Finance Committee, 20 Jun 2006

Meeting date: Tuesday, June 20, 2006

Official Report 280KB pdf

---

Efficient Government (Technical Notes)

The next item on our agenda is consideration of the Scottish Executive's updated efficiency technical notes. Our budget adviser, Professor Arthur Midwinter, has produced a paper, which has been circulated to members along with a copy of "Efficiency Technical Notes March 2006".

I will invite Arthur Midwinter to speak briefly to his paper and then invite comments from members. Arthur, welcome back.

Thank you, convener. Unfortunately, the updated efficiency notes publication is one of those documents in which the devil is in the detail. I propose simply to summarise what I see as the main problems without giving examples. If members want to go into the details, we can deal with those later.

As we have four new members since we began this exercise, I feel that members need to be fully aware of the points that have been made in the past if they are to be able to criticise the latest document properly. That is why I have produced a background summary of our previous criticisms. Particularly for the new members, I should explain that since the introduction of the efficient government initiative, the savings target for cash-releasing savings has increased. In June 2004, it was originally set at £500 million. In November of that year, the target increased to £745 million but that figure also included Scottish Water, which is outwith the departmental expenditure limit. The figure for cash-releasing savings in the new efficiency technical notes is about £810 million.

The two previous documents had two fundamental problems, which I categorise broadly as relating to the identification, monitoring and measurement of savings and to the tracking and reporting of the growth in outputs in front-line services. In several reports, the committee has asked a series of questions about the savings assumptions, which were based on the original Gershon model in England, the quality of the cost and output data and the absence of baseline data, particularly for 2004-05. The committee has expressed concern about delivery problems when agencies at the grass roots are relied on to deliver savings; the savings exercise in spending review 2004, when savings were reallocated along with the new growth moneys through the Barnett formula; and the measurement of those savings by the Executive. The Executive estimated output growth at 5 per cent per annum on the basis of the national accounts approach, which is not particularly helpful. We must also be able to measure inputs and outputs for something to be an efficiency saving. Such criticisms were made over two series of notes.

When the committee reported in December on those matters and drew attention to the fact that the figure for the Executive was still significantly below comparative savings levels in Whitehall, its report was heavily criticised in The Scotsman for making unfair comparisons with Whitehall and for questioning the hard evidence from Scotland while taking Gordon Brown's information at face value. Those criticisms are hopelessly misplaced. First, it was the First Minister who made the comparison and it is the committee's duty to scrutinise robustly the spin and the substance of Executive action. We would fail in our duty if we did not draw attention to the matter, given that the First Minister made the claim. Secondly, the notion that the efficiency technical notes as they stood in December constituted hard evidence of saving is difficult to take seriously. In fact, the notes have been subject to widespread criticism.

The new efficiency technical notes are much better. I am pleased to say that the text contains several examples that meet the committee's recommendation in its December report that much clearer identification should be given of input and output data and of baselines. Executive officials in the efficient government group have revised their template to require departments to provide that information. In roughly 75 to 80 per cent of cases, I am content with how departments have responded. As we have disagreed with officials in the past, it is important to say well done to them for pursuing the matter. As I said, I am much happier with the document.

We now have something like 57 efficiency technical notes for cash-releasing savings. I will spend the rest of my time on dealing with the remaining information gaps, because that is where problems arise. I will summarise, but my paper and the notes contain much detail.

In about a dozen efficiency technical notes, the statements of outputs are still too vague to be converted into the ratio of input to output that we need for efficiency savings. About a dozen ETNs also exclude all development costs—up-front moneys that are spent to allow a project to happen—usually on the ground that the costs were incurred for reasons other than efficiency. Whether such costs should be included is a matter for debate. I certainly believe that an element of development costs should have been included. Some ETNs included projects that were decided on before the efficient government initiative was launched and I have doubts about whether they should be claimed as efficiency savings under the exercise. In half a dozen ETNs, development costs are still reported as uncertain or unknown at this time.

In some cases, I would describe the checks on quality, which we need, as being not fit for purpose. That is because the proposal is often to use the targets that the Executive sets in the Scottish budget or the statutory performance indicators that Audit Scotland creates for local government, both of which are quite useless for this purpose because the performance of the local authorities rises and falls against them year on year—it has nothing to do with efficiency savings. For quality checks, we need specific outputs that relate to the specific projects and not general statements of that nature. There is also a lack of hard data in the documents on the reallocation of savings. That goes back to the original decision to allocate the money by combining the savings with the growth moneys through Barnett. To be fair, the minister acknowledged the criticism. In asking departments to make revisions, he said that, where possible, they should highlight for us where the money has been redistributed. However, very few of them are able to do that.

There are two standard departmental responses in the document: either that the moneys released resources that were then made available according to ministerial priorities; or that they were retained by the department for use on departmental priorities. Neither of those responses tells us how the moneys were used. The information is not quantified, so we have no way of knowing whether the growth in output that we want to see in front-line services is taking place. Unless we get to the specifics, we cannot deal with the matter. In those cases, the checks on quality are not fit for purpose. I have listed the ETNs that I think are problematic at the end of the paper.

I have two overall conclusions on the document. First, I remain doubtful that the actual savings will reach £810 million. I have supplied the committee with a short note on one example of an ETN that I queried directly with the officials who are concerned with the rail franchise. As members will see, the ETN was originally inserted as a cash-releasing saving. The officials now accept that it does not quite fit that definition, as there was no saving on the baseline. That one will be reclassified. The decisions on a number of projects were taken prior to the beginning of the process. There is a problem with development costs and a lack of output data. If there are no specific output measures, it is not possible to claim that something is an efficiency saving—it may be a saving but it is not an efficiency saving. I cannot stress that enough.

Secondly, the growth in outputs will not be recorded at 5 per cent per annum, as we were originally advised. I welcome the Executive's recognition of the need for what it describes as the more equitable approach that it will take next time to the allocation of savings between departments. Clearly, some departments got off very lightly in the exercise and were able to produce proposals for savings that were already part of their development plans. Therefore, despite the improvements that have been made, I continue to hold the overall view that monitoring of savings is not wholly robust and that the monitoring of growth of outputs is minimal.

Okay. Thank you very much. It is probably worth placing on record the fact that the Executive has responded positively to some of the messages that the committee has been sending it over a fairly long period of time. We welcome the greater definition of the measurements and better monitoring of efficiency savings.

Obviously, a number of areas that Arthur Midwinter identified continue to show a lack of definition or deficiency in terms of outputs or baselines. The document is clear in identifying those deficiencies and we will focus on them. It is worth placing on record the fact that we have been fairly robust in our dealings on the issue. As a committee, we should recognise that the process has been productive; we are getting some tangible information and improvements.

Is the improvement that we are getting in any sense comparable with the information that comes from down south, to use the comparator that we should not use? When we consider efficiency technical notes or their equivalent for United Kingdom departments, do the same issues arise?

I have enough difficulty keeping up with the amount of information here. I have not attempted to look at the paperwork south of the border. Given that they have 10 times as much as we have, it would be quite difficult to deal with.

The comparisons were about the ambition of the targets rather than the detail of the savings that were made, although I saw a report in the Chartered Institute of Public Finance and Accountancy's journal, Public Finance, that said that the forecasts were that the English totals would exceed the targets. I do not know what the detailed basis for those forecasts is.

We must be satisfied that the efficiency technical notes stand up to robust scrutiny against the measure that the Executive has set for them. I am quite happy to place on record the fact that there have been improvements in the situation. However, it is also apparent that there is a long way to go before this process becomes as robust and deeply rooted as the committee would like it to be.

In the case of the rail franchise procurement process, which concerns a sizeable supposed efficiency saving of £24.7 million, the effort that had to be made to prove that there is a weakness in that figure and that it should not be counted as an efficiency saving is an illustration of how much has still to be undertaken. The fact that the impact of the efficiency savings has been felt disproportionately by the Finance and Central Services Department, the Health Department, the Justice Department, the Enterprise, Transport and Lifelong Learning Department, the Crown Office and the Education Department and that, we believe, a host of others have been left largely untouched shows that we are a long way away from having a regular, dependable, reliable and robust process by which we can test the Government's efficiency. We welcome the progress that has been made but there is still a lot to be done.

I think that you have given a fair summary of where we are at. Because of how this was handled in 2004, we may not resolve the situation in this round. The original decision to put all of the money in one pot and not require people to say what they were doing with it means that we are unlikely to get the kind of information that we would want to be able to say confidently that the increase in outputs is 4 per cent, 5 per cent or whatever. Likewise, the more equitable approach between departments will have to wait until the next spending review exercise.

However hard the committee has tried to get to the bottom of the validity and robustness of the efficiency technical notes, our task has been made impossible by the way in which we started on this journey some years ago.

Professor Midwinter, I had a look at the information regarding the schools building programme, on page 46 of your document, which is one of the examples that you drew attention to. You say that the saving is a time-releasing saving, not a cash-releasing saving, and you use it as an example of the kind of problems that arise.

I am interested in what could be done to resolve the issues around situations such as that. In point 8, you say:

"The school building programme is not being pursued on efficiency grounds alone and therefore it would not be appropriate to net off any development costs".

Would you propose that a proportion of development costs be considered?

In point 12, you say:

"There are no formal mechanisms for monitoring the delivery of these efficiency arrangements".

Some kind of mechanism is needed to measure the efficiency savings, but if mechanisms do not exist at the moment, is it cost-effective to put them in place, given that they might cost more than any time savings that are achieved?

I have a fairly straightforward view, which is that savings of that kind are being quantified for no other reason than to report them. Government has always made savings like that; it is part of an on-going process and happens every time the Government invests money, does something in a new way or redesigns something. In the past, no one ever had to report such savings. A project of that nature should not be included in an efficient government exercise. Making savings should be the criterion that drives a savings exercise. When departments have on-going work, they are able to return the figures as an efficiency saving, but my view is quite simple: they should not be in there as efficiency savings.

That goes back to how the system was developed in the first place. Efficiency saving exercises should concentrate on savings that are generated as a result of the exercise being created rather than allow departments to include projects the spending proposals for which were approved before the efficiency exercise started.

I do not know if that helps. I am quite happy to acknowledge that such projects have efficiency gains, if we want to describe them in that way, but for me they are not part of the savings drive. They would have happened anyway and five years ago they would not have been reported as an efficiency saving. Even worse, we have people trying to cost those savings. Economists get into trouble when they start trying to quantify something that is not cash-based.

You mention the schools building programme in your report but you do not list it as one of the problematic projects in the appendices.

I used that example because of the statement that the projects are not being pursued on efficiency grounds. That causes confusion between projects. Some departments have had to find efficiency savings in their budgets and report to the Executive on what they are doing. Others have been able to say what they were doing anyway and, as a result, have made an efficiency gain that can be quantified—although the gain is in time—and can report that the department has made a saving. I do not think that that should be part of the exercise.

Have you analysed what proportion of the efficiency technical notes reflect genuine efficiencies, where a department has had to look at its budget and make a saving, and the proportion of savings that are spurious, because the departments would have made them anyway?

I have listed a dozen or so projects about which I have doubts because they fit into the category that you describe. In the main, it is my job to advise the Finance Committee on the budget. I have considered only the cash-releasing savings because, as I said earlier, I have severe doubts about the merit of doing the time-releasing exercise in the first place. There are question marks over whether about 12 of the 57 ETNs ought to have been included in the first place because they were driven by policy development rather than efficiency savings.

Like others, I think that the document is slightly more transparent than the previous editions were, but there is still a long way to go.

Convener, do you mean to follow up on the dozen or so issues on which we need more specific and transparent advice? Will we be writing to the minister to ask specifically for that?

Also, there are queries in appendix 2 over development costs.

If there are very specific questions, we can raise them in writing at official level with the department. We will take evidence from the minister once the ETNs are published and will have the opportunity to quiz him on any issues then. You may remember that we are in correspondence on project A/C3, on human resources transformation. That correspondence has been circulated to members.

So should we just leave the matter for now? I am trying to decide on the best approach. Should we wait until we meet the minister and have further information, or should we try to get more information now?

If we need to follow up specific issues, we can do that with officials, but general policy issues are probably best left until we speak to the minister.

There are philosophical disagreements about whether development costs should be included. I have challenged the Executive's line on that and the committee has queried it in the past. It is for the committee to raise that with the minister. We will have no difficulty in getting information from the officials about details that members want to raise. However, whether the development costs should be included involves a political judgment that the minister must defend.

What about the items in appendix 1, on which we seek more specific information?

I guess that such detailed comments will eventually find their way into the December budget report—that is what has happened in the past. The Executive will respond to the recommendations at that time, as it has done with the previous couple of reports, and say whether it will do something. We have a list of issues on which we think we need better output data. I presume that the Executive will respond once the list is converted into a set of formal recommendations in the budget report in December.

The paper is terrific and summarises the issues tightly. It contains much that I found valuable. The conclusion that I have drawn is that if, rather than being Arthur Midwinter, you were KPMG auditing a major company, the exercise would have been the equivalent of qualifying the accounts, because it has so many caveats built into it.

The fee would be more expensive.

I will accept KPMG rates.

I am keen to focus on inputs and outputs. I worry about outputs a lot. If we were examining the Ford corporation, outputs would be having more employees, dealerships and adverts and producing more vehicles. However, in essence, that is not what the shareholders or analysts are interested in; they are interested in outcomes, which are basically how profitable the business is, the strength of the balance sheet, the cash position and the value of the shares. I have to believe that, somewhere in the programme, there are outcome values that we can get our heads round and hold Government to account on, especially for cases in which we can identify the baselines and move forward from there. I suspect that, in most of the local authority areas with which members' constituencies are coterminous, frontline services such as free care and school buses are shrinking dramatically. We received evidence from Colin Mair, the chief executive of the Improvement Service, but he failed dramatically any real-world tests. Given that you have identified the programme as a wheel with bearings that are giving considerable wobble, my key question is whether that is a sufficiently robust platform on which to build public sector reform?

That is a much larger question than is covered in the paper. We are considering the management of public finances and best value, which are presented to us in the "Framework for Economic Development in Scotland" as the mechanisms through which the Executive will improve the productivity of the public sector and so contribute to economic growth. However, it is clear that there is a mismatch between the way in which the Executive measures economic growth and the way in which it does these exercises.

I do not know whether you picked up on this, but under the current approach for measuring economic growth, the Executive makes no distinction between back-office services and front-line services. Thinking simply about the efficiency exercise, I am not worried—we are getting towards having a model that we can use to measure efficiency savings. However, I have real doubts about whether that will throw any light on the bigger question of how the savings contribute to economic growth. It is certainly feasible to build on where we are now simply as an exercise in measuring the change in outputs. However, I do not think that we will ever be able to make the link between the exercise and the economic growth targets.

I agree. I reiterate that, when taxpayers and the general public look at the engine of government, they are not going to see from the exercise £810 million of incremental outcomes that will impact on their lives.

From the beginning, the Executive made it clear that the exercise was about inputs and outputs; it did not present it in outcome terms. The outcomes are a bigger question the answer to which we will pull together when we examine the success of the mechanisms as a whole at driving the outcomes that we want. We will look at the efficiency technical notes, the best-value regime, the baseline review when it comes out and the pre-expenditure assessment and consider whether the whole package helps to—

I agree with everything that you say, but the fundamental point is that the efficiency technical notes document is grotesquely complex. At the weekend, I was invited to comment on a new advocacy of lean management techniques. The efficiency technical notes are a million miles away from that.

When I look at what is happening around the planet, I see that even cities such as Fort Wayne in Indiana in the States are buying into genuinely implementing savings that people understand and which they can see being made on an incremental basis. Instead of a three-year outlook, they have a perpetual outlook on where they are going and there is perpetual improvement, year in, year out. The comparison with the efficiency technical notes is just night and day.

I have had my say.

I presume that there will be a further version of the efficiency technical notes in due course.

There will be two things. In the next month or so, we should receive the annual report. I assume that, after that, the efficiency technical notes will be revised again. It is about eight or nine months since they were revised. They were supposed to be out in March but they were slightly late in coming out. I hope that we will get the final set before the end of the parliamentary session, if the Executive is on schedule.

Also, a report by Audit Scotland is due to come out at some point.

I think that it is due out in the autumn. Audit Scotland will consider the efficiency technical notes over the summer and respond to them. Audit Scotland is required to give assurances that the savings have been delivered.

In addition to reflecting on the detail of the issue, I suggest that we start to think about how a better system of management of efficiency savings could be constructed in the context of the next spending review. It seems to me that there are lessons to be learned from the processes that the Executive and the committee have been through. We should think about the process of specifying the system of efficiency management and consider how we can do that better, because we know that it will have to be repeated in the context of the next spending review, irrespective of the content that is chosen. It would be useful for us to undertake that exercise.

That would be a constructive contribution.

Are members agreeable to that exercise being explored?

Members indicated agreement.

I suggest that there are lots of practitioners out there whom we might pull in to give evidence, even just to a subset of us.

It might be as well to get Arthur Midwinter to do some preliminary work on the issue, and then we could—

I certainly found it helpful when we had the practitioners in when we were talking about how to refine the budget. I think that the idea is worth while. If you let me do some exploration and come up with some ideas, that would be good.

As members are content with that, I thank Arthur for his paper, which is very useful.