and up-to-date figures on Scottish taxation, given the debate that is taking place on Scotland’s future. Are you aware of that? Are you receptive to the idea?

      • Robert Chote: Our remit is set fairly precisely by Parliament. Any expansion of or change in our role would have to come from that source. Delighted though I would be to wade into the placid waters of Scottish politics, I must fall back on the UK Parliament’s instructions on what we should and should not look into. We need to stick to the mandate that we have been given. However, with the advent of the Scottish tax forecast, that mandate has been changed once already, and I cannot rule out the possibility that my political masters will change it again.

      • Mark McDonald: I am sure that we would welcome your input with open arms.

        The Scottish Government has a fiscal commission working group, which is looking at issues beyond just those that relate to the Scotland Bill. Have you had any dealings with that group? Would you be willing to work alongside it on some of those wider economic planning issues?

      • Robert Chote: We have not had any engagement with that group. Again, I suspect that that would fall outside the remit that we have been given, but I would be happy to chat with its members and see whether we can be helpful. I need to stick to the task that I have been entrusted with by Parliament.

      • The Convener: It sounds as if some of the political parties here are happy to take instructions from the UK Parliament.

        Moving swiftly on, I turn to “The Scotland Bill Proposals for Forecasting and Reconciling Income Tax Receipts”, which was produced by Professor David Bell, who is the committee’s financial adviser. On page 12 of his report, he mentioned the mean absolute percentage error. He said:

        “it is not surprising that MAPE increases from 3.0% to 5.9% for forecasts for the same period”—

        which is a number of months—but

        “It increases further to 7.6% for forecasts made in the preceding year.”

        There is an issue about accuracy, which members have touched on several times. What macroeconomic data would the OBR need to have available to produce more accurate forecasts for Scotland? I realise that you have a specific remit. If you could get data from the Treasury, HMRC, the Scottish Government and so on, what additional methodologies would you require to give us the same level of accuracy that you can provide for the UK?

      • Robert Chote: One set of issues that arises is less to do with our ability to forecast and more to do with data availability. Some of the data relating to determinants for income tax—data on mixed income and self-employment income, for example—are, as far as I am aware, not available for Scotland. A breakdown of expenditure and income is being produced on an experimental basis with some lag and so on.

        The obstacle to producing a Scottish macro forecast as an underpinning for a Scottish fiscal forecast is not only a lack of resources to put together the model to do that—we would need to do it in a way that was consistent with the UK forecast as a whole—but the fact that some of the data that underpin the UK-wide forecast are not available at a Scottish level. The raw material that you would need to input into the model is not there. I am not fully up to speed with how much of that data the ONS or others intend to generate, but I think that you are some way away from having the data infrastructure available to be able to do something more dramatic on that score.

      • The Convener: Professor Bell has touched on a number of occasions on the point that we need to improve our data resources.

      • Paul Wheelhouse: The issue that I am about to address, which relates to the rather dull but worthy subject of stamp duty, has been dealt with in part by Mark McDonald. The methodologies that you have considered as alternatives for forecasting stamp duty are largely dependent on historical data and econometric modelling of previous trends relative to UK policy changes.

      • Robert Chote: Unhelpfully for Scotland, the data do not go back far enough.

      • Paul Wheelhouse: Yes. Notwithstanding your answer to Mark McDonald, can you provide any further detail about the current breakdown of receipts from stamp duty by the different rates? You mentioned that the top 1 per cent of properties in Scotland are worth more than £2 million each. Can we access data that would give us a breakdown of the figures?

        If not, is there any prospect of working with Registers of Scotland, which is the agency in Scotland that collects information on sales of both commercial and residential properties, to collect more real-time data than you would be able to do through your proposed methodology? Such data would give us a better understanding of how the different markets in Scotland and the UK are working, because there are differential trends in residential property, in particular, and recently you have indicated that there are also differential trends in commercial property. We may be doing quite well in that respect relative to the rest of the UK. Could you comment on those issues?

      • Robert Chote: I am not sure that I have any additional data. HMRC produces regular statistics that provide a breakdown of the information by tax bands and show how much revenue you are getting. The information shows that X number of transactions in a particular band are generating Y amount of revenue. I do not know whether HMRC plans to provide similar data broken down on a Scotland and rest-of-the-UK basis consequent to all of this.

        You or we could certainly inquire about that, because HMRC produces a substantial breakdown of the individual tax forecasts. It shows where the liabilities lie, in what sort of bands and so on, for not only stamp duty but other taxes. It might be worth seeing whether it is possible to extend that. Obviously, we are keen to do anything that we can to increase the transparency of what we are doing, subject to the availability of reliable data to do it.

      • Paul Wheelhouse: You might not see the data, so you are at a disadvantage, but members will know that we get almost weekly reports from Registers of Scotland that inform us of the number of sales in the previous month. The information is broken down to quite a detailed geographic level. Might it be possible to use that very accurate set of data—which is being improved as we invest in a new land registration process, so it should become more complete—to give more real-time understanding of what is happening with stamp duty so that we can model and project the figures forward?

      • Robert Chote: To be honest, I do not know the extent to which HMRC already makes use of that set of data and whether it has made the efforts that we have to establish whether there is a more sophisticated way of tackling this tax forecast. I suspect that, if that set of data is out there, HMRC is probably aware of that and will be looking at it. We could certainly follow up the matter.

      • Paul Wheelhouse: I would be grateful if you would, because I think that that would be very helpful to the committee.

      • The Convener: I thank committee members for their questions. I have one or two more.

        When we had our informal session last year, you talked quite a lot about the impact of demographics. How much of a part do demographics play in your forward projections?

      • Robert Chote: They play relatively little part in the projections that we produce in the economic and fiscal outlooks, as they have only a five-year horizon. We focus on demographics in the fiscal sustainability report, in which we do 50-year projections. We will return to that matter on 11 or 13 July—it is one of those two dates; the date is set and is in the report somewhere. We will want to see whether the population size and structure projections from the ONS that we use have moved significantly from those that we deployed last year. We will want to keep an eye on them and their impact on the projections. The main focus on demographics is in the fiscal sustainability report, simply because of the timescales.

      • The Convener: Earlier, I touched on borrowing. Professor Bell said in his paper:

        “Borrowing powers are an essential mechanism for managing risk. When considering the amount of borrowing available to the Scottish government, the size of the total amount should be guided by the principle that a Scottish resident is at no greater risk than an English or Welsh resident of having to adjust to in-year spending cuts that might be caused by forecasting error.”

        Do you agree with that statement?

      • Robert Chote: That seems to me to be a perfectly reasonable statement. As I have said, the borrowing powers discussion and the recommendations predated the early discussion of what the methodology might look like. I presume that the issue that you raise is one of the reasons for the discussion about the Holtham and Calman methodologies. It may be playing into that. We have not directly addressed the extent to which that approach in itself essentially helps to achieve that aim. However, as we see our experience with forecasting evolve as we get closer to the point at which that is done, that will provide more data on which to have an informed debate about whether the size of the borrowing powers is appropriate.

      • The Convener: I was going to ask you about that finally. Are you confident that we will have enough data to ensure that it remains the case that Scottish residents are not disadvantaged because forecasting errors are larger than one would hope?

      • Robert Chote: Decisions on the choice of Holtham on borrowing powers, for example, are outside our remit. All that I would say is that I would always caution people about our economic and fiscal forecasts. I am always the first person to point out how uncertain such things are. It is in the nature of forecasting that something will crop up that one did not expect or did not have experience of in the past. That is why the job is so entertaining to do. Insulating national economies of whatever size from such things is not straightforward.

      • The Convener: Absolutely. I thank you very much. The session has been fascinating, and I look forward to seeing you at lunch time.

      • Robert Chote: Thank you very much.

        10:58 Meeting suspended. 11:06 On resuming—
    • Youth Employment