Finance Committee, 16 Nov 1999

Meeting date: Tuesday, November 16, 1999

Official Report 229KB pdf

---

Expenditure Plans (2000-01)

On behalf of members, I welcome again to the committee Professor Brian Ashcroft and Stephen Boyle, who made important contributions to our briefing session in August. It is nice to see you again and I thank you for making time to see us. Both of you are going to make presentations, which I hope will enable the committee to ask more informed questions of the Minister for Finance and his civil servants. I hand over to Professor Ashcroft, who is going to start.

Thank you. We greatly appreciate the opportunity to give evidence to the committee. I say that on Stephen Boyle's behalf as well as my own. Members will appreciate that we have not had much time to put the presentation together, as our thoughts are still evolving. We hope to write an article to appear in our "Quarterly Economic Commentary" in December, which will give a fuller analysis of the planned changes. Today, we will try to give a flavour of them.

Stephen and I have decided to organise our evidence as follows. I will give my view of the level 1 data. Stephen will then examine the details of the level 2 data. A handout including graphs and my speaking notes will be circulated to members before the end of the meeting.

Figure 1 shows the budget baseline changes for the period 1999-2000 to 2001-02. The graph compares the situation before and after the announcement of expenditure changes in October. The figures are adjusted for inflation, so they reflect real spending. As far as I can detect, the figures after the announcement are somewhat higher than the figures before due to what the Scottish Executive calls "flexibility and transfers".

The graph refers to planned expenditure. Therefore, as one would expect, planned expenditure before and after the announcement track each other from 1994-95 to 1998-99, but thereafter diverge. Specifically, the higher figures after the announcement appear to be due to the carrying forward of underspent moneys into the planning period 1999-2000 to 2001-02. The new total for 1999-2000 is £309 million greater at constant prices than under previous plans.

In his speech on 6 October, the Minister for Finance mentioned the transfer of working capital surpluses from national health service trusts to current expenditure. I am not certain about all the aspects of that and I was unable to get clarification from the Scottish Executive finance department yesterday, but I will pursue that point.

For 2000-01, the figures are almost identical to those under the previous plan. There is therefore little change overall, but some change this year.

I move now to figure 2.

Would you prefer to complete your presentation and take questions afterwards, or take questions as we go?

I do not mind. What is the convention of the committee?

There is no convention, so we will proceed in whichever way is most comfortable for you.

In that case, I will complete the presentation and take questions afterwards.

Figure 2 examines the change in real level 1 planned expenditure for the period 1998-99 to 2001-02, comparing the plans before the minister's announcement, which were largely as outlined in the final departmental Scottish Office report, "Serving Scotland's Needs", published in March this year, with the new plans announced by the minister on 6 October. A positive number, therefore, indicates that, under the new plans, spending on programmes will increase by more or will decrease by less during the period than was originally outlined in "Serving Scotland's Needs".

Against that background, it can be seen that the total budget is almost static, as one would expect. There is a small increase of £13 million at constant prices. At programme level, the major increases in planned level 1 expenditure are in spending on children and central Government education, which is up £28 million in real terms; transport and the environment, up £23 million in real terms; enterprise and lifelong learning, up £16 million in real terms and outlays on the Scottish Parliament, which are up £15 million.

The major decreases are in expenditure on communities programmes, down £35 million in real terms—I will come back to that—and in the unallocated capital modernisation fund, down £26 million in real terms. However, all is not as it seems. Although the graph shows a fall in the communities budget in real terms of £35 million compared with previous plans, the decrease may only appear to be such.

In fact, as the minister outlined in his speech on 6 October, the Executive has decided to forgo making early repayments of Scottish Homes debt. That will lower current outlays, but will not lower current expenditure on housing under the programme. The Executive's decision to reduce debt repayments will allow money to be reallocated to spending on housing. In a sense, therefore, the decline in the communities budget hides an increase in expenditure over the planning period.

More money will have to be found in future, however, to repay the debt, or future spending on housing will be lower. On 6 October, the minister noted also that the freeing up of surplus working capital by NHS trusts would reduce the published figures for spending on housing, but not the investment in housing. I am not clear why that should be the case. I am trying to understand.

The overall picture, therefore, is that there is not a great deal of change, but that money is being reallocated under the new plans to children and central Government education, transport and the environment and enterprise and lifelong learning.

Figure 3 does not compare planned expenditure before and after the announcement, but outlines the absolute change in real planned expenditure. The figure shows that the total budget is projected to rise by £380 million in real terms in the period 1999 to 2001 to 2001-02. Within that budget, the programme that will enjoy the greatest increase in spending is health, where expenditure is up £209 million in real terms. Expenditure on rural affairs will increase by £56 million, on children and central Government education by £54 million, on communities by £39 million in real terms and on enterprise and lifelong learning by £33 million.

The programmes that experience a real fall in their budgets in the period 1999 to 2001 to 2001-02 are justice, with a fall of £51 million, Scottish Executive administration, where there is a drop of £20 million, EU structural funds, down £12 million, and transport and the environment, down £11 million.

Another way of looking at the changes is to express them as a percentage of the base figure spent per programme. A lot of money is already being spent on the health service. We can see from figure 4 the percentage change in spending by programme over the period 1999-2000 and 2001-02, at 1998-99 prices.

The figures show a couple of anomalies. Forestry enjoys the biggest increase—67 per cent over the two-year period. However, that, in part, reflects its transfer to the Scottish Parliament. In addition, Scottish Executive associated departments have an increase of 64 per cent. Looking at the mainstream programmes, we can see that the children and central Government education programme experiences a real increase over the period of 17 per cent, rural affairs increases by 11 per cent and capital expenditure for local authorities increases by 8 per cent. Only two programmes have a significant decrease—spending on the Crown Office decreases by 13 per cent and Scottish Executive administration falls by 12 per cent.

I will close by examining the importance of the planned changes to the overall budget. We know that the programmes vary significantly in the scale of expenditure, so that health, which is one of the major programmes, has a significant expenditure.

Figure 5 identifies the share of the total budget change accounted for by the planned changes in each programme. It can be seen clearly from figure 5 that the health budget enjoys by far and away the greatest share of the increased budget. The budget increases by £390 million between 1999 and 2001-02 and the health service enjoys a large share—almost half—of that increase. Fifteen per cent of the increase goes to rural affairs, 14 per cent goes to the children and central Government education programme, 10 per cent goes to communities and 9 per cent of the overall budget change goes to enterprise and lifelong learning. The fall in real expenditure on justice is about 13 per cent and the Scottish Executive administration decrease represents about 5 per cent of the total budget change.

The graphs are an analysis of the bald figures of level 1 expenditures, comparing before and after plans and comparing the new plans—the change from this year to 2001-02. Stephen Boyle will speak about the level 2 expenditure.

Stephen, just before you start, it would be helpful if we could ask Brian some questions about his presentation.

In your article in the Fraser of Allander "Quarterly Economic Commentary" in September, you made a point about how the Government was not in keeping with its priorities in terms of the expenditures on industry, enterprise and training. You said:

"It is difficult to resist the conclusion that the industry, enterprise and training budget has had to bear a large part of the burden of the set-up costs of the Scottish Parliament."

How would you revise your comments?

There is some truth in my statement. It partly refers to level 2 expenditure, which Stephen is going to talk about. While I do not want to pre-empt what he wants to say, the level 2 figures show that, in money terms, there is a drop in spending on Scottish Enterprise in 1999, from £400 million to £376 million. In cash terms, spending picks up again to £432 million. That spending is still there, for the present year.

However, you went on to make the point that it was the type of spending in that area that was not in keeping with the Government's priorities.

It seemed to me that everyone else is tending to move away from cash subsidisation to investment—to softer forms of assistance to promote competitiveness and to stimulate innovation, where Scotland has a real problem.

At that time, the Scottish Office was spending about £10 million a year on promoting innovation. That has been cut to about £5 million. On the level 2 figures it is hard to identify what the spending on innovation is because it is not disaggregated enough.

Last week, the Chancellor of the Exchequer made an announcement about entrepreneurship. How does that affect us?

I am not clear about how it affects us. It will affect us only in terms of any increased spending that might feed through.

How might that happen?

I am not sure. Could you remind me of the specifics of the chancellor's proposal?

He made various announcements that glamorously described ways of encouraging entrepreneurship.

The problem is that our spending is determined by the allocation to Scotland, which is driven by the Barnett formula. That applies to changes in funding to comparable programmes in England and Wales. Parliament decides on the allocation of the block that is given to Scotland and can spend as much on enterprise as it likes.

Figure 1 illustrates the before and after situations and clearly shows an expenditure bulge in 1999-2000. The Administration has given commitments to additional real expenditure in certain policy areas, particularly in relation to education. If I understand this graph correctly, you are saying that such investment would be possible in 1999-2000, because of the expenditure bulge, but might not be possible in subsequent years, unless money is taken from other budget headings.

The position will be unchanged, more or less, by the end of this period. As far as I understand it, the bulge in 1999-2000 reflects the carry-forward of underspend in previous years. In one sense, it could be argued that that is not a real increase and that an examination of the outturns would show a different situation: there would be a dip in one year and an increase in the next.

The total budget, compared with previous plans, hardly changes: it is about £13 million. The increased expenditure that is available for some of those programmes comes from elsewhere in the budget. However, the reallocation is only short term. For example, funds have been reallocated from non-productive expenditure, such as paying debts, to productive expenditure on housing, but those debts will have to paid at some point.

So we are not comparing apples with apples, but plans with outcome.

We are comparing plans, which is a problem.

It appears to me that the minister is using outturn to inform his 1999-2000 figures.

That is true, and we do not have the outturn figures before us.

So we are not considering plans only, but plans and an outturn.

The figures that we were given were planned figures. I have no figures for last year's outturn, but it seems that there was an underspend that can be carried forward.

I want to follow up what John Swinney was saying about identifying the underspend in previous years. In previous years, was the underspend claimed by Westminster?

I am not certain, but I think the Treasury rules were that the underspend could be claimed back. It was not carried forward within the Scottish Office budget—there was a new playing field next year. Given the constitutional arrangements now, the Scottish Parliament can transfer between heads and transfer between time.

I am sure the Minister for Finance said that this was an accumulation of underspend over the previous two or three years, certainly on the health service side.

In his speech on 6 October, the minister said:

"The new arrangements for end-year flexibility, introduced by the Chancellor of the Exchequer, have enabled us to carry forward significantly more money from last year to this and subsequent years."—[Official Report, 6 October 1999; Vol 2, c 1030.]

That is clearly a new arrangement. It appears that it is a consequence of the Scottish Parliament. I am not sure that those changes apply in England.

Wherever this money was taken from, was it a previous accumulation? At the end of each outturn, if there was an underspend was it taken back to Westminster?

My feeling is that it would probably have been taken back and that the accumulation is from the first year of the comprehensive spending review.

So it is a one-off?

That is my feeling, but I do not know. It would be helpful if we had outturn figures as well as planned figures, for the reason that one can get a better understanding of the disposition of expenditures against plans.

As there are no more questions, Stephen Boyle can now make his presentation.

As Brian mentioned, I will focus on the more detailed level 2 data. I will highlight the key changes those data indicate in the Executive's spending plans. I will compare the end year of 2001-02 with the last full outturn year, which was 1998-99.

I have had to take a slightly different approach from the one taken by Brian Ashcroft. I do not have an equivalent set of before figures from "Serving Scotland's Needs", so it is difficult to do a before and after comparison. I have also dealt with a different period from the one Brian covered, because I will compare the final year of this spending programme with the last year before the Parliament came into existence.

I will address four points. First, I will explain my understanding of the differences between the level 1 and level 2 data. Secondly, I will outline the approach that I have taken to analysing the data. Thirdly, I will consider where the principal changes have occurred—I will say more about what I mean by principal changes later. Finally, I will comment on lessons I think we have learned in the past couple of days about the way this information and the manner in which it is provided may affect the budget process in future.

What are the differences between level 1 and level 2 data? An obvious difference is that the level 1 data, which Brian commented on, have 19 separate headings whereas the level 2 data have 70 separate headings. On the face of it, considerably more detail is available in the level 2 data than is the case with level 1 data. However, there remains a considerable degree of aggregation at level 2. Health spending and local government spending together account for 60 per cent of the total. Even within the health total, one line accounts for about £4.5 billion. Therefore, while we have more information than is available at level 1, much more could still be revealed.

As we have additional information at level 2, it permits a more detailed analysis of the changes between the major budget headings. It also allows us to undertake some analysis, which I think members were moving towards in their questions to Brian Ashcroft, of the reallocation of spending within the major headings. For example, it allows us to comment on how moneys may be reallocated within the education heading, between student support and further and higher education.

Level 2 data do not allow us to undertake analyses within those headings. For example, if we wanted to understand what was happening within further education, it would not be possible to tell from these data how moneys have been allocated within the FE heading. If we are interested in how Scottish Enterprise or Highlands and Islands Enterprise spend their moneys across their programmes, the level 2 data do not allow us to make that assessment. Those are the principal differences between the level 1 and level 2 data.

Unlike the level 1 data, which Brian Ashcroft dealt with, the level 2 data that have been provided to us by the Executive are in cash terms only. Whereas Brian could present comparisons in real terms after making adjustments for expected inflation, the level 2 data are not set out in real terms. That makes it very difficult to make any sensible judgments about future patterns of spending under the various headings.

To make that comparison possible and more realistic, I have applied the assumed inflation rates in the level 1 data to the level 2 data. I will present level 2 data in real terms, at 1998-99 prices.

I have made one other adjustment to make a comparison. There has been a reallocation of funds from current local government spending to the communities budget. Footnote 4 on the table specifies that. For the sake of comparison, I have added the number which is now in the communities budget back on to the local government total.

I want to focus on what I have called the big numbers. As Brian demonstrated when talking about the level 1 data, there are cases of very large percentage increases or decreases in a particular aggregate. The seven-figure sums involved are relatively small, so I have mainly disregarded them. I will focus on the biggest changes in percentage terms, but also in absolute terms.

Let us examine what I have called winners in absolute terms. I remind members that I am comparing the end-point year with the last year before the Parliament came into existence. The largest single real-terms increase over that period is revenue spending for hospital and community health services, which runs to over £300 million. Over the period as a whole, the total real-terms increase in the budget is about £880 million. That accounts for a very large proportion.

The next largest component is general pharmaceutical services—over £150 million. Taken together, those two sub-components of health expenditure account for a considerable proportion of the increase in the budget over the period.

I will not go into detail on the other components of change, which are noted in the graph, other than to say that additional expenditure—substantial amounts—is evidently planned for further and higher education and for pre-school education.

To return to a point Brian made about the treatment of Scottish Homes debt, there is a reduction, as I will explain in a moment, in Scottish Homes expenditure, but there is a broadly compensating increase in the resources available for the new housing partnerships.

Those are the principal increases in absolute terms. If we turn to the principal increases in percentage terms, a slightly different picture emerges, partly because hospital and community health revenue spending runs to more than £3 billion. Even the fairly large cash increase means that it does not feature as one of the largest percentage increases. The largest percentage increases, as you will see, are in social inclusion and new housing partnerships. There are also large increases for a number of education elements.

The largest fall in percentage terms is in European regional development structural funds moneys—more than 30 per cent in the period to 2001-02. There are also significant falls in the budgets for environmental protection, Scottish Homes—as I mentioned—and for student support.

I will make a few points about some of the implications for the budget process of the availability of the level 2 data. There is no question in my mind that the availability of these data represents significant progress, compared with the level 1 data that were previously made available to us. They allow us to undertake an analysis of whether moneys are being allocated or reallocated in line with priorities.

However, if it were possible, a strong case could be made for the availability of level 3 data. First to enable you and others to understand where trade-offs could be made, for example within the further education or enterprise budgets. It would also enable you to understand—to return to a question that was raised earlier—whether expenditure is being tailored towards the Executive's priorities; for example whether, within the enterprise budget, resources are being targeted in accordance with priorities and are being reallocated where appropriate.

Finally, this is not intended as a plea in mitigation or as bleating, and I realise that this is year one, but it would help us and the committee if in future years there was an opportunity for a slightly more leisurely perusal and analysis of the numbers.

Thank you Stephen.

Both witnesses have said that there has been a fair bit of debt postponement. How sustainable is that? As a result of it, what outcomes might the committee be required to review in trying to handle that situation in future years?

I have to confess I have not had the opportunity to study that issue in detail and I could not make sensible comments on it at this stage.

All I will say is that in his speech on 6 October, the minister said that previously the Scottish Office had planned to repay the Scottish Homes debt earlier than the original agreements had laid out, but that the Executive is reverting to the original repayment agreements. The sum that was set aside for early repayment, and which therefore was released for other spending, was £59 million. That was clearly simply a reallocation of funds across time: you will need to find that money soon.

In the scheme of things, that amount of money is not great—the budget is £16 billion and we are talking about fairly small amounts—but it is something of a chimera in that it gives you extra spend now but it must be paid for later.

In my final year as a councillor, we went through an exercise that had councillors whooping with joy at the ability to postpone debt even further. There were massive movements to push everything into future repayment, which made councils appear to balance their books. However, eventually, money will have to be found in the budget.

That is right. I do not know what the terms of the contract are, but presumably it will not be a fixed-rate contract. If interest rates rise, therefore, the real debt paid will probably be greater. Because of that, and because the debt will run over a longer period, there will probably be less to spend in the end. Also, the net interest payments would presumably be less if the debt was repaid more quickly.

Professor Ashcroft told us earlier how he thought we should interpret the underspend of previous years. What happened to the underspend in previous years? What is happening to it now? We seem to be benefiting from having it all to ourselves, whereas in the past it presumably went back to the Treasury. Is that your interpretation of where the extra money is coming from, Stephen?

My interpretation is that the ability to carry over underspend is not a function of the existence of the Parliament, but of the comprehensive spending review arrangements. Whether or not the constitutional change that brought about the Parliament had taken place, the Scottish Office would have had the ability to run a three-year spending programme rather than annual programmes.

All I can do is draw your attention to what the minister said in the Scottish Parliament on 6 October. He said:

"The new arrangements for end-year flexibility, introduced by the Chancellor of the Exchequer, have enabled us to carry forward significantly more money from last year to this and subsequent years. The arrangement is new; it is linked closely with the move to three-year settlements within the comprehensive spending review. In the past, unspent provision had, in many cases, to be returned to the Treasury, but now, the extent to which it can be carried forward from year to year has been extended and we are making full use of that facility."—[Official Report, 6 October 1999; Vol 2, c 1030.]

I do not know the details of what could and could not be carried forward in the past and what can be now. We need to see the outturn figures as well as the planned figures to make a reasoned judgment about what is happening.

In the past, have you been able to see underspend in the outturn figures and work out where the money went?

I have not done that analysis. We could probably do it, but it would take a bit of time. The outturn figures are published. We should get them now, rather than having to wait.

I want to follow up what Stephen Boyle said about the helpfulness of the level 2 data. Obviously the figures are helpful, but they are slightly limited. Had not we had Stephen Boyle here to deflate them for us, it would have been difficult to make head or tail of the movements. Why do we not get the deflated figures? Should not that be standard practice?

As Stephen said, some of the budget lines are huge. We need level 3 disaggregation if we are to make any substantive comments about the allocations. The committee must take that forward.

I was surprised to see that we got information in cash terms, as that is almost meaningless for the period we are talking about. Stephen's comments have therefore been very helpful.

Stephen will correct me if I am wrong, but the level 2 figures give us a better assimilation to the bottom line of figures that were already published in "Serving Scotland's Needs"—they are added up to a bottom line.

Yes, and they are in one place.

Yes, that is it exactly. We now have a spreadsheet that tallies up all the figures in "Serving Scotland's Needs". We never had that before and it is to be welcomed.

I wish to stress Adam Ingram's point: we must request the level 3 information as that will allow us to consider meaningfully the way in which patterns of expenditure flow.

Stephen spoke about the substantial percentage increase in social inclusion spend. To what extent will that increase have come about because of, for example, reductions in the European regional development fund? Is the Administration making reallocations simply to compensate for ERDF reductions, or are there other sources of expenditure—previous headings—that may have been gathered together to be compensated under the social inclusion heading? That seems to be a large shift of resources and I cannot quite understand where it has come from.

The answer is that I do not know. It is not possible to tell from the level 2 data whether the increase in social inclusion spend is intended to substitute for the reduction in ERDF. One cannot draw that conclusion. All that can be observed is that one has risen and the other has fallen.

My interpretation of the figure 1 chart that Brian Ashcroft showed us is that, because of the changes made in the comprehensive spending review, the opportunity has arisen to create a one-off bulge, the impact of which will be smoothed over a three-year period. For example, we are spending 98 per cent of our capacity every year, but we will see that smoothed over in outturn figures over a three-year period.

Yes, because the CSR sets the agenda for that spending—in total.

This financial statement has been presented to us as a combination of plans and outturn for one financial year, which skews the process.

The amount of new money is less than as presented because some of the new money was simply a reallocation—as far as I understand—from an underspend in the previous year to a planned spend in the subsequent year.

That can be done only once, because of the way in which we are going about planning expenditure.

That is right.

I wish to make two detailed points, the first of which goes back to what I said about last week. The detailed programme is rolled out after the chancellor's announcement, like Helen Liddell's announcement about the business schools' arrangement with France and so on. Is not it difficult to disaggregate that kind of expenditure in order to see what comes to Scotland? It is virtually impossible.

Yes, it is.

Does that come under the heading of annually managed expenditure, or is it just a reserved budget?

To be honest, I am not sure. I would have to look into it.

If any of that money is to come to Scotland, it must come under AME. Searching questions could be asked about what is coming.

It is important to disaggregate that kind of expenditure, in order to see what is happening and to identify what share we are getting. Cambridge got a huge amount—£70 million—for its tie-up with Massachusetts Institute of Technology last week. What is Scotland getting?

I will move quickly on to Stephen's point about general pharmaceutical services—is that the generic drugs budget?

As I understand it, yes. However, it is difficult because, while I told members what is on the spreadsheet, there is no disaggregation.

It is not just a question of getting more detailed figures. I was with Tayside Health Board last Friday and read Public Finance magazine last night and learned that generic drugs have multiplied in price six times since last December. I do not know whether you know about that. Inflation in generic drug prices is huge and it is causing huge problems for health boards. Presumably the Executive knows about that and is making allowances for it, so it is not a real increase in expenditure.

That comes back to the point I made at the outset about the figures that we received being set out in cash terms. The simplifying assumption that I made was that inflation would be identical across all budget headings for the next three years. That is a hugely simplistic assumption. In some components of the budget, such as pharmaceuticals, inflation may be much higher and in others it will be less.

One of the things that the committee should track—I will not pretend that it is easy—is real trends in the level 2 headings and below.

This is the sort of matter that we can find out about only through parliamentary questions. It is important; all the health boards are under huge pressure on their drugs budgets at the moment. Tayside Health Board has a £1 million overspend already. That has had huge repercussions for its budget. Perhaps the Executive is more aware of this than it has indicated in the past.

We have mentioned the need for more disaggregation, down to level 3 figures. Stephen's point is valid: because we got the data only in cash terms, he has had to apply a general deflator. When the Scottish Executive presents figures in real terms, it is legitimate for this committee to ask the Scottish Executive what deflators it is using to bring the figures into real terms.

Is the Executive applying a specific head deflator, such as general pharmaceutical services, where we know that there is significant inflation, or is it applying a general deflator? If the general one is applied to, for example, the health budget, the figure may appear to be rising for that head but it could be falling. It is important that the methodology that the Executive is using on deflation is made available to the committee so that members can have a much better understanding of the real, underlying changes. At the moment, members cannot understand that.

My understanding, from examining the level 1 data, is that a uniform deflator has been assumed across all budget heads.

My feeling, from "Serving Scotland's Needs", was that there was some difference in the deflators, although I was not clear where they came from.

I will return to John Swinney's point about the bulge—the one-off windfall that we will allegedly have in the coming year. Surely if we are to establish whether that is what will happen, we must carry out an analysis backwards on the outturn figures for previous years. It is naive to expect that because the Scottish Parliament is here we will suddenly spend 100 per cent of the planned budget in the next one to two years and that there will be no underspends in any budgets.

The world does not work that way. I do not think that the argument that this is a one-off benefit will stand up to scrutiny. At the end of planned expenditure budgets each year there will be underspends in some areas and overspends in others.

There will not be a 100 per cent spend every year when outturn figures come through. It would be useful for this committee to have an analysis of previous years' outturn figures to see how much was underspent and whether it was taken back to the Treasury. That would give some indication as to what underspend there may be in years to come.

I am arguing that the graph illustrates that the minister has announced planned expenditure for the next year and has added on the surplus that has been carried over from the previous financial year. This is the first time he has been able to do that, so has given us a revised, inflated, planned figure. He will never be able to do that again because it will meander along as in previous years.

I am trying to get to the nub of this: has an inflated planned figure been announced—with a flourish—that will be valid for one year only? Will we then revert to a much more smoothed-out level of expenditure? The graph seems to suggest that that is what will happen.

Yes, it shows a one-off gain.

No, I do not think that that is exactly true. If you present the data in terms of planned expenditures, you can always—if you underspent last year—bring that forward into next year's plans. You then get the rather bizarre situation that your ex post planning data do not change, but your forward planning data can change as a consequence of the underspend in the previous years. That is why you need outturn data to make any sense of it.

That is because you are not comparing apples with apples.

Exactly—but that is the whole point. The outturn figure for 1998-99 has not been published, as far as I know, so we cannot compare. If you always present things in planning terms, the previous years are the previous plans, and they stick; but the future years can change according to the degree to which you carry forward. You are not comparing like with like. That is why you must have the outturn figures to make sense of it.

It would be useful, convener, if we could have an analysis of the underspends in previous years.

I do not think that that will necessarily be as easy as you might think. Ten or 15 years ago, the issue with budgeting in the United Kingdom was not underspend but overspend. UK Governments systematically overspent on their budgets for a whole host of reasons. There has been greater discipline in public sector budgeting throughout most of the 1990s, so there have been underspends.

At a UK level, we have typically had three bites at planned expenditure. This year we would have plans for this year, next year and the year after; next year, two of those years would be revised. So which is the relevant comparator? Do you compare the first version of the plan with the outturn, or do you compare the third version of the plan with the outturn? The exercise is well worth doing, but it is not necessarily straightforward.

I want to develop the theme of comparing like with like. Stephen, do you have any difficulties reconciling the level 2 headings that are provided in your paper with what is in "Serving Scotland's Needs"? It appears that some pre-school education money may have gone into schools. Has there been some redefining of headings?

Furthermore, why is there nothing beside the heading of social work? Where do those figures appear?

I have had a very quick look at "Serving Scotland's Needs" and I have tried to line the level 2 data up with it. In places, there appear to be inconsistencies, or at least discrepancies, that need to be explained. I do not know the reasons for them.

We will obviously have to follow that up, convener.

Looking through this, there is a big decline in the level 1 figure for the justice budget, but then, when we get down to the level 2 figures, it does not seem to appear. Do we have any idea what that reduction is? The Scottish Executive administration costs are down as well; I would have thought that they would be increasing. Perhaps that is why it is taking eight weeks to reply to letters.

As I said, my presentation covered a slightly different time from Brian's. I went back a year further than he did. I have also converted my figures to real terms. Over the period 1998-99 through to 2001-02, the justice level 1 line increases by £29 million in real terms. The major changes within that are a reduction in legal aid of £11 million, an increase of £14 million in miscellaneous, which includes such things as civil defence, and an increase of £17 million in prisons.

Does that discrepancy arise because the analysis covers a different period?

There is a big jump between 1998-99 and 1999-2000, which accounts for the discrepancy between our analyses.

Out of interest, do we have any idea what Scottish Executive administration is? I cannot understand why the amount for that has gone down so significantly, certainly at level 1.

There is only one line for Scottish Executive administration. I assume that this year and next year there are set-up costs, which will begin to fall out.

Is the reason for the increase in the social inclusion budget the huge transfer from the local government budget to communities?

There is a transfer from local government current spending next year and the year after of £57.6 million in cash terms.

So was that money in the local government budget before? It was not ring-fenced.

There is no indication, as local authority current expenditure is one line of £5.5 billion. All we know is that a sum has been deducted from that and reassigned to communities.

When the Minister for Finance appeared before the committee, I asked him a question about NHS trusts that neither he nor his officials could answer. He agreed to come back. If the minister is working with those problems—and, as we have heard, we are working with even more problems—how can we ensure that we have on-going support for some sort of management accounting audit process that will enable us to be kept on track? Can we bring in outside advice to keep us looking in the right directions and under the right stones?

We are in a learning process anyway, but the Minister for Finance and his department cannot give us the information that we need because it seems to come from other departments. There seems to be a lack of flow of information through Government, let alone down to this committee.

We can take advice or evidence from any source that we think appropriate. If we find a good source, I think that we should do that.

I agree with David Davidson. I think that what he suggests is very important in allowing us to function as a committee and to ask the right questions. We are going round in circles here and are poking about in the dark.

It is terribly important that we have special advisers, although obviously the committee has to decide on matters itself. Mike Watson will know that special advisers are sometimes appointed for just one inquiry but, on an on-going matter such as spending, we need permanent advisers.

On call.

They should be on call from one year to the next. If we chop and change advisers, we will get more confused than we are already.

We will have to consider that.

I would appreciate the views of Stephen Boyle and Brian Ashcroft on the difficulty involved in tracking the transfer of burdens and responsibilities between different parts of Government. We have touched on that in relation to local government and social inclusion. I suspect that local authorities are increasingly assuming burdens for aspects of service provision, but it is difficult to track from where the resources, if any, have come to pay for that. What is your impression of the availability of the information that would enable the committee to strip down that £5.5 billion line in the local authorities budget to establish how that budget was accommodating the additional burdens local authorities are having to bear as a result of legislation that we or others may have passed?

I am not sure what the information would be; I would have to look into it. It is interesting that the local authority budget is increasing in cash terms, as well as in real terms. Something must be driving that, but we are not being told what. Is it being driven from the bottom up? Presumably, it must be coming from the local authorities. However, it would be useful for the committee to have that information.

The budget is going up in cash terms from a low base. In 1997-98 it reached its nadir of £5.384 billion and it has crept up since. The 1998-99 figure will be a decline in real terms from 1997-98 but, in that period, real burdens on local authorities have increased. There is a squeeze.

There is not necessarily a correlation between expenditure and the services that a local authority has to bear. We need that information and we need to find a way of pulling all the local authority accounts together. That is the problem.

Let us take the example of housing. If we were to decide to cut this spreadsheet differently—to cut it in terms of the total amount that is spent on housing in Scotland—it would be impossible on the information that is currently available to get that figure swiftly.

The figure for the amount spent on social housing?

The amount spent in local authority support for housing, local authority grants to housing associations and so on. It would be not be at all easy to amass a figure for the total amount spent on housing in Scotland by the various agencies.

That is because we would need the local authority line to be broken down, and we do not have it broken down. That is the case with education as well.

There are two approaches that we could take to this kind of issue. Some of the tables in "Serving Scotland's Needs" were helpful in explaining where there had been reallocations of responsibilities, but trying to track them is exceptionally difficult. One would have to reach judgments and make compromises about what the data were saying.

I have not looked at those figures in much detail recently, but the Chartered Institute of Public Finance and Accountancy also produces some fairly detailed data on local authority spending, both capital and current. Those might help the committee to get disaggregations of the local authority lines quickly.

I hear what you are saying, Stephen, and respect the Chartered Institute of Public Finance and Accountancy, but is this not something that the Scottish Executive should provide?

If the committee wanted to make progress on this quickly, the information is there. I do not know how quickly the Executive would be able to respond.

Could we try to avoid talking over one another, as it is difficult for the official report to follow?

In Fife alone, local government has had to absorb £47 million in salary increases over the past six years. We can get the broken-down figures for that sort of thing. However—and I would be grateful to our witnesses if they could provide feedback on this—the committee must also discuss how we can ensure that the different subject committees examine the budgets for their areas of responsibility. They need to work on disaggregating their budgets as far as possible and to feed that information back to us. John Swinney was right to highlight the additional costs that local government has had to absorb without assistance from central Government.

If we are going to investigate how effective local government has been and the extra pressures to which it has been subjected, we must examine the best-value programme as well. We need to get back some of the figures and reports on that, so that we can compare performances across the different areas of local government. We cannot say that the pressures are unsustainable unless we know how effective the delivery of service at local level has been. This committee should take some ownership of that process, because we exist to ensure that every pound of the Scottish budget is spent effectively. Local government spending accounts for a third of our budget and we need to spend some time scrutinising it.

Another area that we could do with some information on is local authority funding streams from sources other than central Government, as a number of local authorities have been involved in property speculation—perhaps I should say property development—with arm's-length companies. Other local authorities are using the private finance initiative in various areas. These things do not apply evenly across Scotland, but information on them can make quite a difference to the committee's view of what goes out from the centre. We could do with some input when we are dealing with such subjects and with health matters.

There is only so much that we can do. We are getting down to health board level. I have spoken to two health boards in the past 10 days and they have told me that they are down to the bone in terms of efficiency savings. They are now having to turn their attention from non-clinical matters to clinical matters, which worries me because it means rationing. However, we cannot start taking evidence from individual health boards—the Health and Community Care Committee would have to do that.

Surely some of the areas that David Davidson and Keith Raffan have mentioned—especially in relation to local finance—should be examined by the subject committees that are responsible for them, rather than by this committee.

We are all having difficulties with the figures, for various reasons. Perhaps it would be useful to get experts from the Scottish Parliament, or from outside, to recommend ways forward and how information should or could be presented in a way that makes it easier for the committee to understand. That could mean ensuring that we see figures for actual spend against planned spend, or the rate of inflation against each of the programme headings and so on.

Stephen and Brian, if there is any specific information that you think we should be asking the Executive or the Minister for Finance for, it would be helpful if you would tell us. We want to know which questions to ask to elicit the information that will enable us to make proper judgments.

As Brian Ashcroft said, we plan to write a paper on the subjects that we have discussed. Among the points that we will address will be the significant gaps in the information. I presume that you will want the report before it sees the light of day.

We have mentioned several areas already. We feel that the level 3 data, where they are available, would help in the analysis of trade-offs. Clearly, you need to have the data in real terms as well as in cash terms. To judge those data appropriately, you need information on the deflators that are being used, from which you can assess the implicit judgments about inflation in each of the heads.

That has emerged already.

It is crucial if any sense is to be made of the data. You would also need up-to-date information on outturn expenditure as well as on planning expenditure. That information is crucial if you are to make a judgment.

I am not deriding what we are doing here today, but we cannot have a serious discussion about the figures that are in front of us because, as Brian Ashcroft outlined, we need standard formats. That would enable us to get to grips with the figures, to start looking at where the trends are going and to do some real work.

Over the next few weeks, our challenge is to get to grips with this subject by working with these gentlemen, or others, to ensure that the committee comes to an agreement with the Executive on what we need to enable us to have a confident discussion on some of the issues that arise out of the Scottish Executive's budget.

In fairness, the Minister for Finance has made it clear in previous meetings that he is prepared to produce figures that can be more widely understood. We have to make the most of that.

Brian Ashcroft's summary on the type of information that would be useful is absolutely what we need. We asked for and have received level 2 information, which saves us a lot of headaches in trying to put columns together in "Serving Scotland's Needs". Some of the detail in that document does not really show us level 3 information or how small the large numbers can become. We should follow Brian's suggestions about the kind of information that we should request from the Minister for Finance.

We have covered quite a few specific points. Stephen Boyle and Brian Ashcroft will put together a report that the committee will see. We will try to feed that report's points into the way in which the committee operates. Furthermore, we will have to ensure that the Minister for Finance receives the suggestions about the form in which information should be presented.

Unless there are any other comments, I suggest that we draw things to a conclusion. I thank Stephen and Brian for their time.

Will Stephen and Brian also be involved with the committee when their report becomes available? Perhaps we need to have an on-going input from them.

We will have to discuss that separately. It is understood that we will need some general reference point—if that is not too crude a way of putting it—to keep us properly informed. That might not involve just one or two people; it might well be horses for courses.

George Lyon has raised an important point. There is a need for continuity. Although we might want horses for courses for specific inquiries, it is urgent that the committee considers appointing permanent advisers, particularly on this issue.

Whether such an adviser would be permanent is a matter for consideration.

I think that we should have a permanent adviser, but we can discuss that in committee.

Elaine Thomson rightly raised the issue of what other committees are doing with the budget. Have you established relations with other committees through the conveners committee so that we do not get an issue dumped on us for a yes or no? We need flows of information from other committees about the budget process.

The conveners committee has had only one meeting since we last spoke about that and there was no opportunity to raise the matter. The point is that committees need level 3 information to function effectively.

The financial issues advisory group is clear about the involvement of subject committees at stage 1 of legislation. We will need certain information in a specific form from those committees. Perhaps we should decide what form that information should take and suggest—humbly—that other committees follow our recommendations.

But that has already been agreed.

I know, but we could consider the format.

Right. I thank Brian Ashcroft and Stephen Boyle for joining us.

Meeting closed at 12:03.