Our second item of business is to take evidence on the United Kingdom economic and fiscal outlook. I welcome to the meeting Robert Chote, chairman of the Office for Budget Responsibility. Mr Chote will begin with a short opening statement.
Thank you very much indeed, convener, and good morning everybody. It is a great pleasure to be back and thank you very much for asking me.
I will open with a few remarks on how the forecasts that we produced at the time of the budget have moved since previous forecasts, both in terms of the economy and the public finances. Perhaps I should say something after that about what the forecast for the public finances over the next five years is implying for the change in the shape and structure of the state, relative to the pre-crisis period, in order that you get a sense of what the aggregate movement has been over the crisis period and the consolidation period on which we have embarked.
I will start on the economy. On the forecast that we produced in December 2013, by the time that we got to the budget it was clear that there was a bit more momentum in economic growth coming into 2014 than we had anticipated. The growth number for the first quarter of 2014 was actually the same as what we predicted back in December, but there had been upward revisions to previous gross domestic product data for 2013 that suggested that there had been more of a pick-up in the economy in 2013, following a period of relative stagnation in 2012 and earlier. We therefore elevated our growth forecast for this year up slightly to 2.7 per cent growth over the previous year. That was in line with the average of outside forecasts at the time that we made it, although the outside average has now nudged up another 0.1 per cent in the wake of subsequent data.
I guess that the main reason for the economy growing a bit more quickly than we had anticipated a few months previously is that consumer spending performed slightly more strongly. Notably, that seems to have been the result of people drawing down savings more quickly than we had anticipated, rather than the long-awaited pick-up in real income growth, which is important for the long-term sustainability of the recovery. Our forecast thus assumes that there will be a slight slowing in the quarter on quarter pace of growth throughout this year as consumer spending moves back into line with the underlying growth in incomes, rather than assuming a continued rapid fall in the proportion of people’s incomes that is being saved, which we would not expect to see continuing at the sort of rate that it was over 2013. We will therefore have a slightly slower growth rate in 2015 than in 2014, because of the way in which the arithmetic works. That is not a forecast that we are expecting the economy to run into choppy waters in 2015; it is about the fact of the slight slowing during the course of this year that we anticipate as consumer spending moves back more into Iine with income growth.
Looking out over the five-year horizon, we expect the economy to grow by about 2.5 per cent a year over that period. Needless to say, consumer spending is as always an important driver of that because it is the larger share of spending in the economy by some margin.
We have a pick-up in business investment. The numbers there have been revised up, showing stronger growth in business investment through 2013 than we had anticipated at the time of the previous forecast. What we do not have is any additional contribution to growth over the next five years from net exports, so we are not relying on additional improvement in net trade performance in order to deliver the growth pattern over the five years; it is much more down to consumer spending and the pick-up in business investment.
Government spending cuts act as a drag over the five-year period. We are actually surprised that they have not acted as more of a drag already, partly because of how public sector output is measured, but we are expecting that to be more the case as we move forward.
As I said, these forecasts are not a million miles away from the outside average. The Bank of England is slightly more optimistic about the pace of growth over the course of the year and is expecting 3 per cent plus, partly because, in addition to forecasting what is going to happen in the future, the bank also forecasts how the Office for National Statistics will revise past history. That is not something that we try to do. The main reasons for the bank’s optimism are that it expects a swifter pick-up in business investment than we do and slightly more consumer spending activity.
We think that there is a little bit less spare capacity in the economy than we thought back in December. That seems to fit with the fact that unemployment fell rather more rapidly in the early part of the year than we had expected, which suggests that some spare capacity was used up. As ever, however, there is enormous uncertainty over any estimate of spare capacity in the economy at any given time.
With regard to public finances, as the official data on borrowing will have been published and made available only five minutes ago, I do not know the outturn data for the latest set of numbers. However, the forecasts that we have produced suggest a picture pretty similar to the one we produced at the end of last year. Borrowing is slightly lower than we had expected, partly because of a strong showing from stamp duty as a result of activity in the housing market and partly because of lower inflation than we had expected, which reduces the cost of servicing index-linked gilts. We are also expecting the budget deficit to continue to shrink year in, year out over the next five years, primarily because of the on-going cuts in public expenditure as a share of GDP.
I think that it is worth going back to the pre-crisis period and contrasting where our forecasts now imply where, on current policy, the Government will be at the end of our five-year forecast in 2018-19 and where we were prior to the financial crisis, say, in 2007-08. We expect that by 2018-19 the overall budget will be back in balance for roughly the first time in 18 years. At that stage, taxation and spending would be at about 38 per cent of national income, which is not a million miles away from where they were the last time we had a near-budget balance, which was back in the early 2000s and early on in Labour’s term in office in the United Kingdom.
With regard to how we get to that balance, I think that there are some striking comparisons to be made with where we were in 2007-08 prior to the financial crisis. First, back in 2007-08, we were running a budget deficit. In the run-up to the financial crisis, the budget deficit would have been about 2 to 3 per cent a year over the period; the forecasts imply that the Government wants to get rid of that deficit, which means that it has to find 2 to 3 per cent of GDP from somewhere.
At the same time, we know that in 2018-19 we are going to have to spend a bit more on debt interest than we spent back in 2007-08 because of the amount of borrowing that has been going on in the meantime. We would also expect the welfare bill and social security spending as a share of GDP to be slightly higher in 2018-19 than it was in 2007-08, partly because retail price inflation has been relatively high over the period, which means that the generosity of many social security payments has risen relative to the living standards of people in work and the size of the economy. We are going to have to find a bit more money for that.
Given that the Government is not planning to raise a great deal more tax revenue than it was back in 2007-08, how will it get all of this to add up? The answer is a continued squeeze on public expenditure, which is where the bulk of the continued work on fiscal consolidation will happen over the next few years. Capital spending will be a bit lower in 2018-19 than in 2007-08, but the Government has indicated that it does not want to squeeze it that much further. The real squeeze, therefore, will come on the day-to-day non-investment running costs of public services. It is hard to make a long historical comparison here, but as a share of national income I think that that is heading to levels that we have not seen since at least 1948. If you are trying to close a budget deficit, spend a bit more on welfare and debt interest and not raise very much more in tax, you will need to squeeze the day-to-day running costs of public services to make everything add up.
I hope that those introductory remarks are of some use, convener.
Absolutely. I had intended to ask you a whole series of questions but, after those remarks, I will ask you a series of different ones instead. Colleagues will be pleased to hear that it will not be a whole series of questions, because, as always, I want to ensure that they have an opportunity to ask their own instead of my stealing all the juicy ones at the start of the session.
On your last point, is it true that about 80 per cent of fiscal consolidation will come through a reduction in public spending?
It depends on what counterfactual you are using—in other words, what you would have expected to happen in the absence of policy changes—but the answer is yes, that is about right. The Government has aimed for an 80:20 split in most of its consolidation plans, and the picture still looks pretty much like that. That in effect takes tax receipts back to where they were prior to the crisis and reduces spending sufficiently to get rid of not only the increase in spending that happened during the crisis but the budget deficit back in 2007-08. If we take those two things together, I think that you are right—the split is roughly 80:20.
As far as timing is concerned, I find it striking that most of the tax increase component of the plans has already happened; indeed, the increase in the standard rate of VAT was the main element of that. Given that we have had most of the tax increases and most of the cut in capital spending, we are left with the squeeze on spending on public services as the place where most of the remaining lift will come from.
I am sure that colleagues will want to explore that further.
You mentioned the levels of economic growth. I take it that we are more or less back to trend growth, but we do not seem to be making up for the years of lost output; after all, the economy shrank by more than 5 per cent at the peak of the recession. The Institute for Fiscal Studies has advised that real incomes will not reach 2001 levels until 2017-18. Is that about right?
Yes, that is roughly what we are showing. Indeed, that is what you would expect. You need to take into account not only the pace of economic growth but the increase in the size of the population over the period. GDP per capita, which is quite closely although not precisely linked to real living standards, looks weaker in comparison. In that sense, we are further away from reaching the pre-crisis peak than we would be if we did not take population into account.
Indeed. If the economy is growing at 2 or 3 per cent and the population at 1 or 2 per cent, per capita growth will be only 1 per cent.
In the overview of your executive summary, you say:
“The outlook for productivity growth, which underpins income growth and the sustainability of the recovery, remains the key uncertainty.”
Over the past few years, economists have been wrestling with this question of why productivity has not really grown in this recession. In normal recessions, although unemployment increases, per capita output also increases; however, that has not happened. Is that partly because employment growth, which you refer to in paragraph 1.24 of the executive summary, has tended to be based on part-time rather than full-time work? What is the current position with full-time equivalent posts rather than the total number of people who are in employment?
There is an element of that, but the key point to bear in mind is that the productivity puzzle remains a puzzle if you take both elements into account and look at output per hour as well as output per worker. The underlying puzzle is that although until relatively recently a growing number of hours was being worked in the economy, not a great deal more was being produced. Obviously, that has been adjusted for the effect of self-employment, part-time employment and short-time working. What you have suggested helps to explain part of the big picture, but it does not help to get rid of the underlying puzzle, which is not merely why the private sector created so many jobs when output was not going anywhere but why, even though more hours were being worked over the same period, there was not a great deal more output.
In your report, you say that income growth for self-employed people has not really matched expectations, which is why tax receipts from them have been lower than expected. Is one explanation that, because wages have risen less than inflation, relative labour costs have been lower than, for example, investment in further capital? We often hear about businesses being capital rich, but has that had an impact on productivity?
09:45
Some people would run a story of which the relative costs of labour and capital and substitution between the two were a part. Most people’s view is that it is hard to run that argument strongly enough for it to explain a great deal of the difference.
I would love to say that we have cracked this since we were previously here, but nobody has done that. However, one explanation to which people continue to return and which still has validity for most people is the idea that the wider difficulties of the financial system, such as credit conditions, mean that capital has not been reallocated as efficiently as would normally be the case from relatively unproductive firms that might go out of business to younger firms that could be innovative and highly productive, which it would ordinarily be hoped would grow rapidly.
In looking at that story, some people would emphasise the zombie firm side—the failure to kill off weaker firms, which is a combination of wages being relatively subdued, interest costs being relatively subdued and banks not wanting to crystallise losses, so firms are keeping going. Perhaps more important is the inability of productive and innovative firms, which could expand rapidly, to get hold of the working capital that they need to expand. It is hard to quantify precisely the contributions of all those things, so one must assume that the explanation is a combination of a lot of them.
Another issue, on which more light might be shed in the summer, is whether output is being undermeasured. In relation to the 1990s recession and recovery, subsequent statistical revisions have made the recession look shorter and shallower and the recovery look stronger than the earlier vintages of national accounts data suggested.
For example, back in the 1990s, three years after the pre-recession peak, people thought that the economy was running at about 2 percentage points below that peak. However, the national accounts data today suggest that we should now think that, at the same point, output was running at 2 percentage points above the peak. A 4 percentage point difference is a substantial revision in comparison with the size of the fall in GDP in the course of that recession.
We have no way of knowing whether that situation will be repeated and whether new data will show that the recent recession was shorter and shallower than previous months’ data shows, but that is clearly a possibility, given what happened in the 1990s. It is striking that it took some years for the statisticians to change the numbers on the 1990s recession in such a way that they noticeably rewrote history.
We might get some such information this summer. Quite a big set of national accounts revisions is coming up. The committee might come on to the fact that that might have a big impact on measured public debt and make the cash size of the economy look about 2.5 to 5 per cent bigger than the current data suggests. However, how that will affect the path of real GDP down and up during the crisis is less clear.
We always come back to the sobering fact that we are dealing with the relatively early drafts of economic history. Anybody who says that their forecasts are great might find that history confounds them, and people who are depressed about forecasts might suddenly discover that they are not as dumb as they thought that they were.
Where are we on the level of structural unemployment in the UK relative to pre-recession levels?
Our view, which we share with the Bank of England, is that the long-term sustainable level of unemployment—the level that is consistent with stable inflation—might be about 5 or 5.5 per cent, although that is highly uncertain. The unemployment rate is a little below 7 per cent at the moment. The Bank of England has what it thinks of as a medium-term sustainable level of unemployment of about 6 per cent, on the basis that if the rate falls rapidly to that, wage growth might be expected to pick up.
We look over a five-year horizon, whereas the Bank of England looks over a two to three-year horizon, so we focus more on a longer-term view of structural unemployment. It is assumed that some of the people who are in the long-term unemployment that is automatically associated with any rise in the jobless total will be re-employable in time—that takes time—so a 5.5 per cent unemployment rate can probably be sustained in the long term. That implies that some spare capacity remains in the labour market and is one reason why we still have an output gap. We have a view that the economy is running about 1.7 per cent below its full potential, because there is scope for employment to be higher than it is while still being consistent with maintaining the inflation target in the longer term.
You have prepared a briefing on Scotland specifically, particularly on the devolved taxes. I will ask you a couple of questions on that before I open up the discussion to colleagues around the table.
My first question is on the OBR’s forecast figures for the Scottish rate of income tax. The March 2012 figures predicted that in 2016-17 SRIT receipts would be £5.633 billion, but your latest prediction, from March this year, is that they will be £4.974 billion, which is a drop of £659 million, or almost 12 per cent.
In December 2013, you predicted that receipts from the SRIT would increase from £4.377 billion to £5.551 billion, which is an increase of £1,174 million, or 27 per cent, in the four years from 2014-15 to 2018-19.
Will you talk us through those revisions and where we are now?
Sure. Think about it at two levels: the UK position as a whole and then the Scottish share. In the UK position as a whole, a striking feature of the recovery to date has been the fact that earnings growth has not picked up as rapidly as we had anticipated, which is linked closely to the productivity puzzle. You would assume that, once productivity started to increase, that would make firms more confident about paying people more and we would get back to the sort of world in which earnings grow more rapidly than inflation. That has not happened as quickly as we and most other forecasters anticipated, and that explains some of the downward revisions to the overall level of income tax receipts.
The picture from year to year is also complicated by changes to the top tax rate—the introduction of the 50p rate and then the 45p rate. The fact that both those changes were pre-announced created incentives for people who were in a position to shift their income forward or backward in time to do so, to take advantage of the change in income tax rates. That has made assessing the underlying path of self-assessed income growth particularly difficult, which has clouded quite a lot of this.
Why do we expect things to pick up in the future? We assume that productivity will eventually start to pick up and that incomes will start to grow more rapidly than inflation. There is some evidence that the gap has now closed. It is early days, but we assume that that will start to happen in the course of this year and that, in time, we will get back to a more typical pattern of incomes rising by 2 per cent or so above inflation. We will not get to that by the end of the five years, but we are moving in that direction, and that will bring back something that we have not seen for quite a while: fiscal drag. That is the idea that, as incomes outpace inflation, more of the incomes of more people will be pulled into higher tax brackets, raising the average tax rate and helping to raise the overall receipts moving ahead.
A key point to note about the Scottish share specifically is that we had something for this forecast that we did not have for previous forecasts—the survey of personal incomes for 2011-12, which at the moment is all that we have to go on to work out the Scottish share up until the point at which HM Revenue and Customs will flag people as being Scottish taxpayers or not. The survey showed that the Scottish share was higher in 2011-12 than the previous forecast suggested, so we adjusted for that and pushed it through the forecast.
We are a bit more optimistic about earnings growth in this forecast than we were back in December, which pushes things up a bit, and the budget measures have had an impact, although their impact on the Scottish share is not enormous.
A key element is a further rise in the personal allowance, which tends to impact on Scotland relatively more than it does on the rest of the UK. On the other hand, some of the measures to deal with tax avoidance and increase those payments would be expected to impact on Scotland less than on the rest of the UK, so there are pros and cons in terms of the impact of budget measures on the Scottish share. However, those are relatively small compared to the underlying adjustments to the UK forecast as a whole, which are more important for tracking both where the revisions have been made and why you would expect to get more in five years’ time than you get today.
There are real issues with block grant adjustment.
The last issue that I want to deal with is land and buildings transaction tax. I want colleagues to have a chance to ask questions about that.
There has been a significant reduction in Scotland’s share of land and buildings transaction tax—stamp duty land tax, as it is now. Over the past five years, it has fallen from 6.7 per cent of the UK level to 4.1 per cent. The Cabinet Secretary for Finance, Employment and Sustainable Growth has suggested that an average of those five years should fuel the block grant adjustment. However, Danny Alexander, the Chief Secretary to the Treasury, has said that
“the value of such a one-off adjustment would erode over time and would therefore cause a windfall gain to the Scottish Government and a windfall loss to the UK taxpayers over quite a number of years.”—[Official Report, Finance Committee, 4 September 2013; c 2867.]
The latest OBR forecast is for a doubling of the take from Scottish land and buildings transaction tax from £283 million to £580 million—in fact, that is an increase of more than 100 per cent. In its stage 1 report on the Land and Buildings Transaction Tax (Scotland) Bill, the committee said that the OBR’s forecast was “wildly optimistic” and was £127 million more than the actual figure. I note that the OBR has stated that house prices in London increased by 12.3 per cent in the year to December 2013 as opposed to only 0.5 per cent in Scotland. I realise that this is a fairly long-winded question, but the bottom line is that, as I am sure other members agree, we have concerns about the OBR’s forecasting of LBTT receipts given the history of that issue. Could you talk us through that?
I understand the concern. You should have concerns about anybody’s forecast for a property transaction tax.
It is important to remember that, in predicting what you are going to get out of the tax, you are interested not simply in house prices but in the volume of transactions as well. One reason why there are much bigger swings in the amount of receipts that you get from such taxes relative to, say, income tax and VAT is that there is uncertainty not only about where the prices are going but about where the transactions are going.
Transactions are relatively compressed compared to what you might think of as the long-term, sustainable level, which, in itself, is uncertain. You have to take into account the average number of times that people move house or the number of times in a given number of years that people will move house. That implies a level of transactions, and you assume that, over time, you will get back to that figure.
Two or three years ago, we would have expected the economy to be picking up and, therefore, the activity in the housing market and the consequent receipts from that activity to be picking up more rapidly, as there is a greater volatility on the way up and on the way down. That did not happen. However, it has now started to happen with a vengeance and one of the reasons why we will end up borrowing less than we expected this year is that the housing market is now more buoyant than we anticipated.
The additional uncertainty in the Scottish share concerns the extent to which the dramatic increase in receipts that we are seeing at the moment is a London effect—or, perhaps more accurately, an expensive houses effect. There are more expensive houses in London relative to the rest of the country, so it is not clear that it is a regional issue per se as opposed to a house price distribution issue. We have assumed that London—the top end of the market—will continue to outperform for the next year or so, and that has been taken into account in the Scottish share forecast that we have produced.
I would love to say that we are moving to a period in which the stamp duty receipts will suddenly become vastly more predictable, but that would be a rash thing to promise. At the moment, a combination of rapidly rising house prices and policy measures is increasing the amount of transactions, which is producing considerably more money.
10:00It is striking, as you say, that the Scottish share is relatively volatile in that case. As you know, for most of our devolved tax forecasts we assume that historic rates of the Scottish share of total UK receipts will continue into the future. In some cases, such as income tax, that is a reflection of the fact that the share does not move very much. In this case, it is a reflection of the fact that the share moves quite a lot but in ways that are hard to explain, so we average the share over a number of periods and assume that the average share will continue. We have to take that into account.
The issue of how to address the block grant under those circumstances is, fortunately, way above my pay grade and I leave it to the cabinet secretary and the Chief Secretary to the Treasury to fight that one out between themselves.
I thought that you would want to avoid that issue, to be honest. The Chief Secretary to the Treasury wants to maximise the revenue for the UK and the cabinet secretary wants to maximise the revenue for Scotland.
The issue about LBTT is that the OBR was about 14 per cent out in its predictions. That is a significant difference, yet your most recent predictions are that revenue will be higher over the next three years than you estimated in March 2012 when there was that overestimation of about 14 per cent. That is why there are some concerns.
The price story has moved on and we have assumed that the price increases that we expect over the five-year period will be bunched together in the near term instead of happening later on. That is a reflection of recent outturn data and the pace of what is happening in the most recent figures. The forecasting errors in the total forecast for the past year were to do with the fact that the stamp duty receipts were more buoyant than anticipated, which is hardly surprising given what is going on in the housing market.
There is, of course, a new and related issue about what will happen when Scotland moves from SDLT to LBTT. If it was difficult to forecast what was going to happen when we were sticking with the same system, it will become even more difficult to forecast what will happen if there is a new system. We look forward to talking to whichever bit of the Scottish Government or revenue Scotland is thinking about what the rates and the structure will deliver. I do not know whether that will all be in place to enable us to look at it in time for the autumn statement forecast at the end of this year or whether it is more likely to be pushed into the early part of next year. You may know more about that than I do. We will talk to the experts who are putting the system together when they are in a position to talk to us about it.
Thank you. I now open up the discussion to colleagues around the table.
I will start with the Scottish tax forecasts—we are very interested in them and particularly in income tax if we go down the route of further fiscal devolution. You explained some of that to the convener. I was interested in what I read about the effect of the budget in your Scottish tax forecasts paper, although you seemed to suggest today that the effect is not as great as I perhaps expected from your paper.
In table 1.2 on page 5 of your paper, which shows the Scottish share of income tax, the budget does not seem to have any effect on that share. Does it have an effect that is balanced out by something else?
The main effect is from the new information on where we were back in 2011-12, because we now have the next round of the survey of personal incomes, which gives us the outturn data. That suggests that the Scottish share of income tax was 3.08 per cent in 2011-12, rather than the initial estimate of 3.03 per cent.
Why does the share drop over two years to 2.92 per cent?
That is partly because we expect London prices to grow relatively rapidly in comparison with the rest of the country, so we will get more receipts from London and the south-east relative to Scotland.
In the table, the Scottish share is shown as flattening out over the following five years.
We basically assume that the London effect—or the expensive house price effect—will last for a while, but we do not assume that there will be an ever-continuing widening of the gap in prices between London and Scotland.
History suggests that, over time, some regions or parts of the housing market lead others. In the longer term, a more neutral assumption needs to be made about house prices in aggregate rising more in line with earnings and about not adding in something specific to the share to adjust for a continued change in relative housing market performance. That must be looked at each time we have new data.
The table shows that the budget has no discernible effect on the Scottish share of income tax.
On income tax specifically, if you look at table 1.4—
I was asking about income tax before, so I was a bit mystified by your answer.
I am sorry. Table 1.4 shows what has driven the forecast change in income tax. The table’s title refers to March but it should refer to December.
The budget 2014 measures line, which is second from the bottom in the table, shows that, although the changes are not trivial, the sums involved are not enormous. Towards the end of the forecasting period, the dominant effect is due to the announcement of a further increase in the personal allowance, which has a direct effect on the cash implications and the Scottish share of income tax.
The top line of the change takes into account the fact that the share in 2011-12 in outturn data was higher than we expected. We basically take that change on board and push that through the rest of the forecast, so that it affects every year thereafter by roughly the same amount.
We have taken into account an adjustment on the treatment of gift aid repayments. The bottom line of the table shows the UK forecast as distinct from anything Scottish. The fact that we have had relatively strong data in the near past is pushing the amount higher.
A combination of the measures and implications is coming together. As I said, the budget measures are non-trivial, but £30 million out of £5.6 billion is not an enormous sum.
The budget measures will always be taken into account in block grant adjustments.
As I said, it is not for us to decide what the block grant adjustment is. We try to be as transparent—
I know, but is that the position in principle?
In principle, we carry out such a diagnostic with each forecast, which would allow people to look at the effects of the measures. Whether the negotiators wish to take that into account is up to them.
Do we have information on non-savings and non-dividend income, which is the relevant income tax base for the block grant adjustment, and on how that is changing?
I do not think that anything particularly strange is going on in the difference between the total income tax base and the specific stream that would push the figures dramatically. We have spelled out the position as best we can. That is an issue of what HMRC can do for us to remove the bits that are not included, so that we get rid of some of the uncertainty that is in self-assessment.
I return to the wider UK issues that you have discussed. Your basic point is about people drawing down savings and cyclical growth. You mention that business investment is picking up, but you also talk about the persistence of low levels of investment. What is your general view on the situation and whether investment is picking up?
The level of investment is low and the Office for National Statistics has revised down the level, so investment has looked weaker than earlier vintages of data suggested. However, the latest data also suggests a relatively rapid pick-up through 2013, with an increase of about 8.5 per cent from the beginning to the end of the year. That is not a year-on-year comparison.
The fact that business investment has picked up is one reason why we have nudged the overall growth forecast slightly higher relative to December. However, I was careful in the outlook document to point out that we should not count that particular bunch of chickens before they hatch. Given the scale of the revisions, we would probably want to wait and see whether we are experiencing a sustained pick-up rather than another branch in a relatively volatile path.
Such as it is, the news is relatively upbeat. As I said, the Bank of England’s forecasts say that business investment will grow in double digits for the next three years—at 12, 13 or 14 per cent—while we have forecasts of 8, 9 or 10 per cent or thereabouts. We are in the range of outside forecasts, but it is not surprising that the range is relatively wide when we are dealing with a series of figures that is extremely volatile from quarter to quarter and is revised significantly from one vintage of data to the next.
I will look at two of the factors that you mention with reference to 2018. Obviously, there will be more on debt interest, but you also talk about
“two scenarios where interest rates are”
even
“higher than assumed”.
How significant would their effect on the fiscal position be?
They have different effects. The reason for putting in those scenarios is that there has been a lot of debate about how quickly interest rates will rise and whether they might rise more rapidly than people expect. The key point from a fiscal position is that the situation depends on why interest rates are rising more rapidly.
We can think of one explanation for interest rates rising more rapidly as the good or benign one—it is that the economy is doing better, so people’s incomes and spending are rising more rapidly. That is beneficial in the sense that it reduces the headline level of borrowing and it perhaps makes it more likely that the Government will get the debt to GDP ratio falling again in 2015-16, as it hopes to do, although our view remains that it is more likely than not that it will not achieve that. However, it could do that if the economy was stronger, and one of the consequences of that would be higher interest rates.
If interest rates are rising because of risk premia and difficulties in emerging markets, for example—in other words, if a rise in interest rates is not accompanied by an increase in people’s incomes—that is less benign and is not as good for the fiscal position. If we assume that it tells us more about the structural weakness of the economy, it might make things more difficult for achieving the target of cyclically adjusted balance. It also matters for individuals and families because, if interest rates were rising because people’s incomes were rising, we would worry a lot less about people’s ability to cope with them than we would if they were rising for another reason.
I am sure that the financial policy committee of the Bank of England will have an eye to whether we need to worry more or less about that matter. We focus more on the fiscal consequences.
Will you talk more about social security? The story about social security is very much about some people being excluded from it. The presumption is that there will be less unemployment in 2018. Are demographics or other factors driving that?
The comparison with the position prior to the crisis arises partly because there has been a period in which the inflation rates to which the generosity of many welfare and social security payments is linked have remained relatively high, even as the economy and the earnings of people in work have been relatively weak. That has pushed up the generosity of welfare payments in terms of the share of national income that is received per recipient.
There are certainly all the usual effects. As there is a cyclical recovery, we see unemployment falling, which reduces welfare pressures. However, a substantial increase in the generosity of benefits when inflation was 5.2 per cent in a particular September was locked in, and that was used for the subsequent years’ increases. I use the word “generosity” in a relative sense. There is a conceptual issue about whether we judge how generous benefits are by what can be bought with them or in relation to the living standards of people who are still in work.
One reason why the situation has evolved as it has is that we have been through a relatively extended period in which earnings have been weak relative to inflation. Retail price inflation and consumer price inflation have been higher over the years since the financial crisis and even higher than what Alistair Darling expected when he produced his final forecast, which assumed that there would not be a recession. There has been a quite unusual period in which earnings and the economy have been very weak, but retail price inflation has remained relatively high, partly because of imported costs and oil costs, for example.
I am glad that Mr Chote quantified the term “generous”. I am not sure that many of my constituents at the sharp end of welfare reform would feel that it is a particularly generous system. However, that is just my observation.
You talked about there being a degree of growth in the economy, Mr Chote. However, in your executive summary, you said that “productivity ... remained disappointing.” You said in your opening statement—I think that I am quoting you exactly—that cuts to Government spending are acting “as a drag” to recovery. Can you quantify that for us? If cuts had not happened in a certain area what might we have expected to see?
10:15
There are two ways of looking at that, one of which is that there is a direct contribution from public spending to GDP that we can look at in isolation. However, that is only part of public spending. The bits of public spending that count directly as part of GDP are basically the Government’s consumption of goods and services and the capital investment that the Government makes, but not welfare payments, which are about transferring money from one person to another but not increasing the total amount of output in the economy.
We can look at how GDP moves over time and ask what contribution the changes in Government consumption of goods and services and Government investment have made. I cannot give you the numbers off the top of my head, but we produce tables showing the relative contributions of those changes compared with things such as net exports and private consumption. What is striking about that data is that, ironically, over the period in which we and most other people have been overoptimistic about the overall growth of the economy, we have overestimated the extent to which public spending cuts would be a direct drag. That is because we did not take adequate account early enough of the way in which the output of much of the public sector is measured for the purposes of the national accounts.
For example, if we squeeze spending on teaching and schools, and measure the output of schools by the number of pupils passing through the school system, perhaps or perhaps not adjusted for exam results, we can cut spending quite a lot and say that we will remove teaching assistants, which does not change the measure of output. That means that there is actually less of a drag than anticipated and it looks as though productivity is doing very well in the public sector specifically. However, if we decide to have cuts in education by keeping the level of spending per pupil the same but not educating anybody whose name comes after the letter R in the alphabet, we will have a direct link between the cut in spending and a cut in the measured output.
In education and a variety of areas of public services in which a significant portion of the output is directly measured, we overestimated the degree to which cuts would be a drag on GDP in the early years. We now anticipate that there will be such a drag, but we have been surprised by how little there has been. So that is one issue, and it can be quantified in that way.
There is then a broader discussion about what impact we think fiscal consolidation as a whole has had on GDP, which means the consequence of not only the direct contribution of what the Government consumes and invests but the impact of the decisions that it makes on welfare and taxes. That takes us into a debate about things called multipliers, which involves estimating the effect on GDP of a given change in the amount of Government borrowing. The effect will differ depending on whether the cuts come through capital spending, welfare, VAT increases or other tax increases.
On our analysis of that, for the forecast produced by my predecessor in June 2010, when the coalition came in and announced the first and only significant net near-term fiscal tightening relative to what it had inherited, a set of multipliers was used that were based on a fairly wide range of academic estimates. A set of numbers was plucked out that was within that range. If we apply those numbers mechanically, we can say that the fiscal consolidation has probably taken about 1.7 per cent off the level of GDP—that is, the level of GDP would be about 1.7 per cent higher in the absence of that fiscal consolidation. We get that figure from directly applying the numbers, and can say that the fiscal consolidation will have had a negative impact on growth over the past few years.
Looking forward, that will have less of an impact on growth, because we approach the point at which the depressing effect of the new year of fiscal consolidation is offset by the previous years of fiscal consolidation fading away. Basically, the same amount of water is going into the bath from the tap as is coming out of the plug, and the level remains constant overall. That is as best as we have managed to quantify the effect as regards the mechanical application of the multipliers that we used back in June 2010.
There is a broader debate, however, around the question of those multipliers being a good measure of the impact on the economy of doing a little bit more or less fiscal consolidation. There is much broader uncertainty around the fact that markets respond to fiscal consolidations in a non-linear way. There might be a point at which the markets suddenly get nervous if we do not do very much at all, and that can have more dramatic effects.
I would not place an enormous amount of weight on this, but we quantify the effect and we have a report in October that looks back to the performance of previous forecasts, where we set everything out. So there is a quantified estimate, and the level effect is about 1.7 per cent. As I say, however, I would treat that with a reasonable amount of salt, as one always has to do with such estimates.
Treating that with a reasonable amount of salt, you talk about 1.7 per cent of GDP, but what would that be in cash terms, roughly?
Well, erm—
Perhaps I should have asked you to bring a calculator.
It is roughly £1.6 trillion multiplied by 0.017, so—
Perhaps I should have brought a calculator myself.
I do not have a calculator with me, but I am sure that somebody can knock that out for you.
If we make a comparison with comparable economies, how is the UK doing now, as opposed to before the financial crisis of 2008?
Do you mean in terms of the economy or in terms of public finances?
I mean the recovery in terms of GDP.
In comparison with past recessions in the UK as well as internationally, the peak-to-trough fall in the recent recession was big but not without historical precedent over the previous three or four recessions. It is the weakness of the subsequent recovery that has been striking—the absence of a V-shaped pattern of recovery on the lost ground, which the convener talked about earlier. Both internationally and in comparison with history, the extended weakness of the recovery is the striking thing.
However, bearing in mind what happened to the numbers in the 1990s, we need to be careful in suggesting that that will still be the pattern when we doubtless meet here again in 10 years’ time, as history may have been rewritten. For the time being, however, it is the weakness after the trough that is striking, at least as much as, if not more than, the peak-to-trough decline.
I admire your confidence that we will all be here in 10 years’ time. I was given to understand that the UK’s recovery was the weakest of all the G7 countries. Is that correct?
I do not have the comparison with me, but that would certainly fit. As I say, I would not necessarily suggest that the peak-to-trough fall is the dramatic thing, but it is a less flattering comparison to note where we are X years after the peak.
Alongside the budget, the Treasury published a report called “Impact on households: distributional analysis to accompany Budget 2014”. I refer to the Treasury’s own figures. The report says that the accumulation of cuts to public services and tax and benefit changes implemented between 2010-11 and 2015-16 would be equivalent to £757 for the average household, or 2.1 per cent of income. For households in the bottom income quintile, the cuts are equivalent to £814 or 3.4 per cent of income. The concern is that that is the Treasury’s own analysis of its budget. Do those figures sound roughly correct?
I am afraid that the Westminster Parliament has explicitly told us that that is outside our remit. With my previous hat on, I would have given you an informed answer to that, and I am sure that Paul Johnson would give you one if you asked him.
We will perhaps ask him, then.
You have spoken about wage growth, and the convener mentioned the IFS view that real income levels will not return to 2001 levels until 2017-18. You said that that sounded about right. When you said that wage growth remained disappointing, were you referring to that or to something else? Can you quantify what you meant?
I meant that growth is disappointing relative to the expectations in previous forecasts. We have found—and have tried to adjust for this in subsequent forecasts—that, relative to what we had anticipated, employment tends to perform better than we expected while wage growth does not pick up as much. That is linked to the productivity puzzle.
The fiscal implications of that are interesting. It has offsetting effects, because the total amount of labour income is boosted by employment but weakened by relatively weak earnings growth. Generally speaking, you get more bang for your buck through the growth in tax receipts from people’s earnings than from a growth in the number of people in employment. Many of those people in employment will be on relatively low incomes and will not produce a great deal more tax whereas, if everybody’s earnings are growing more rapidly, that helps to shove some relatively well-off people further up the tax bracket, as we discussed earlier.
That has been a fairly consistent pattern, but it will be interesting to see, when we come to the next forecast, whether it remains the case. The data that we have had so far this year certainly continue to show employment and unemployment similarly performing strongly relative to most people’s expectations.
That would help not only tax receipts but economic recovery, as those people would have more money to spend productively in the economy.
Absolutely. We have underlined that, to get to a position in which productivity growth is such that people’s real living standards are rising and incomes are outstripping inflation by a healthy margin over a continuous period of time, it is important to have a long-term sustainable pick-up in consumption, which is the largest chunk of spending in the economy.
That is why, as I said, we have been slightly cautious about the sudden pick-up in growth that occurred during 2013, and the extent to which private consumer spending has contributed to that. It seems to be more the result of a reduction in saving than of the long-awaited pick-up in real incomes. The saving ratio fell by approximately 2 per cent of real household disposable income during 2013, but it is not realistic to expect that it will continue to fall by two percentage points. In the near term, we will need to see income growth picking up the reins.
The reason why we are forecasting a slight slowing in growth during this year is that we assume that we will move back to a position in which consumption is driven more by underlying income growth than by saving behaviour. Having said that, the difference is not enormous, and we will wait to see the quarterly growth figures. Forecasting quarterly growth paths in the near term is an art at least as much as a science, and we will wait to see what emerges.
That 2 per cent cut in income to which you refer is presumably the same one that I just mentioned. I know that you cannot comment on that in detail, but it is interesting. My question was about wage growth, but I am not clear that it is covered by the 2 per cent figure, which refers to cuts to public services, benefit reforms and tax changes. The situation of those people could be further exacerbated by the failure of wage growth to kick off.
As I said, one way to take a step back from the issue is to return to the fact that GDP per head will take a lot longer than GDP in total to recover and reach pre-crisis levels, because of what has happened to the population over that period. It is therefore not entirely surprising that real incomes are still some way away from their pre-crisis peak.
In your opening statement, you spoke briefly about there being no growth in net exports for the period of the forecast. Given that we see a growth in net exports in Scotland—although I appreciate that you do not have differential figures—what leads you to draw that conclusion?
10:30
It is a combination of the views that we take on board on the growth of the world economy and world trade. We are a relatively small body, so we do not do a complete bottom-up global forecast on export demand. For example, we look at the numbers that are produced by the International Monetary Fund and the Organisation for Economic Co-operation and Development in their world forecasts and world trade forecasts. We would adjust those figures if we felt that there was a powerful reason to do so, but we are not taking an unusual view of the progress of the world economy relative to what the international institutions are saying.
Another thing to note is that the UK’s export market share has been in a fairly steady decline, which we assume will continue at a somewhat slower pace than in the recent past. That is partly an automatic reflection of the fact that, if we had a world economy that consisted of the UK and two other countries in the emerging markets, and those other countries traded more with each other, the UK could be standing still but our share of overall trade would fall because countries B and C would be trading more with each other and increasing the overall magnitude. We have to take some account of that.
For the eurozone, the picture is clearly better than people were anticipating some while back, but it is not motoring away in a way that would lead us to take a dramatically more positive view. The fall in the exchange rate did not, as some people might have anticipated, result in a dramatic improvement in export performance.
For all those reasons, our view is that we will not get much more of a net contribution. That is basically down to the view that we take of the world economy, which leads us on to world trade and UK export performance relative to market share. On the import side, there is a pick-up in consumer spending and in investment, which is bringing imports in too, so those two things—roughly speaking—cancel each other out and there is a balance.
We do not have disaggregated figures for Scotland, England, Wales and Northern Ireland.
No—and certainly not with a sufficiently short lag for them to be terribly useful.
I will ask a daft question. Are you privy to policy changes, such as the sale of the Post Office and the change in policy on pensions, in advance? Do you offer budget projections on those policies, or do you recalibrate figures after the event?
We produce a forecast that is published as the chancellor sits down after giving his budget statement, and I do a press conference on it two or three hours later. We have to have a forecast that incorporates the impact of all the measures that will be announced in any budget or autumn statement in real time.
The way in which that works is that, about six to eight weeks prior to a budget or autumn statement, we start producing forecasts on the basis of what would happen if the chancellor did nothing and just sat on his hands. We produce what is called a draft pre-measures forecast a few weeks before each statement. We then produce successive drafts as we take more of the judgments that we need to take, and as more economic data comes in and we get closer to publication date. We go through about four of those drafting rounds as we move along.
Simultaneously, the Treasury will come to us, again about six weeks prior to the budget, and give us a draft scorecard, which is a list of measures that the chancellor is thinking of announcing in the budget. We look at that list and work out which of those measures we will need to devote a lot of analytical effort to and which will be relatively simple to analyse in terms of estimating the impact on the economy and the public finances.
As the six to eight week process goes on and we refine the pre-measures forecast, we can make inquiries. For example, if it is a tax measure, we talk to HMRC and consider what we think is a sensible costing for each measure and what impact it will have during the forecast period.
We know when we see the first list, six weeks ahead, that half of the items on that list will never make it to the finishing line. That is partly because the Treasury will think, “We want to do something on pensions and we’ve got three options.” It puts down the three options but we know that it will never want to do all three. Basically, it is a series of options.
As we get closer to the budget, some things that were not on the list at the beginning will appear on the list. One of the important things that we have to do with the Treasury is ensure that we are able to give adequate attention to each of those measures in order to be able to say that we have had enough time and information to reach a judgment on whether those measures will cost what the Treasury says that they will cost.
As the committee will appreciate, we have a debate about that. We will make suggestions such as, “We think that the avoidance of this tax measure will be greater than you are assuming,” and the Treasury can decide whether it wants to take that advice on board. As it turns out, there has never been an occasion on which the Treasury has gone to the Parliament with an estimate of what a particular measure will cost or raise when we have felt constrained to say that we do not believe the estimate. That tells us something about the Treasury’s responsiveness during the course of that discussion and scrutiny process.
At the end of the day, however, with every number that the Treasury puts into its list of policy measures and how much they will raise, we have to say, “Yes, we agree,” “No, we don’t agree,” or “You didn’t give us enough time or information to reach an informed judgment.” So far, on the vast majority of measures, the Treasury has taken on board our suggestions that a measure may cost or raise more or less and has come out with a number that, at the end of the day, we have been happy to endorse.
The short answer is yes, we see policy changes in advance. We have to see them in advance in order to produce a forecast that works on the day. We give advice to the extent that we say, “This is what we think this will cost or raise.” We do not give advice such as, “This is an extremely stupid policy,” “This is an excellent policy,” or “Why don’t you do something else instead?” It is more limited to, “If you go ahead with this, this is what we think it will cost or raise.”
Were there any unforeseen consequences to the post office and pensions policy changes?
There are usually unforeseen consequences with every policy measure. One of the difficulties is that the number of measures for which we have the time, information and opportunity to do a rigorous analysis, after the event, of whether it raised or cost what we thought it would is relatively limited. Usually, the occasions on which we do that are when there is a new policy measure in a subsequent budget or autumn statement that is in a similar area to something that has happened previously. In order to have an informed view of what the new policy will raise or cost, we need to go back and scrutinise what the previous one raised or cost.
One striking issue about this budget—and a number of measures, such as the pension measure of bringing forward the amount of money that people have to pay up front in disputes over tax avoidance and so on—is that there are about four measures that, taken in aggregate, are quite positive for the public finances during the five years of our forecast horizon but are much less positive if we push the horizon further into the future. We felt that for those measures, it was particularly important to present what the forecast would look like over a 20 to 30-year time horizon, because we felt that focusing just on what they were expected to raise or cost in the near-term five years gave an unduly flattering impression of their underlying impact on the public finances. We wanted to show that the £1.2 billion a year that those measures would raise in the first five years would dribble away fairly quickly beyond five years. That gives a fairer impression of the impact.
We have to be particularly wary of measures that shift money from later to earlier or from earlier to later, as distinct from measures that basically raise the same amount of money or cost the same amount of money each year into the future, because that can paint an unflattering or an unduly flattering picture.
You touched on the question of savings, which you also mentioned when you were speaking to Mr Gibson and Mr Hepburn. If I understand you correctly, you are saying that people are saving less at the moment, not that they are eating into their savings.
There is an aggregate number for the savings ratio, which incorporates some people taking on more debt and some people saving less. We do not look at it at a micro level, but in aggregate the share of people’s incomes that they are saving has fallen over that period.
The changes in pensions would perhaps encourage people to spend earlier and then not spend later. I think that you said in response to Jean Urquhart that there would be a movement in spending, rather than an overall increase.
Yes, there is a series of measures, such as the changes to annuities, the treatment of tax-free individual savings accounts, and the new granny bond, and that set of changes is likely to influence saving behaviour. We felt that it was sensible to look at those in aggregate and we decided that we would not adjust the economic forecast for the effect of those things in aggregate, not because we are certain that they will not have an impact, but because some of them go in one direction and some go in another and one cannot even be clear about what the aggregate sign is likely to be. We looked at them in the round and said that we did not feel that it was sensible to make an additional adjustment to the forecast.
The annuity changes will obviously encourage people to spend money earlier, but they may also encourage others to save more in the first place because they will be less concerned about their money being tied up in the future. Even on one measure, things could go in both directions. You would also expect the annuity measure to have different effects on the incentives of different sorts of people. It is likely to be relatively attractive to bring forward money for people who have expensive debts that they want to pay down and save money on, and it might also be relatively attractive to people who have large amounts of assets. It is less likely to be attractive for the people in the middle of those two groups. We are assuming that roughly 30 per cent of people will take their money earlier as a consequence of the annuity decision, taken in aggregate.
It is best to look at that sort of measure in the round and to try to reach a judgment, and because things are moving in both directions we have not made a further adjustment to our savings ratio forecast, compared with the pre-measures forecast that I mentioned to Ms Urquhart.
Some people have suggested that, although people were hit in 2010 and realised that they had not been saving enough, their behaviour might change in future. Would you make a judgment on that or would you just wait and see what happened?
We have not made a judgment on that. We have identified it in the policy costings document as one of the key areas of uncertainty around the package of fiscal measures in the budget. There are quite large measures affecting the incentives that people have to save, and to save in particular ways, and the unforeseen consequences might be considerable. Ex ante, however, it does not look as though we can say with confidence that there will be movement of X amount in one direction or another.
It is striking that there is an awful lot of savings floating around, and people can move money in quite large volumes quite quickly—for example, as we noted in the report, billions of pounds are moved relatively rapidly between deposit accounts and sight access accounts. It remains to be seen what impact that set of measures will have. A lot can depend on whether all the Sunday newspaper personal finance sections say of a particular measure, “This is great, fill your boots,” or whether they do not, and that can have quite an influence on the eventual impact.
Thinking about changing behaviour on house prices and the volume of house sales, you mentioned long-term trends. Presumably, the assumption is that people’s behaviour in relation to houses—how often they buy or what prices they pay—is not changing, even though we thought that people were paying too much for houses.
10:45
We do not reach a view on the right level of housing transactions or have a model that is heavily deterministic in that respect. As far as transactions are concerned, we assume that, in the long term, we will return to people moving house with roughly the frequency that they had for a longer period. If we are way below that level, quite rapid growth in housing transactions will get us back to normal levels. Needless to say, however, there is considerable uncertainty about what the eventual normal level will be and whether history is as good a guide to that as we hope it will be. You just need to take into account the information that you have available.
It is good that you mentioned the information that you have available, because in your report you refer to the changes that the ONS is making. I realise that you are not the ONS, but I have to say that I do not have a particularly good handle on what those changes are. If I am reading paragraph 1.39 on page 16 of the executive summary correctly, you seem to be saying that changes in one direction will be set against changes in the other and that, overall, there will not be that much impact. How will those changes affect us if we are trying to compare different years?
Some changes will directly affect the size of the public sector’s debt; there will be some effects on the deficit, but their magnitude is less important. Some things that were previously not counted as public sector debt will be, and we expect that to add about £140 billion to the measured size of public sector debt in 2014-15.
It is a wee bit scary for debt suddenly to appear out of nowhere.
It is, but it is important to bear it in mind that nothing in reality has changed; all that has changed is the way things are measured. For example, the treatment of Network Rail as part of the private or public sector will have an impact in that respect.
Will you have to go back and change the last 20 years to compensate for that change?
Yes, and we will be urging the ONS to give us the longest possible run of consistent data to allow us to draw this inference.
With regard to year-to-year changes, the addition to net debt in 2018-19 will be slightly smaller than £140 billion. Although the level of net debt will be higher, it will be higher by smaller amounts in future than in near terms. As a result, even though the debt will be higher, the probability of its falling in any given year will be slightly higher than it was.
The second complication is that people usually look at Government debt not just in cash terms but as a share of national income. At the same time as it is increasing the measure of public sector debt by, for example, changing the treatment of Network Rail, the ONS is making other changes that will increase the overall cash size of the economy. For example, more of the investment made by companies will be counted as an output instead of some intermediate function that gets washed out in the public finances. The higher level of debt and the increase in the cash size of the economy will both offset debt as a share of GDP: the debt numbers will push it up and the increase in the cash size of the economy will pull it down, although not by enough to get us back to where we started. As a result, the net debt ratio as a share of GDP is likely to look higher when we next meet than it does today, looking back as well as forward.
That is quite interesting. Perhaps we should get the ONS along to the committee.
If you are that keen, I note that there is an annex to the report that I suspect the clerk has very helpfully not inflicted on you. I would describe it as of niche interest but, for that small and relatively noble band who are interested in these sorts of things, it goes into some detail on the exact contributors to the change in net debt.
That is helpful. Thank you.
You talk about forecasting land and buildings transaction tax once you know what the rates and bands will be. Why would you forecast LBTT? That would be on behalf of the Scottish Government, but you are not directly working for it.
That is true, but we will still need to take a view on it because we will have to come up with an aggregate forecast for the public sector finances for the UK as a whole. We will want to talk to whichever body produces the forecasts for the Scottish Government—you may know whether it will be revenue Scotland or the possible new independent body—about whatever numbers it comes up with. If we feel that it is necessary, we may adjust those figures for the numbers that we put into our data. However, we will still need to cover that tax in order to produce a comprehensive view of the public finances. We will talk to whoever turns the handle in Scotland on it in the same way that we talk to HMRC about every other tax that we look at.
We can continue to talk to HMRC about SRIT, but we will want to discuss the fully devolved taxes with whoever the experts are—the people who are involved in designing them and working out what is going to be raised through whatever rates and rules are put in place. We will either use those numbers or adjust them as we see fit.
In the short term, you will forecast both what SDLT would have been if we had kept a Scottish share and what we expect LBTT to be—the two might be different.
Yes. To produce a forecast for the UK public finances, we need the forecast of the receipts from whatever tax is being paid at the time. If we know that SDLT is going to fall out for 2015 and that, thereafter, there will be a new and different structure, to produce a forecast for the public finances that will go forward to 2019-20 we will need to forecast on the basis of the tax that will be in place.
The other purpose of your forecast will be to inform the block grant adjustment, which is based not on LBTT but on SDLT. You need that one as well.
Exactly.
You have stated:
“in the final week before the Budget, HMRC officials provided us with a final set of Scottish forecasts ... Due to the confidentiality of the measures we were unable to involve the Scottish Government in this stage of the process.”
Can you explain why that was the case?
The pre-measures forecast is based on whatever the policy is prior to the budget. We can and do have the Scottish Government representatives at our discussions about how we think that these things are likely to evolve and how we should interpret the past information. However, we cannot share with the Scottish Government what the chancellor is planning to announce to the UK Parliament in the budget before he has announced it.
I wonder how that will work, assuming that we remain part of the UK. Let us say that there was some dramatic change in landfill tax at the UK level in the UK budget. Would that not need to be discussed with the Scottish Government? Would that be your responsibility, or would someone else do that?
That is a different issue. It is unlikely that the chancellor would discuss measures that he was thinking of announcing in a budget prior to announcing them to Parliament. There can be consultation on such changes, but we produce a forecast on the basis of unchanged policy and then try to take into account what the chancellor is going to announce on budget day, which he will keep to himself.
The UK system is obviously different. We consult on our budget before we get there. You would then have to forecast what the Scottish Government might do in response to what the UK Government had done, and it strikes me that that would be almost impossible for you if you had not spoken to the Scottish Government.
On that basis, we would have to work on the assumption of unchanged policy and that would probably have to be swept up in a subsequent forecast. If a change in UK policy stimulated a change in Scottish policy after the event in a way that had not been predetermined, we would have to catch up with that after the event.
We are required by Parliament to produce a forecast on the basis of current policy. It would be inappropriate for us to guess how the Scottish Government would change policy in response to a change in UK Government policy.
It is an interesting area. Certainly on income tax, if we had control over the 10 per cent rate, I think that the Scottish Government might well react to what happened elsewhere.
Do you mean if the UK changed other parameters, for example by cutting or raising rates?
That is right.
I suspect that we would not be very popular if we prejudged how the Scottish Government was likely to set its income tax rate.
Perhaps it should be a requirement for the two Governments to talk to each other, rather than for the OBR to be involved.
Yes. Again, there is a question about timescale. If things were done in a way that was consistent with having an agreed view at the time of the budget, we would be able to take that on board, but I presume that that is not where we will end up.
The other issue is that, with regard to when decisions are made on the Scottish rate, we have to produce our forecasts according to the timetable for the UK budget and the autumn statement. I guess that there are questions about when it will be clear what the Scottish rate will be and in which forecast we could incorporate that as existing policy. We do not know quite how the relative timetables will work out.
Yes, because—you made this point earlier—if you give more notice of changes, that can allow certain taxpayers to move their affairs around.
Yes, but that is a slightly different issue. If a tax change is announced in advance, depending on the nature of the tax, that creates the potential for people to anticipate what is being done.
Which is why all Governments tend not to want to announce—
Indeed. I suspect that there is rather more scope for that with self-assessed income tax than there is with landfill tax. There are some areas of tax in which pre-announcing a change is not desirable. If changes in capital gains tax would create a powerful incentive for people to realise gains at a particular point in time, it would be necessary to be wary of that.
As I said, forecasting income tax has been made particularly complicated over the past few years, not merely by the fact that the rates have moved up and down but by the fact that the movements have been announced some way in advance and that they have affected people who—unlike wage slaves such as us—can shift their income from one place to another relatively easily.
Thanks very much.
My first question is a brief one on business investment. Since you published your economic and fiscal outlook, has anything happened to make you more optimistic or less optimistic about your projections for business investment?
We will not come back to that and do another forecast until we get to the autumn statement.
Given the amount of data revision that we have coming up over the next few months, it would not be worth while worrying too much about that in the near term when so much could change before we get to the next forecast. There is nothing that I could point to that would suggest a significant change at this stage, but I would not want to suggest that we have been looking closely at the issue to reach that conclusion.
Thank you.
I will move on to an issue that has not been touched on. It relates to paragraph 4.52 on page 111 of your economic and fiscal outlook, which has the headline “UK oil and gas revenues”. I will be sensitive in how I ask my question. Will you outline to the committee how you reached your conclusion that the figure for 2013-14 will be £4.7 billion, that the figure for 2014-15 will be £3.7 billion and so on up to 2018-19, when you predict that it will be £3.5 billion? What is the process that the OBR goes through in reaching such conclusions?
One of the delights of forecasting oil and gas revenues is the number of determinants that go into it and the fact that they move around quite a lot. That means that the receipts in this area are highly volatile. Changes of 40 to 70 per cent from one year to the next are not unusual. Forecasts often end up being revised significantly. The reason for that is that a number of variables need to be taken into account, one of which is obviously production. The second one is price and, because the price is set in dollars, we have to worry about the sterling-dollar exchange rate. We also have to worry about the level of expenditure in the industry, because quite a lot of that can be set off against tax, so doing more capex reduces the amount of receipts that come in. There are then any changes in the tax system that are taking place at the same time. To come up with a forecast, it is necessary to look at all those things.
11:00On production, we look at the data and projections that are available from the Department of Energy and Climate Change and Oil & Gas UK. Broadly, the picture is that we have repeatedly been overoptimistic in assuming that the fall in oil production would level off, as it has not done so—it has continued to decline. That said, we continue to assume that, as oil production has fallen for 13 years on the trot, there will be a period in which production stabilises. That assumption is a reflection of the amount of investment that is going on in the industry. We are assuming that that will be sufficient to arrest the decline in production. As I say, we have tended to assume that that would happen before—DECC’s production forecast has assumed year after year that the pace of decline in production would slow down or stop, but it never has and has continued. We hope that, at some point, the trend will cease to be the friend and we will see the flattening off that we are expecting.
We then have to take a view on prices. We used to use the futures curve right the way out over five years. However, we became concerned about that, because trading is very thin beyond the first couple of years and it is not clear whether it gives sensible information, so we are using futures and basically assuming that prices will be flat thereafter. It is possible to make a variety of other more or less complicated assumptions, but that is a good trade-off between simplicity and taking on board the information that is in the market to begin with. As you will be aware, organisations use numerous different methods—some have higher oil price forecasts over the next five years than we do and others have lower ones. We make an assumption about movement in the exchange rate based on relative expectations, relative interest rates and so on.
We look at the information and intelligence that we can get from DECC and from the industry via DECC about what is going on with capital and other expenditure. In looking at the weakness of forecast receipts, there is a twin thing to consider. We hope—perhaps against recent experience—that the relatively large amount of investment that is going on in the industry will arrest the decline in production but, at the same time, if there is a lot of investment, the firms are able to set that off against tax, so you do not get the receipts early on. You will get more receipts later as a consequence of higher production, but that will be offset in the near term.
In coming up with all these numbers, we basically look at that set of determinants. As you see in table 4.11 in our economic and fiscal outlook, in each forecast we try to split out how the aggregate forecast for receipts has changed as a result of each of those things. For example, you will see that oil and gas production has not had a great deal of impact between December and March, which is hardly surprising because it is only three months and there is not a great deal more information. If we had done the comparison with a year ago, we would have been taking on board the fact that production had disappointed again.
There are then market movements, prices, expenditure and the fact that less receipts are coming in in the near term as a result of expenditure, because if there is more capital investment, it is possible for the firms to set off more of that against tax and that affects the receipts. That is basically the exercise that we have to go through, but it is one of those areas of receipts where there is a relatively large number of determinants that are, in themselves, quite volatile and hard to predict, which is why the outturn and the forecasts move around quite a lot. I am afraid that that will be true for whoever tries to do your forecast for these sorts of things, not only for us.
One of the criticisms that have been made of the OBR is that you have not taken into account the capital investment and expenditure made by oil companies. To what extent have you taken that into account? Is the criticism accurate?
It is not accurate. Anyone who came down from Mars and looked at the path of oil production over the past few years would see that it is a pretty straight downward line, and I suspect that when that puzzled Martian looked at our expectation that oil production is going to go flat over the next number of years he would ask, “Why on earth are you not expecting the continuous decline that you have seen over the past decade or so to continue?” The reason for that is precisely the increase in investment that you have described.
You have talked about a number of elements, but I think that two stand out: production levels, and the revenues that would flow from those levels. Is the OBR judgment based on the view of whoever you have spoken to that production is going to stabilise rather than increase?
Yes.
And your figures for the revenues that we collect are shown in table 4.11.
Yes.
I do not have any more questions on that.
Finally, you called SDLT a fairly volatile and unpredictable tax. Is there any difference in the volatility and unpredictability in the commercial and residential sectors? Your projections seem to change less for commercial properties than they do for residential properties; is there any obvious explanation for that?
It is probably true that volatility in the commercial sector has been somewhat less. As for the explanation for that, I have to say that I do not know the extent to which all of this is linked to the pricing and availability of mortgages and whether over time they move in a way that generates greater uncertainty. I suspect that the planning that businesses carry out might result in a less volatile outcome, but I am afraid that that is as close as I can get to an explanation. There might be other structural reasons for that.
Thank you.
On the oil and gas issues that Gavin Brown has just explored, I was intrigued by Mr Chote’s reference to relatively high investment. I understand that investment in 2013 was the highest that it has ever been. Is that your understanding, too?
Yes, I think so.
So instead of “relatively high” we could call it a record high.
In nominal terms most things are at record highs most of the time.
Do you accept that your position on the oil price is fairly pessimistic compared with other assumptions?
I would not have said so.
You assume that in 2016-17 the oil price will be $99 a barrel. Is that correct?
The last time I looked at the comparison—which was probably the last time I was here, when I gave a talk on this—we had a range of figures. I seem to remember that the Scottish Government produced a set of oil price assumptions, and ours was at one extreme. Some think tank here produced another set, and we were at the bottom end. When you added the two together, we were somewhere in the middle.
I am not talking about some think tank. The Department of Energy and Climate Change has assumed a price of $120 a barrel; by contrast, the Scottish Government has been more cautious than that, although it is somewhat more optimistic than you have been. Your estimates are lower than many others.
There is a variety of other estimates, including those of the International Monetary Fund, the International Energy Agency, the Organisation for Economic Co-operation and Development and the Bank of Norway. Take your pick.
You assume that production will be flat but, in its last UK forecast, the industry itself forecast an increase from 1.4 million barrels to 1.7 million barrels a day by 2017. Why is the industry wrong and you are right?
For the same reason that DECC has been wrong over time. If you look at past comparisons, you will see that things do not always turn out as the industry or, indeed, DECC has suggested.
Or as the OBR has suggested.
As I have said, we have taken the view that, given the scale of investment, we would expect the steady decline that we have had in the past to level out. We have assumed as much in the past and, like the industry and DECC, we have been historically disappointed that production has continued to decline rather than stabilise as people have been expecting for some time now.
The industry has invested a record amount—£14.4 billion—and you estimate that that will result in flat production, while you assume that the industry, which invested the money, is wrong in projecting an increase in production.
When the industry wrote to me a year or two ago, it said that there was a record amount of investment and, subsequently, it was disappointed by the level of production.
To go from the sublime to the ridiculous, we will move from talking about billions of pounds in income to talking about the aggregates levy, which involves marginal sums. Your Scottish tax forecasts document suggests that income from the aggregates levy will increase by £4 million over the next four years, which is a marginal figure. It is interesting that you say that that is because of a change in aggregates extraction levels. What is the change that you base your assumptions on? You say that there will be a decline. What are the factors in that decline?
Again, we are relying on data that we can get from the relevant departments. I presume that the information is more an extrapolation of recent trends that is informed by some view of the future rather than an assumption of a dramatic change in the path. I cannot give much more detail than that.
I understand that Scotland’s proportion of aggregates extraction is higher. You predict that the UK figure will increase by £31 million over the next four years and that the Scottish figure will increase by £4 million, which is almost exactly what would be expected on a per capita basis.
Yes—the share is remaining relatively constant.
It appears that, if Scotland produces proportionately more aggregates, we are to produce less than the UK average over that period. Is that interpretation of table 1.10 right?
I assume that the figures are based on an assumption that the proportions will be relatively constant. If the Scottish level moved out of the path on which the rest of the UK is moving, the figures would shift. I can check, but I think that the assumption is that the share will remain relatively constant.
You might not be able to answer my next question, either. For as long as I have been looking at economic figures, the value of using GDP to measure a country’s strength and compare it with other countries has been discussed. The issue has become topical again recently, although it is always there in the debate among academics, economists, politicians, geeks, anoraks and whoever. Do you defend the use of GDP as the measure by which you forecast the economy’s strength?
Your comment is right. The issue is horses for courses—what is GDP being predicted for? We are trying to predict the path of the public finances; we look at where spending will go and where tax and other receipts will go on the basis of existing policy. When we consider that most taxes are levied on some proportion of people’s incomes and spending and of transactions in the economy, real GDP—which people read about in newspaper articles and focus on as showing how well the economy is doing in this quarter relative to the previous quarter—is less important to us than the size of the cash economy and how that breaks down into different categories of spending and receipts.
I agree with you that probably more attention is given to our forecasts for real GDP, because most people are interested in that as some sort of index of how well the economy is doing, even though it would not be at the top of the list of the three or four variables that we would consider if a fairy godmother came down to say that we could predict a variable with perfect certainty to help us to predict what would happen in the public finances.
11:15There is a related set of questions around whether we claim that GDP is a good measure of material wellbeing. In that case, we run into the fact that GDP, which is an aggregate, does not take into account how income is distributed, whether we regard producing weaponry as adding to wellbeing, any environmental consequences and so on. However, those are not arguments for not looking at GDP; they are arguments for looking at something else if we are asking a question to which looking at GDP cannot provide an answer. For example, the ONS has a big programme of work on wellbeing for which it produces alternative and additional indicators. We can have a different debate about whether that work is value for money—whether it tells us anything new or interesting.
We can certainly point to the flaws in using GDP for a variety of purposes, but they are not necessarily the purpose for which it was produced in the first place. I commend to the committee an excellent book by my former colleague at The Independent, Diane Coyle. You will be glad to hear that it is a slim volume of tribute entitled “GDP: A Brief but Affectionate History”, which looks at both why we should still care about GDP and why we should not care about it too much.
Thank you. So when Professor John Kay raises doubts about GDP and politicians use GDP as a benchmark, we should tread carefully.
I think so. As I said, it depends very much on what question it is claimed GDP gives the answer to. If it is not the right question, you need to be wary. Another reason to be wary of real GDP is, as I said, because of revisions. The path of the fall and recovery in GDP in the 1990s now looks completely different from what we anticipate for GDP or even what, in retrospect, GDP appears to have been in the mid-1990s. There should be caveats on whether the official measure of GDP is even measuring what it is supposed to be measuring accurately, let alone on whether it is measuring the right thing in order to answer whatever other questions you might want answered. That is not an attack on the ONS. Measuring or summarising activity in the economy of tens of millions of people is not a straightforward thing to do anyway, and the idea that there is some magic number that summarises everything that anybody might want to know about the economy is, I am afraid, not on the table.
GDP should have a Government warning attached when a Government uses it as its benchmark.
The Government has been supportive of the idea of the ONS doing more work on wellbeing. However, whatever the number of people who look at the work that is done on wellbeing measures, I think that whether that tells them a great deal that they could not have guessed otherwise or derived from other measures is in the eye of the beholder.
Okay. Thank you very much.
It is a shame that it is a couple of months past Valentine’s day, because we could all have got that GDP book for our other half.
It is the gift that keeps on giving.
I think that because GDP is internationally recognised, there would be a lot of suspicion from Opposition parties if any Government tried to move away from using it. That is one of the difficulties with not using GDP that we have discussed at committee previously.
We have concluded questions from the committee, but I want to touch on an issue that has not come up yet. We had quite a lot of questions on some of the sexier issues—the big macroeconomic ones, for example. However, like Michael McMahon with the aggregates tax, I want to ask about the landfill tax. Of course, the landfill tax is important because of its impact on the block grant adjustment, which is why on a number of occasions the committee has discussed some of the OBR forecasts.
In March 2012, you predicted that in 2016-17 there would be £157 million in receipts in Scotland from landfill tax, but your most recent prediction states that there will be receipts of £100 million for the same year, which is a difference of more than 36 per cent. The Scottish Government estimates that landfill tax receipts will continue to fall and will be around £40 million by 2025, but the OBR’s forward projections appear to suggest that there will be stability in the receipts, because of stronger than expected UK landfill tax receipts. Are you going to discuss that issue in greater detail with the Scottish Government? If so, has the Scottish Government come back to you with what it thinks the annual decline in the receipts will be over the long term?
It has not come to us with any alternative numbers. Obviously, in the discussion of the pre-measures forecast, Scottish Government officials were present virtually, if not in the room.
If you look at pages 10 and 11 of the clerk’s note, you will see that the main change in the forecast since the last time was that better data from the Department for Environment, Food and Rural Affairs on the path of landfill that is charged at the standard rate was taken into account. I think that the Scottish Government has had concerns about whether policy was driving that down more rapidly than was forecast anyway. The fact that DEFRA’s numbers now suggest that more of that is going on at a UK-wide level may address some of that anxiety.
I recall that that background note for this meeting referred to the Scottish Government saying that there was an expectation of £107 million in landfill tax receipts in 2015-16, but a decline further on—if my memory serves me correctly. Its number did not seem to be very different for 2015-16, but obviously it was producing a forecast that goes way beyond our five-year horizon.
Yes. It expects landfill tax receipts to go down to £40.5 million by 2025. The issue is that Scotland’s share of landfill tax income appears to be holding steady in your predictions at 8.9 per cent. We have a concern that that might be an exaggeration, because of the policy of trying to go for greener policies to reduce landfill.
That is clearly an uncertainty, but the Scottish Government has not come back to us with an alternative set of numbers that it thinks that we ought to look at instead. Chart 1.2 of the Scottish tax forecasts paper obviously shows a decline in landfill charged at the standard rate assumed for the UK as a whole, but I suspect that it will be rather difficult to reach a prior view on the relative expectations of both Governments that landfill will go down, who is expecting more and how well founded those judgments are. Obviously, if there is any information that people want to bring to us to look at over that horizon, we would be pleased to look at it.
Okay. That is something that we can pursue.
I thank you very much for your contribution. You have been in the hot seat for nearly two hours. Is there anything that the committee has not touched on during its deliberations that you want to talk to us about?
No, I do not think so. Obviously, I enjoyed our earlier discussion about the possibility of an independent fiscal institution for Scotland. From our point of view, looking over the next few months until we meet again, the issue of where the forecasting expertise will reside for the non-SRIT devolved taxes is obviously of particular interest to us. I want to ensure that we are in a position in which, however the Scottish Government decides to arrange those things, we can come up with the best possible working relationship in which our relative independence is respected if there is an independent body for Scotland and in which we are also able to exchange information in the most straightforward process. If different views are taken—people will always take different views on such things—I would like to be in a position in which we at least know why our views are different, even if we decide to stick to those views. I will be very interested to see how the mechanics of that unfold. I understood from the note that most of that remains uncertain for the time being, but I presume that it will clarify itself before we meet again.
Yes. We are waiting for the Government to respond to the committee’s report, which was submitted on 7 February.
Once again, I thank you very much. Your contribution was very enlightening and is greatly appreciated. I hope that you enjoy the rest of your day, particularly your first trip to St Andrews.
Thank you very much.
As that is the end of the public part of the committee’s deliberations, I close the public part of the meeting to allow the official reporters, Mr Chote and guests to leave.
11:24 Meeting continued in private until 12:24.