Budget Setting and Financial Scrutiny
Good morning, colleagues and welcome to our meeting this morning. We have received apologies from Richard Simpson.
The first item on our agenda concerns external research that the committee commissioned. Professor Colin Talbot is here to speak to that research, which, as members know, relates to the budget-setting process in other legislatures in the United Kingdom and other parts of the world. An extensive paper has been produced. After Professor Talbot has said a few words, I will open up the meeting to questions from members.
Professor Colin Talbot (Pubic Futures):
Thank you for inviting me to give evidence—it is nice to be back in Edinburgh.
Our research relates exclusively to international comparisons. We have not examined the Scottish Parliament's budget processes, except to consider what the official documents say about what the Parliament is currently doing. The lessons that we have been able to learn from international experiences are based purely on those experiences. However, I am involved in budget processes elsewhere in the UK. I have given evidence three times to the Treasury Select Committee of the House of Commons on the new spending review process and public service agreements. At the moment I am acting as specialist adviser to the Public Administration Select Committee of the House of Commons, which is considering the setting of targets across UK public services.
As members will see from the case studies, we selected a group of colleagues from the countries concerned to conduct the research for the report that we have produced. That approach has strengths and weaknesses. Its strengths are that those who carried out the research have far more local knowledge than I do about what happens in their countries. However, as members who have tried to wade through the case studies will have found, sometimes translating what happens in other countries into UK domestic experience is difficult, because of cultural and administrative differences between jurisdictions.
We also tried to examine the wider literature on reforms of budget processes—in particular, reforms to improve participation in, and the openness and accountability of systems. The 11 conclusions that we have presented in the report draw on that review of the wider literature and the case studies. The conclusions are the part of the report that is likely to be of most interest to members, so I shall take a few minutes to go through those.
In Scotland there has been an attempt to introduce a system that is very different from the processes that existed under the former Scottish Office. However, reforming budget processes is inherently difficult for a number of reasons. It is difficult to convey simply the complexity of modern states and their financial activities to Executives and members of legislatures, let alone to the general public. There is also a problem of inertia. My experience of working as a consultant with organisations and of observing what happens from outside is that new processes take a long time to bed down; it takes time for people to start behaving differently. In Wales, many civil servants have taken a long time to change the way in which they thought about their accountabilities and relationships with Whitehall departments since devolution, but I am sure that there is not the same problem in Scotland.
Any problems that the Finance Committee might have encountered are part of the bedding-down process. The committee might need simply to keep doing what it has been attempting to do. However, in response to issues that were highlighted by the research, the committee may want to make some small changes to encourage the sort of process that it is trying to develop. I will summarise those issues briefly.
The first point relates to longer planning horizons. For good reasons, the budget process that was introduced in Scotland at devolution was based on a cautious approach. At the time, the UK-wide public expenditure system was going through substantial changes and no one was sure how it would settle down.
We now have much more experience of how the system is running at UK level. It is interesting to note that, in almost all of the case studies that we examined, people are examining ways in which to develop longer planning horizons for public services bodies, but there is enormous variation. An example that is not among our case studies is what happens at state level in Sweden. There, people have been looking at differential planning horizons for different parts of the public sector, depending on the relative stability of the finances and the needs that they are addressing.
There are some areas of the public sector in which quite long planning horizons can be set, because it is known that the situation will stay relatively stable and that priorities will not change much. In other areas, the situation might be much more volatile. That is somewhat akin to the distinction that has been made in UK public finances between annually managed expenditure and the new departmental expenditure limits. In almost all cases, there has been an attempt to introduce longer planning horizons. From what I have read about what is going on in Scotland, and from the case studies, I conclude that there is something in the Scottish processes that needs to be reviewed. Perhaps you need to think about adjusting the planning process to bring it more in line with the spending review process at UK level, so that it fits better in terms of planning on a slightly longer horizon.
The second point that we came across is interesting and is a possibility for Scotland. In a number of cases, we came across examples of legislative budget offices.
I am sorry; I missed out a point. The second point that I meant to make was on the Executives and legislatures. One of the things that emerged from all the case studies was that in all of them there is a very different relationship between the Executive and the legislature than exists in the traditional Whitehall model. All of them have struck more of a balance between Executive and legislative input to the budget formation process than in the traditional Whitehall model, where the legislature has no input whatever into the budget process, either in formulation or in signing it off. All the case studies revealed something closer to the model that you are trying to develop in Scotland.
The existence of some sort of legislative budget office has increased the capability of some legislatures by acting as a counterweight to the place where all the information is generally held in the Executives of those different countries. Such an office provides a legislature with better analyses, better guidelines and more advice, and with a pool of experts to help it to examine budget issues in more detail. I certainly think that that should be considered and examined in the Scottish context.
The fourth point concerns institutionalised budget consultation. It is interesting to note that, in most of the European examples that we studied, and in some others that I know about, there is nearly always some form of consultation process. That might take place between the Executive and civic society or between the legislature and civic society, or both. Very often, there is some sort of formal council that is made up of civic society actors. In the Flemish and Tuscan examples, there are detailed cases of such bodies' being consulted formally as part of the budget process. Although there has been a small degree of that in Scotland, it is something that the Parliament might want to consider for the future. It might be a way of involving more people in the discussion.
A number of the points that we have made interlink with one another. For example, if you start to think in terms of longer planning horizons, that opens up more space for consultative processes to take place. If you are simply working on an annualised basis, consultation is obviously more difficult.
The fifth point concerns more innovative forms of budget consultation. At the end of that section of the report, there is a list of things that have already been done—by the Enterprise and Lifelong Learning Committee, for example. There are numerous examples in our case studies of all sorts of consultation processes taking place, and I know of many more. There are many innovative ways of trying to get more people involved and engaged in something that is not just for officials and for the legislature.
The next point is on linking budgets to objectives and performance. That is something that has happened in Wales to a slightly greater extent than it has in Scotland, and it is certainly happening in most of the other jurisdictions that we studied. There is a fairly clear tide towards trying to account not only for what is being spent in the public sector, but for what is being achieved by it. Attempts are being made to link the allocation of resources more closely to what is being achieved in performance terms. That means better housing conditions, better social conditions, better health and better personal or international security. I am interested in that subject, which is why I am acting as an adviser to the Public Administration Select Committee on it. That general trend is apparent in most of the Organisation for Economic Co-operation and Development countries, and some of the case studies contain specific lessons about how that is being done.
My last point, before I go on to talk about some of the trickier issues, is on rolling reviews of programmes, departments and performance. One of the things that has arisen in Australia, in one or two of the states of the United States and in other places has been the idea that there should be attempts to identify particular areas of expenditure and programme activity for periodical in-depth review. That approach is taken rather than attempting to have an all-encompassing process that looks at everything every year—or every budgetary cycle, if the budget is operated on a spending-review cycle. Over time, a number of items will be looked at across the board, but no attempt is made to consider them all at once.
One thing that is not in our case studies, but which I know about from previous work, is a programme of evaluative reviews of areas of federal expenditure in Australia in the 1990s. They went through all those areas one at a time over about five or six years, which provided an opportunity for in-depth coverage and the possibility of wider consultation and involvement on specific issues. That might also relate to my previous point about opening the process up to more innovative forms of consultation. If more in-depth reviews are carried out, if more interest is generated and if more bodies are involved in a specific area of activity, more innovative consultation becomes possible.
I will not dwell too long on my point about the staged budget process, other than to say that the sort of staged budget that the Scottish Parliament has introduced is generally in line with the reforms that are taking place in a number of other countries.
Much more difficult is the issue of simplified budget documentation and budget formats. That is a much trickier problem, and the lessons from the case studies that we have examined are much more mixed. Essentially, the matter boils down to the argument that making documents simpler means that it is easier to make aggregate strategic decisions about the allocation of resources. The difficulty with that is that specific interest groups and sectors often have an interest in the minutiae of specific small allocations of resources and how they are conducted. The use of the more strategic aggregate numbers can sometimes prevent participation and involvement, because people cannot see how their specific concerns link to big budget numbers about health, education or whatever area it is.
The case studies show that it is always a tricky problem to try to simplify documents sufficiently for people to be able to understand what is going on, while enabling them to have access to the information that concerns them. A number of issues are linked to that. For example, accruals accounting has very mixed effects. It gives us a very clear and accurate picture of what is going on in public finances, which is extremely useful in planning public finances, but it has a downside in that it is much more confusing for the general public, for legislatures and even for Executives. I have worked with a number of Government departments trying to implement resource accounting and budgeting, and even senior mandarins now have trouble getting their heads around what some of the accounts for their own departments actually mean. There are swings and roundabouts in that approach.
My final point is a bit more speculative and might seem slightly off the wall in relation to the situation in Scotland. Although the effects of the Barnett formula and the funding regime for the Scottish Parliament might make it appear that the finances are relatively stable and predictable over time, they vary over time in a number of ways. As a result it might be worth while—if the committee is attempting to address the issue of more long-term planning in terms of budget expenditure—to have in place long-term estimates of what income is likely to be. One idea that we picked up on—which is used in a number of places, in particular in the United States—is that of having consensus revenue estimates. In those, external experts are asked to produce predictions about the likely finances of public bodies over the next three to five years, and therefore to predict what the budget envelopes are likely to be so that planning can be done more realistically.
Those are the general points. I have skated over an in-depth report. The total length of the report, including the case studies, is 300 or 400 pages, so it is difficult to summarise all the issues in 10 minutes. However, those are the major points that we picked out and which we thought would fit best with the direction in which Scottish policy is moving on the issues. A lot of other points in the report are probably not applicable in the Scottish context.
Thank you very much. As you say, it is a very full report; it is also interesting and will be of considerable use to a future Finance Committee after the elections in May, if that committee wants to broaden the way in which it examines the budgets.
As the convener says, the report is interesting. Longer planning horizons have been discussed on and off over the past three or four years. Are you suggesting that in order to fit in with the spending review, which has undoubtedly caused a number of difficulties in scrutiny of the budget, we should move away from an annual budget to consider a two-year budget?
There might be a third way, if I can use that phrase. In Westminster, there is still a spending review and an annual budget. The spending review sets out the broad parameters for public expenditure. First, it sets out the parameters in terms of the gross amounts to be spent by the public sector. Secondly, it sets those out—supposedly firmly—for three years in advance, although in practice it is for two years because the spending reviews are conducted every two years. Thirdly, the Chancellor of the Exchequer makes annual adjustments. In some cases adjustments are made within the year; substantial amounts of extra money have been allocated to, for example, health.
The use of longer planning horizons does not preclude also having an annual budget cycle, but it means that it is hoped that some of the bigger decisions can be taken in a biennial cycle that is linked to the spending process. Some discussions on the big issues, such as allocation of the budget on health versus education or health versus social services can be held on a two-yearly cycle. More detailed discussions about allocations within the wider budgets can take place annually. There are ways of attempting to do both.
Is there a danger that the three-year process—now a two-year process—is in a sense becoming a myth because chancellors, or whoever, have to yield to the temptation to deal with political reality and priorities regardless of what happens in an expenditure review? To what extent is the two-year process merely an exercise that is gone through, while the reality is that changes are made every year or half year as political imperatives dictate?
If you consider the process as a whole, that would be a fair point. If you consider total public expenditure, there has not been a huge amount of variation from the proposals within the various spending reviews. The Government has kept fairly strictly to the spending totals. Some shifts have taken place within that, but the shifts have varied across the public sector, so some areas of public expenditure UK-wide have been much more stable as a result of the spending review process, although there has been more instability in some areas. In most of the sectors in which there has been instability the changes have been in an upward direction, so people have not been screaming too much about that. If spending had been going in the other direction to the same extent I suspect that there would have been more reaction.
I wait with interest for one of my colleagues to do research on how much the variation in public expenditure under the new spending review process is greater or less than under the old public expenditure survey, which included a three-year look ahead at public finances. My hunch is that year 3 under the PES would not be much different from year 3 under the spending review process. That said, in certain areas of public expenditure the stability has been welcomed by public sector managers overall. They are obviously able to do more long-term planning. One of the crucial points is that sticking to a completely annualised budget system makes any sort of end-year flexibility impossible, or almost impossible. It makes it impossible for public sector managers to plan over time horizons that are longer than a year, so having at least some notional longer-term planning horizons can help.
You mentioned the possibility of having longer estimates of revenue in the future. To what extent is that possible in a devolved administration where by definition—certainly in our case—most of the revenue is determined by somebody else's decisions?
In the examples from the United States in which revenue estimates take place, tax income is estimated which, as we all know, is inherently unstable. Gordon Brown is currently finding that out. Those legislatures estimate several economic and fiscal issues that are inherently indeterminate in the future and difficult to gauge, in order to gauge what their likely budget will be. I suggest that that is not dissimilar to the situation in a devolved Parliament; they consider external factors over which they have no control, so to that extent the Scottish Parliament and Scottish Executive is in the same position. The use of guesstimates about where you think things will be in three years time is as plausible for the Scottish Parliament and the Executive—in some ways it is more plausible—as it is for people who have to make guesstimates on gross domestic product, turnover, tax take and so on.
How effective in eliciting change is public consultation in other Administration areas?
Effectiveness varies. In the time that was available for the research, we were not able to carry out empirical research on what the impacts of the consultation processes on budgets were in the various jurisdictions. The message that we got from our colleagues, in discussions with them and in the case studies, is that the impact is very mixed.
In Belgium, for example, one of the concerns is that consultation had too much influence on the budget and that the decentralisation of control of the budget and the input of a range of actors in the budget process disempowered the politicians. One of the agendas in the Flemish Community in Belgium is about re-empowering the politicians to put them back in charge of the budget process. One of the dilemmas that we point out in the report is that there is always a tension between representative democracy—the Parliament and the Executive—taking decisions about where budgets ought to go, and consultation that involves other people in the process. There are examples in which consultation has had a considerable effect, but there are many examples in which consultation has had an effect at a minor level, which might not have a big effect on the budget overall, but might be important for individuals. People have been able to lobby on and get funding for particular activities and areas of action. That is important to them.
There are many other examples of where the process has become relatively routine and bureaucratised. I suspect that that is true of the Italian case. We understood from the Tuscan example that much of its consultation has been fairly routine and does not have much impact on the decision-making process. The answer is that the effectiveness of consultation can be very mixed.
Do you feel that there has been dilution of accountability, in particular in the Flemish model?
There is bound to be some such dilution. If the decision-making process is to be shared and opened up to consultation, and that consultation taken seriously, there is bound to be a certain amount of dilution of politicians' accountability, because other people have been involved in making the decisions. Executives and legislatures must make political judgments about how far they want to go towards sharing that responsibility.
We have cited some of the more extreme cases: California and Oregon, for example, have legislatures that have abandoned responsibility by allowing decisions to be made by plebiscite or referendum. I do not think that Scotland is moving in that direction, which is why we did not recommend it.
Further to what you just said, was there any evidence that single issues and tabloid-type stories could sway, blur or damage the decision-making process in a plebiscite system?
There was no such evidence from our case studies because none of the places that we considered has been directly affected by that. However, I know from examples such as Oregon and, in particular, California that false majorities have been created behind a particular reform that everyone can sign up to. The questions might be "Do I want to pay more taxes?" or "Do I want to pay this particular tax?" The answer is usually "No", but everyone can easily answer such questions when they do not have to answer the next question, which is usually "What are we going to tax or cut instead?" In such circumstances, it is very easy to generate false majorities—that has happened—and public sector managers are left to pick up the pieces.
I thank Professor Talbot very much. As I said earlier, you have produced very interesting work and I am confident that it will guide the work of the Finance Committee after the next election.