Finance Committee

Meeting date: Wednesday, December 4, 2013

Official Report 458KB pdf

---

Independent Fiscal Body Inquiry

Our second item of business is the taking of evidence by videoconference from the Organisation for Economic Co-operation and Development as part of our inquiry into proposals for an independent fiscal body. It is my pleasure to welcome to the meeting Mr Sebastian Barnes and Ms Lisa von Trapp. There will be no opening statement, so we will go straight to questions. As is normal at Finance Committee meetings, I will ask a few opening questions and will then invite questions from colleagues around the table.

In my introductory hellos to you before the meeting, I spoke rather rapidly. Given that you may struggle with our accents a wee bit, I will speak as clearly as is possible for someone with my west of Scotland/Glasgow background. I welcome you to our committee and appreciate your taking the time to give us the benefit of your experience in this important area. You have already given us answers to a number of questions in your written submission. I will go over some of those for the benefit of the Official Report and to provide clarification for members.

I would like first to touch on the OECD principles for an independent fiscal institution. The OECD working party of senior budget officials has put together a first set of draft principles. Can either—or both—of you explain what you believe the core of those principles should be?

I reassure you that my best friend is from Glasgow, so I should be okay with your accent. So far, so good, anyway.

I will start by giving you a little bit of background information about how we developed the principles. They have been approved and are no longer considered a draft, but we are taking them forward to become a council recommendation, which means that they will be elevated later this year.

The principles were developed initially in our parliamentary budget officials and independent fiscal institutions network, so they draw on the experience of existing independent fiscal institutions. Some of those are much older, such as the Netherlands and the Congressional Budget Office; some have about 10 years of experience, such as Korea; and some are newer ones that have been coming up. The principles are grounded in institutions’ experiences.

It is difficult to say which of the principles is the most important, as we think that they are all important. They are interconnected and, although they do not all appear under the heading of independence and non-partisanship, many of the principles seek to reinforce independence and non-partisanship as well as transparency. I would, therefore, highlight independence, non-partisanship and transparency as the most important principles.

In drafting our written submission, I thought about a composite indicator of the most important things for reinforcing those principles of independence, non-partisanship and transparency in an institution. Those would be things such as leadership appointments made on the basis of merit and technical competence; clearly defined term lengths for the leadership, which would preferably be independent of the electoral cycle; clearly defined criteria and process for dismissal; and the institution having control over the hiring and firing processes for its own staff, within local labour laws.

The institution should also have the ability to set its own work programme and to produce analysis on its own initiative. It should have a separate budget line and multi-annual funding commitments, where that is possible. There are some examples of countries where that is possible, such as Australia. The institution should have a guarantee of access to information in legislation—that might need to be reaffirmed through a memorandum of understanding—and it should provide full transparency on its work and operations.

You take the view that it is “highly desirable” that such a body should be established on a statutory basis. Can you talk us through your reasoning for that?

We do. We recommend that the body should be established in higher-level legislation partly because, if it goes through some initial growing pains, it is more difficult to shut a body down if it has been established on a statutory basis in higher-level legislation. In addition, that gives weight to the body in the eyes of the public and of the other institutions that it has to interact with.

I will offer a perspective. As well as working for the OECD, I sit on the Irish Fiscal Advisory Council, so I sit on both sides of the discussion. One thing that I would draw from that practical experience is that, when it comes to the principles, the important thing is the mandate. We are talking about the mandate being in legislation. That could be read as being just about the division of labour—in other words, what the council will do—but I think that the mandate has huge significance, because it defines the way in which the council will work. If an individual sits on a council and their mandate is to assess the Government’s forecasts, they will take that assessment business very seriously, because it is a different task from simple forecasting.

It is also extremely important from the point of view of the council’s impact with the public that it has such a mandate. In Ireland, it struck me that as soon as the Irish Fiscal Advisory Council was set up and had produced its first report, we were accepted as being an official body. Our opinion was not just the opinion of X on the fiscal position or the accounts; it was the opinion of an official body. That is extremely important to the council playing its role. Some think tank can easily be dismissed because it does not have an official mandate to do its job. That is why legislation affords a level of protection and importance.

I think that the committee has looked at the example of the Office for Budget Responsibility and the Institute for Fiscal Studies in the United Kingdom. The difference between the OBR, which has been given that mandate, and the IFS, which still does very good work, is that, in the eyes of the public, the OBR carries greater weight, because it has been mandated by the Government to perform its role. That is a good illustration of what Sebastian Barnes is saying.

One of the things that you mentioned was the importance of local ownership. You said:

“To be effective and enduring, an IFI requires broad national ownership, commitment, and consensus across the political spectrum.”

How should we go about securing that? You touched on issues such as transparency, but I would like you to expand on how you believe that we can ensure that we get that broad spectrum of acceptance.

One example of how countries have achieved that is through Parliament. That highlights the importance of the role of Parliament, which brings together the different party groups. There are institutions such as the Slovak Republic’s Council for Budget Responsibility, on the establishment of which a member of each party’s parliamentary group signed off. That meant that there was a broad political consensus that was explicit, which came through the Parliament. That is one way in which we have seen that happen, at least in the initial days.

The situation in Canada was similar—there was a broad political consensus on the body that was established there, which came through the party groups in the Parliament. Having such parliamentary involvement means that the body is seen not just as something that the Government of the day has set up, but as something that all parties stand behind.

Accountability is a key issue. In your submission, you discuss whether an IFI should be accountable to the Executive or the Parliament or both. You say:

“mechanisms should be put in place to encourage appropriate accountability to the legislature.”

Can you say a wee bit more about where you think accountability would sit between the Parliament itself and the Executive or Government?

Again, that will depend on the remit that you give the body. I know that you have received submissions that have recommended that it be set under Parliament, but I do not necessarily have a view on whether it should be under Parliament or under the Executive, or whether it should stand alone. Even if the body is set up in the way that the OBR is, under the Executive, there are important ways in which it can be useful to the Parliament.

Other institutions that have established councils publish all their reports so that they are available and Parliament can use them, but maybe that is not enough. Maybe they need to have a process whereby they officially submit their reports to Parliament so that they are debated. Maybe they need to answer parliamentary questions; we assume that that would happen. I think that Sweden was a country of interest to you. In Sweden, although the council is not linked to the Parliament, the finance committee holds a hearing every year at which it uses the fiscal council’s report as the main basis for analysis of the Government’s budget proposal. Those are some of the ways in which the analysis is useful for Parliament.

The other thing that we have looked at in different country examples is whether Parliament plays a role in the appointment or dismissal process, and that is the case for about a third of the countries that we looked at. Some of them are parliamentary budget offices, so that is obviously normal for them, but Ireland, the UK and the Slovak Republic are three examples in which the Parliament has the right to appoint. I think that in the Slovak Republic the Parliament can appoint or dismiss based on a two-thirds majority, or in some cases a full majority. It is a little bit different for each council member in the case of the Slovak Republic. In Ireland, the council members cannot be dismissed without the approval of Parliament; Sebastian Barnes will correct me if I am wrong about that. In the UK, the Treasury committee has the ability to confirm appointments or dismissals, so there is that extra protection through Parliament.

I can confirm that that is correct. In the Irish case, the appointment is essentially by the minister, and there are terms of four years, but dismissal requires a vote in Parliament, which would be a major step. That gives the council a lot of protection against interference from the Government. There have been cases in Hungary, Canada and other countries in which councils have come under a lot of pressure from the Executive. Knowing that that level of protection exists is good for the fiscal advisory council and for its credibility, because people know that the council can say what is true and make tough calls if it needs to, and that its members will not all be fired. In our case, the budget is also protected.

Accountability can be an abstract term, but I find that, in the Irish experience, the interactions that we have had with the parliamentary finance committee, which we go to after every report, have been extremely useful for the council and have helped to guide its work programme. It has made us think about things in a better way than we would otherwise have done. That accountability is not just a nicety of the way in which the institutions work; it has a real bite. My sense is that Parliament finds us useful because the Irish Parliament does not really have the resources to do the kind of work that the council does; our work is a valuable input for the Parliament, because we have better access to information than the Parliament has, and a greater ability to process it. It is a sound two-way relationship, which is helpful to both sides.

It would be natural for the committee to think about how the new institution could be useful to you and to your debates. The Scottish Parliament is an interesting Parliament, in that you have an enhanced committee system compared with a lot of other Westminster-model Parliaments. The Finance Committee gives the Parliament a scrutiny unit that provides some good analysis, so you are already a bit ahead of some of the other Parliaments that we might work with. Having a fiscal council would give you an additional opportunity to have analysis that might be useful for you.

09:45

Yes. The OBR speaks to the Finance Committee. We take evidence directly from it.

An issue of real importance that we have not touched on before in taking evidence is external evaluation; I would be very interested in your views on that. In your submission, you say:

“IFIs should develop a mechanism for external evaluation of their work—to be conducted by local or international experts.”

You go on to talk about the forms that that mechanism could take. Will you expand on that a wee bit? How common is it for IFIs to have external evaluation of their work?

Some of the older institutions, such as in the Netherlands, have regular external evaluation of both the quality and scientific reliability of their work. I think that that happens on a four or five-year basis, but I would have to double-check the details of that; I think that external academics do that evaluation. We talk about external evaluation a lot in our network, and we are working on it to try to come up with a framework.

Another way in which institutions have created a continuous feedback loop for themselves with external advisers is through panels of advisers. We see that for the Congressional Budget Office, the Netherlands and the Slovak Republic. That is not a typical evaluation process, but it allows people to bring their work or something that is a bit difficult for scrutiny by someone outside and get comments and advice.

In Canada, the business community or academic economists have proactively been gone out to and asked to comment on work in progress. To give a specific example, I think that when Canada was working on costings of F1 fighter planes, people did not have the information that they needed, so they tried to find a jurisdiction that might have worked on similar things. They ended up going to the US and working with the Congressional Budget Office and the Government Accountability Office to get information. They had the work that they had done checked by another jurisdiction that had done similar work.

I do not know whether I have an example of a peer review of a full institution, but I have examples of peer reviews of work. I think that that idea comes partly from how auditor generals have developed peer review systems. We think that that is a good idea, and the independent fiscal institutions in our network want to explore it.

For the new generation of fiscal councils, of which ours is an example, as we have just been set up, that problem has not really arisen yet, but we realise that it will be time to look at it after two or three years. The logic is that there is always the question of who watches the watchdogs, and that seems to be a solution to that. It is important for the councils’ good functioning to ensure that they are doing as good a job as possible and for credibility with the public to show that there is not just a group of boffins who are meeting, but there is proper evaluation.

There is a feeling that the peer review is a promising channel because what the councils do is somewhat idiosyncratic. Having a perspective from someone else who faces the same problems is therefore probably the most useful way to go, rather than having, for example, a perspective from academics who may know about economics but may not really face the council’s practical challenges, which are partly about economics but partly about institutional functioning and how to fit into the wider system. That is why we find peer review a fairly attractive option.

In Ireland, I think that we plan to do a peer review in around 18 months’ time, by which time the council will have been going for two or three years, to ensure that people think that we are on the right track analytically and in playing our role as a council and fulfilling our mandate.

Thank you very much. I open up the session to colleagues around the table. Jamie Hepburn will be the first member to ask a question; he will be followed by Jean Urquhart.

Good morning and thank you for your evidence thus far.

In annex 1 of your submission, you have produced a breakdown of the roles of fiscal bodies in 17 member countries. It was rather interesting to note that only two of the bodies have a role in preparing official forecasts: the OBR and the CPB Netherlands Bureau for Economic Policy Analysis. The other bodies seem to have rather wider roles. I realise that you might not feel able to comment on this, but is the OBR anachronistic and should there be a wider basis to its empowerment? Secondly, how can this inform the process of establishing a body in Scotland?

Again, when you look at the question of local ownership, you have to think about the problem that you are trying to solve. The OBR, for example, was intended to solve a specific forecasting and credibility problem that might not apply in a lot of the other countries. The Netherlands Bureau for Economic Policy Analysis is a much older institution whose independent forecasting role has developed over time; it is also, with 145 staff, much bigger, partly because of the policy costing work that it carries out.

I cannot really comment on whether the OBR is anachronistic but I would say that, according to the examples that we have, it is more usual for a council to assess a Government forecast, partly because Governments do not really want to give up control of forecasts. In that sense, you could look at the creation of the OBR as a very brave move.

As far as forecasting is concerned, an issue that I think is relevant to Scotland is the trade-off that can be made here. A council can bring to the forecasting process another pair of eyes and a different perspective from that of the person who has had to put together the numbers. In Ireland, the council usually assesses the forecast but, since this year, we have also been endorsing it, which we view as being a higher standard, and in order to do that we have started our own forecasting operation. In a sense, therefore, Ireland has parallel forecasts.

There is some gain to be had with independence and in some ways it is good to have a split between those who are doing the forecasts and those who are assessing or endorsing them. However, a real issue in smaller countries is the division of their resources. Forecasting involves a lot of fixed costs; you have to set up databases, look at the indicators that emerge and just go through certain processes, and a question for small countries is whether it is better to have one set of people doing all that or to split activity between two. There is a balance to be struck in that respect, and you might want to bear that in mind in allocating resources. Of course, if you have only one team, a separate question is how and where that team should be governed but the resource question is certainly relevant to smaller economies.

In your submission, you refer to the OECD network of parliamentary budget officials and independent fiscal institutions; I confess that I had not previously been aware of that body. Do all the network’s members interact with the state legislature or Executive, or are there any examples of substate actors? After all, in the short term at least—some of us hope otherwise—any body that is established here might be a substate actor.

You will be pleased to know that members of the Parliament’s financial scrutiny unit have been attending meetings of the network for a few years now. Of course, the fact that we have made a special case of Scotland by inviting it to these meetings might be a result of my personal affection for the country; however, we will continue to invite it and if a new fiscal body were to be set up in Scotland we would, of course, invite it as well. I think that Scotland is the only case of a regional Parliament that has been represented in the network, but the example that you have asked about is not unheard of.

That is interesting.

As far as membership is concerned, the network is a fairly diverse group of people. We have independent fiscal institutions that are legislative budget offices, fiscal councils and countries with neither a legislative budget office nor a fiscal council but which send someone from the Parliament who might work for the finance committee or a unit like the financial scrutiny unit in Scotland. The network is quite a diverse body, but it is dominated by independent fiscal institutions.

So it covers a diverse range of bodies. That is helpful.

In your submission, you refer to the type of information that might be needed by any body that might be established and suggest that it have a relationship

“with HMRC, Revenue Scotland, and other relevant parts of the Scottish Government.”

Moreover, as far as the body’s powers are concerned, you recommend that

“access to information be guaranteed in legislation”,

which is actually one of the OECD principles for independent fiscal institutions. When Robert Chote gave evidence to us, he made it pretty clear that any independent fiscal body that is established by this Parliament or the Scottish Government must have access to information from Her Majesty’s Revenue and Customs. However, the problem is that we do not have legislative competence over HMRC. Do you foresee any issues in that respect? From your dealings with and experience of the bodies that are members of your network, can you highlight any parallels that can be drawn or lessons that we can learn?

Information is pretty vital not only because it is needed but because, as I have seen in Ireland, a fiscal council that does not have a good right to it will spend an unbelievable amount of time chasing it, which will obviously reduce its ability to do other things.

However, although that principle is important, in practice it can be harder to know exactly what information you want or should be put in legislation as a kind of right. Indeed, as we have discussed at the PBO, a lot of these things always depend on good will and good relationships between the institutions. There should also be a memorandum of understanding between institutions about what information will be shared, but organising that kind of thing can be complicated.

A difficult problem that might apply in the case that you are talking about is that there are a number of confidentiality issues to deal with. For example, the fact that a lot of taxes are paid by very few people or companies raises a number of difficult questions about protecting that information that we in Ireland have not gone into. However, what you need is a good mixture of the ultimate legislative backstop of the council having a right to stuff—although I appreciate that in this particular case that might be difficult—and in practice an understanding that this will work if there are good relationships, good will and some semi-formal institution such as a memorandum of understanding. Often you just need to have a conversation with someone to figure out what they know; you cannot really request a specific number.

The tax question that I mentioned might be such an issue, and it would be good to have a relationship in which the institution in question was on board and willing to help. You do not necessarily need to know every last number but you need to be able to call it up, ask, for example, “Where did the revenue go in September?” and get someone to give you a general answer. The Irish council does not have a direct relationship with the institution that deals with taxation. However, we have a relationship with the people in the Department of Finance who do, and they have given the council enough of an idea about specific companies that are having a big impact on the aggregate level. We do not need to know the identities of those companies or exactly why they are doing what; all we need is a sense of the impact of that sort of thing on the budget and the economy.

An interesting point for you is that a lot of the data that we expect to be available when we are dealing with a particular country might not be available to Scotland and, if you set up a new body, you will have to think about the data that might be missing and which would be needed to do the kind of forecasting that you are looking for.

Other witnesses have raised that issue. I certainly think that we will need to consider it.

Mr Barnes has partly answered my follow-up question by talking about the Irish experience. Do you know of any other independent fiscal institution that is a member of your network that does not have a direct relationship with the body in their country with responsibility for taxation and accruing revenue for the national exchequer?

No, I think that they all have a direct relationship, although it might take place at different levels. In most cases, the legislation in question gives them a right of access but, in some cases, that legislation might be a bit open to interpretation and Governments or different agencies will use that to slow down the process of giving information. Nevertheless, access to that information is a standard thing.

Thank you very much.

I point out that the Scottish Parliament is a national Parliament in a multinational state, not a regional Parliament.

I made a mistake in saying that Jean Urquhart would speak next when in fact it was John Mason who asked to do so. My eyesight is deteriorating by the day, it seems.

10:00

I return to Ms von Trapp’s query about what the problem is that we are trying to solve. I am not sure whether we have a problem to solve. Is it basically correct that, because Governments have generally been overoptimistic in their forecasts, we need such a body to come in and bring a bit of reality to the situation? Theoretically, if a Government or a Parliament were a bit more realistic, conservative or prudent, we would not need such a body.

I will make one comment before passing over to Sebastian Barnes, who may be able to provide a better answer to your question. The argument that we often hear from New Zealand is that it is so transparent that it does not need such a body, but that is open to debate.

If you look across OECD countries over a long period, you see that public finances over the past 30 or 40 years have been on a poor trend. Most countries had very little public debt in the early 1970s, but a debt of 100 per cent of gross domestic product is now not untypical. You can argue about whether 100 per cent is an okay level, but it is clear that that rise in debt was not a deliberate policy; rather, it is a result of weak control of public finances. Some countries are an exception to that, but not many. In parallel, many countries have taken on massive pension liabilities that are essentially kept off the balance sheet.

It is against that background that the debate about the need for a fiscal council comes in. I often think that it is a little like the monetary policy debate. In the 1970s, inflation was out of control and there were all sorts of other problems. We then had a big effort to make central banks independent to clarify their mandate. Although the outcome of that effort has many imperfections that we are all unfortunately paying the price for today, in some ways a lot of progress has been made. However, debt has risen—the amount may differ between countries but there are a variety of problems.

The problem that you mentioned is real for many countries. In the UK, there is a perception that there was overoptimism and that it was a problem; indeed, there are many other countries around Europe that are overoptimistic, too. However, that is not the only reason to create a fiscal council. One set of reasons—again, this probably does not apply in your case—is that the Governments in some countries cannot manage the public finances. For example, a Government might write a budget at the beginning of the year but spend a totally different amount by the end of the year—in that sense, things are out of control. A fiscal council could be a way to control that situation in part.

There is a much broader case for having a council. I am familiar with the position in Ireland. In some ways, Ireland ran things pretty prudently—it was about the only country in Europe that consistently ran a budget surplus during the early 2000s. The Government seemed to be doing prudent things, but in truth it did not manage the risks carefully enough. There are probably things in the banking sector that a fiscal council would not have been able to do much about, but it is pretty clear on the fiscal side that Ireland was taking risks that it should not have been taking. Analysis of important issues was lacking, so the only way in which it could find out how big the impact of the crisis would be was for the crisis to happen.

The public finances in Ireland over that period could have been better managed, and a council could have contributed to that by its analysis and its recommendations to the Government and by helping to explain to people what the risks were. As I say, things seemed to be going fine, but there were big vulnerabilities underneath.

Thank you for that very helpful explanation.

Let us continue with the Irish example. If the fiscal council had been in place, even though the political parties and the Government were seemingly quite prudent and running a surplus, it might have pointed out that the surplus should have been much higher, with greater savings or whatever for the future. Presumably the role of a council would be to give such a forecast or comment. All the political parties might have decided that advice was going overboard and chosen to ignore it, but the information would still have been there. Is that your argument?

I think that the argument is that the council would, we hope, have done that analysis and made that point. Of course, people might have dismissed it, but if the council was effective, its analysis was convincing and it had a mandate from the Government, it might have been hard not to listen to it.

The Department of Finance in Ireland would make the point, rightly or wrongly, that it was warning ministers but that those warnings were ignored. In essence, however, those warnings were made and the dynamic played out in private. If the warnings had been made in public by a body that had an explicit mandate and much deeper analytical resources focused on the issue, who knows what would have happened? The hope is that the two mechanisms of the mandate and the independence, along with the credibility of the analysis, would have made it more difficult for politicians to ignore such arguments.

In Scotland, we are debating whether we are a region, a state, a nation or whatever, but at the moment we are, let us say, a substate with fairly limited powers, although we are soon getting more powers over a bit of income tax and one or two other taxes. Is this the right time for us to set up such a body, or should we wait until we have more powers? In reality, we do not have a huge amount of flexibility—we cannot get hugely into debt or other such problems.

Under our principle of local ownership, that is for you to decide. It is a difficult balance and you are in an unusual situation. The logic that you and Lisa von Trapp set out is to consider what problem you are trying to fix. In a sense, if there is a system that is already running, the problems are more obvious. In Scotland, you are taking new powers and will be in a new position, so it is less clear exactly what you should do—and I am not sure that I have particularly strong views on that.

Another lesson from small countries is that they can be very volatile. Typically, the economy is dependent on far fewer activities than a country the size of the UK. Therefore, in that context, the risks are extremely important. A fiscal council will not necessarily tell you what to do in the light of those risks, but it can help to build a framework, to go through the arguments and to protect yourself against the kind of thing that happened in Ireland. That might be one thing that you will want to consider. I am not sure how that will play through in your actual and prospective institutional framework, but that is something that a council can bring.

It was mentioned that, in the Scottish context, we perhaps do not have the information available—at least not publicly available—to fully analyse the Scottish economy and to make full forecasts. Would a fiscal body be a boost to the production of such information because it would demand it?

Yes, I think that that is the case. With respect to parliamentarians, they have many other important activities, but a council would be focused on the one issue. In Ireland, a lot of information is available that would not otherwise be in the public domain and which was previously not even available to the Department of Finance.

It is powerful to have a body whose role is to think about fiscal issues full time. Obviously, it is costly to collect information but, in general, we tend not to have enough of it and we tend to take decisions a little in the dark. Councils can be helpful in getting information collected and in disseminating and interpreting it.

That is great; thank you.

You have given a pretty strong statement about the need for independent financial institutions, particularly in relation to controlling public debt and managing risks. However, your written submission states that councils are “not a panacea” and that

“Alternative institutional arrangements may serve some countries equally well.”

That is perhaps slightly more open-minded than what you said in your previous set of answers. Did you have particular countries in mind when you made that statement about alternative arrangements, or was it rather a theoretical statement?

In the part of the OECD that I work in, we make recommendations to individual countries about a number of things, which are basically peer reviewed by OECD members, and putting in place a fiscal council or an independent fiscal institution of some sort is a pretty standard recommendation in almost all the work that we do. There might be exceptions to that, and perhaps Lisa von Trapp knows what they are. I think that that comment in our submission is an exception clause.

It is more of an exception clause, but we were also pointing to concerns of some of our member countries that do not have one institution that fulfils all the roles that we might think of as being for a fiscal council but which have several institutions that each takes a piece of the roles. For example, Germany has quite a few independent institutions that do different pieces of work and which would probably form a strong fiscal council if they were put together.

You say that the creation of the OBR is brave, which puts a slightly different slant on Jamie Hepburn’s suggestion. Your submission says that

“IFIs face real risks when they produce critical analysis, particularly in their early years.”

Do you have examples of that, or is that a general statement?

Councils have faced different initial risks. One risk is that they are ignored. That does not sound dramatic, but it is quite dramatic for a new council that is trying to establish both itself and its credibility to be ignored, particularly by the Government.

Perhaps the best key example comes from Hungary—I do not know whether the committee has looked at it. The situation resulted partly from the political system there. The fiscal council was set up with full support across the political spectrum but, once it began to produce critical analysis, the Government reacted strongly. The Government’s majority in Parliament was so large that it was able to cut the council’s budget to—I think—€50,000 and dismiss all its staff.

That is the most extreme case, in which the Government shut down the institution, but there are other cases. The Swedish institution underwent a period in which its budget was under threat. The Canadian body’s budget was under threat; that played out in the press in a difficult way between the council and Government representatives. There are risks of threats to budgets in particular.

A Government might fall out with a council if the council had a remit to explore alternative fiscal policies. How common is that in such institutions throughout the world?

Could you clarify what you mean?

I mean alternative policy prescriptions and working out the consequences of following alternative economic policies. Does any institution take on such a role?

Some institutions take on such a role, normally from a positive analysis—I hesitated earlier because I was going to ask about that. Some bodies, such as the Congressional Budget Office, look at options. They do not necessarily say, “These are the Government’s policies and we think you should do something else,” but rather, “Here is a list of options and what their effects would be.” They do not give a normative analysis or tell a Government to do something differently; they try to show the legislature—particularly when they are set up to serve it—the range of options and the potential consequences. That is one way in which that role is done.

That certainly seems to be a useful function of such a body. Perhaps that would be more likely to happen if the body was directly accountable to the Parliament rather than the Government.

There is the well-known case of the Netherlands institution, which looks at alternative policies that political parties propose at election times—that is known as election costing. That has been discussed in the UK context. The exercise is different, but it fits into such work in some ways.

We are in an unusual situation in Ireland, as part of our mandate is very open—we just have to assess the fiscal stance. In that context, we have shown alternative scenarios that we have constructed to the one that the Government is pursuing. That is done for information and to allow people to understand the position.

People have different views about things. We show what taking a different view would look like. That is a useful function for councils to pursue.

We have made the point that we highly recommend that councils do not give normative policy advice, so that they are not drawn into political debates or politicised in that way.

10:15

As you have raised that example, and as it is a live issue in the UK, I will ask: how common is it for Opposition parties to have their policies analysed and costed by a fiscal body?

It is very common for most of the institutions under the legislature—the parliamentary budget offices. There are institutions such as the Congressional Budget Office that are required to cost every piece of legislation. That involves a lot of resources, of course, and it is fairly unusual. In the Netherlands, there are similar arrangements.

Only Australia and Netherlands cost election platforms. It is an interesting idea, which many people find useful. There probably needs to be a period of adjustment as political parties have to start thinking about the cost of platforms. I do not think that most parties do that in a significant way. The process in the Netherlands is an interesting one.

There are countries where pieces of proposed legislation are costed, although there is a limit to their materiality or contribution potential. In Canada, there is a small body with 15 staff, and it cannot cost every piece of legislation that the Parliament might want it to cost. It sets a standard and asks whether the proposed legislation is likely to have a big impact on Canada’s public finances; if it is, it will probably take it on.

I think that, in the Slovak Republic, the costings of specific pieces of proposed legislation may be examined at the request of Parliament, but the body there will not examine everything.

Okay; thank you.

Annex 1 to the paper that you have given us has a list of OECD countries with independent fiscal institutions. There are 17 countries listed, and I have just noticed that about 50 per cent of them had an independent fiscal institution prior to the banking collapse and recession in 2008; about 50 per cent of them have been established since then. Did countries that had independent fiscal institutions before the collapse have slightly better recessions than those that did not? That is a very general question, but is there any link? Is there anything that might show that having an institution cushioned the downturn slightly?

That is a very good question. It is always difficult to assess: the countries that had those institutions may be different from the other countries in other ways. Sweden is a good example of an economy that appears to be pretty well managed, because it took the lessons from the nasty crisis that it had in the 1990s. That applies in a number of respects. It shows up in the way in which it regulated its banks and in various aspects of how the Swedish system works. It shows up in the fact that it created a fiscal council. There was a relatively good outcome in Sweden, and it was probably decided that, on many levels, the council would comprise good people—and they took a lot of good decisions.

Although the body in Australia was created more recently, the country has quite a strong tradition in this area. It is the office itself that is new. The story is similar there: Australia has had a relatively good crisis, partly through good luck—no doubt—but also because the country has done a number of things and has exercised good fiscal management.

It is hard to answer the question, given the coincidence between good institutions—good fiscal councils—and good outcomes across a range of factors. The Netherlands is always cited as an example, and it is an interesting one. At times, things in the Netherlands have been well run over the past 40 or 50 years; at other times, they have been badly run. The CPB—the central planning bureau—has always been there throughout. What happens partly depends on the times, on the culture and on a number of other things.

There are good reasons to hope that fiscal councils will help in such situations, but the evidence is hard to evaluate neutrally. We can see good examples and bad examples over history. If you look at countries such as Sweden, Australia or Canada, however, at some level that is the way that you should probably be running your country and your institutions. Those countries all have fiscal councils.

Thank you. I have a more specific question. Principle 2.3 of the OECD principles talks about “term lengths” and “number of terms” for the leadership roles within a body. Do you have views on the term length and the number of terms that would be appropriate or optimal? Are there trends or things that you think are optimal?

In a lot of countries, term lengths closely match the election cycle or are a bit longer or staggered so that they do not run directly alongside the election cycle. That is partly to enhance the independence of the bodies. We recommend that term lengths should not be linked to the election cycles but should be slightly longer and perhaps renewable only once rather than continuously. It is hard to say because different countries have different ideas on the issue. De-linking term lengths from the electoral cycle is important, however, as is being clear about what the term length is.

The one country in which that might be an issue is Korea, where there is no specific term length but, in practice, the head of the national assembly budget office resigns every time that there is a new speaker. That means that they resign every two years. They are not forced to resign, but they do that in practice and it creates an odd situation whereby they may be seen as being too linked to the speaker, which politicises the role a bit.

In Ireland, we have staggered four-year terms that are renewable once. In the context, I think that that is sensible.

It is quite difficult to find people to sit on the council. If the positions are part time, it can be hard to find people who can combine them with whatever else they do. Also, there are not a huge number of people around who have experience in public finance, economics and interacting in these processes. Therefore, you do not want to make the term length too short; otherwise you will struggle to find people to sit on the council.

Institutional memory is valuable as well, particularly in a small institution. If an institution has hundreds of staff members, the chief executive officer can change very quickly, but if the council is a big part of the intellectual capital of the institution you need continuity and the council is important in providing that.

Okay. I have a final question. You mentioned principle 3.1, which states:

“The mandate of IFIs should be clearly defined in higher-level legislation”.

Does pretty much everything to do with the body have to be defined in higher-level legislation, or can some of it be left to secondary legislation? How much needs to be in primary legislation as opposed to regulation or follow-up?

My view is that the principles need to be in the higher-level legislation to offer protection to the council, which is what you want. My general view is that you do not want stuff to be too prescriptive beyond what is necessary to achieve the objectives. The whole idea is to set a high threshold, and you may regret including detail at the start if things change, so I do not think that it should be there. The primary legislation should be relatively clean on that; the council can then define how it operates in a more detailed and transparent way.

In Ireland, there is also a memorandum of understanding that covers some of the specific issues around forecasting. We have used that MOU once and we are about to revise it, which is a good case for having things in an MOU rather than in legislation. If we did not have that, we would be stuck.

That is helpful. Thank you.

In response to questions from my colleague Malcolm Chisholm, you touched on an IFI’s role in recommending what fiscal policy should be followed. The committee has taken evidence on that and explored the pros and cons of an IFI taking on the normative role of recommending the course that any Government should take. One of the main reasons that has been cited against that role is that it would politicise the IFI. Is that the main reason why you recommend that IFIs should not take on such a normative role, or are there other reasons?

The issue is partly about the division of labour. It is your job as politicians to decide what should be done; a council’s strength is in making a positive analysis of different options, which we have discussed, or of the forecast that presents the best view of the risks. The right division of labour in a democratic system is one question.

Politicisation is a risk. The credibility and the degree of buy-in for councils are important. If a council appears to be backing one side or another, it could lose those precious things. The hesitation comes from that.

In Ireland, we go a little further, as we have to assess the fiscal stance. That does not explicitly ask for a normative view, but it is closer to that. It might be useful for the committee to know that we frame a lot of decisions by having the notion that a range of policies is appropriate and that the council’s role is to give a warning if policies are inappropriate. For example, if the deficit target in a year should be one number or another, the council is fine with that, as reasonable people can reasonably disagree about that. However, if policies seemed to be crazily counterproductive, the council would have a normative role in saying, “You can pursue this policy if you wish, but it is not appropriate.”

We have stated some preferences. We have said that what the Government is doing is fine but that there is a strong case for doing some things a bit differently. We have always said that in a much more muted way, because we are less comfortable with such a role. We always have a framework of recognising that lots of views are appropriate. Our role is to give a warning if the car is going off the road.

That is helpful.

For a lot of councils, the issue is difficult. Even when they do only positive analysis, they sometimes find that they walk a bit of a fine line. Just pointing out a measure’s impact on the public finances can show that it might not be a good idea. A council does not have to say that something is not a good idea; it can show just with the numbers that a country is probably going into risky territory. Most institutions walk that fine line, whether or not they give normative advice, which almost none of them give.

That certainly adds to the evidence that we have taken on the subject.

We have exhausted the questions from committee members. Does Ms von Trapp or Mr Barnes wish to make any further points?

We have done case studies on different countries that are not all currently publicly available. If the committee was interested in reading about a country in more depth, I would be happy to share its case study with the committee. I think that the functions—but not necessarily the set-up, which is under the central bank—of the Slovak Republic’s institution might interest the committee as an example in addition to Sweden and Ireland. That institution was quite influenced by what is done in Sweden.

That offer is helpful. I thank Lisa von Trapp and Sebastian Barnes very much for answering our questions.

I end the videoconference. We will have a five-minute break in which to reorganise for our next session.

10:29 Meeting suspended.

10:39 On resuming—

Our next item of business is to continue to take evidence as part of our inquiry into the proposals for an independent fiscal body. I welcome someone who is no stranger to the committee, Professor David Bell, and his colleagues Dr Angus Armstrong, Dr Katerina Lisenkova and Professor Peter McGregor. I invite Professor Bell to make an opening statement on the panel’s behalf.

The people on the panel are all supported in one way or another by the Economic and Social Research Council’s UK and Scotland research programme and are carrying out a number of projects that are specifically focused on the Scottish economy. The projects will add significantly to the technical capability to analyse the Scottish economy, irrespective of what happens in the referendum.

Four types of model are represented here. I work with David Comerford and David Eiser on a microsimulation model; Peter McGregor works on a computable general equilibrium model; Angus Armstrong and Katerina Lisenkova work on a macroeconometric model, which is the closest thing to the kind of model that the OBR or the Treasury might use to forecast the UK as a whole; and Katerina also works on an overlapping generations model, which is more of a long-run perspective of the Scottish economy. I will not go into the technical details of those models, but we will be happy to answer questions about them should they arise. Our argument is that the models could provide different perspectives on the Scottish economy, of the kind that a future Scottish fiscal body might be interested in.

The first issue that the Parliament knows that it must address is taxes associated with the Scotland Act 2012. The OBR is providing forecasts for them and the question is to what extent it is necessary to provide alternatives to those forecasts. We might explore that in discussion.

One of the key factors that the Parliament will want to have a view on is the size of the block grant adjustment and how that relates to the way that the Scottish rate of income tax operates. Our microsimulation model looks in detail at how income tax is raised across the spectrum of households in Scottish society.

There is a variety of reasons why one might question the accuracy of the OBR forecast, which we might well want to explore in discussions. Dr Armstrong, who had a role in the establishment of the OBR, might be able to help the committee with that specifically.

I overheard in your discussion with your contributors from the OECD that you talked about some of the issues to do with the way in which the body will be set up. I do not want to go into those issues specifically, but we would be happy to discuss them.

An important issue is to whom the independent fiscal body, once it comes into existence, should be responsible. My view is that it should be responsible to Parliament and that its first point of contact should be the Finance Committee. Even within the existing budgetary set-up, it seems that there are useful things that an independent fiscal body could do, whether or not it does forecasts for the Scottish economy as a whole. The environmental impact statement, the equalities statement and, say, a statement on the distributional impact of the budget are all kinds of things that to some extent are done by the Scottish Government but are very much below the radar of the headlines that the budget gets. It seems to me that close examination of such issues could garner greater interest in them if that is carried out by an external body.

I am happy to take any questions.

10:45

Thank you very much.

As David Bell will know, how the committee works is that I will ask some introductory questions and then open out the session to colleagues around the table. When I ask a question, it will not necessarily be specifically for an individual; each or one member of the panel can answer it, or whatever. Whoever wants to make a contribution should do so, please.

The third paragraph of David Bell’s submission says:

“The forecasting record of the economics profession in general is not good ... We need to guard against seeing a forecasting body as a panacea for the uncertainties surrounding public finances.”

How do we guard against that? I think that the issue is sometimes feedback loops. We talked about that last week to an extent with Jim Cuthbert. If such an organisation is established, it can be put on a pedestal such that its forecasts feed back into the markets. What is your view on that and on where the balance should be struck? As I said, your colleagues can contribute as well.

Point forecasts, as we call them, can ruin the reputation of any forecasting body. Your reputation can be ruined if you hang it on a specific number without realising that there are various probabilities associated with different outcomes. Perhaps members have seen that kind of thing reflected in the famous Bank of England fan diagrams, which will broadly say that there is a 20 per cent chance of a growth rate of 1.6 per cent, for example. When the actual outcome goes outside the limits of a fan chart, that is really damaging to a forecasting body’s reputation, and that is what happened with the catastrophic events of 2008.

If an economy is going along and sticking with trend and output is not vastly over or below its natural rate, the forecaster’s job is quite easy, but forecasters come under pressure when there are big turning points. They have not been terribly good at predicting major turning points. Very few people foresaw what happened in 2008. We have learned from that experience, but will the next experience be quite different, and will there therefore be another need to change the ways in which we do our forecasting? One needs to have healthy scepticism about forecasts. It seems to me that such a body would have to protect its reputation by not hanging too much on specific numbers.

Does anybody want to add to that?

Yes, if I may. Good morning, ladies and gentlemen.

Forecasting is one of the more misunderstood roles of a fiscal council because of the headlines and what the convener suggested about the attention that it can have in the markets, for example. To assume that a small group of people has any more knowledge about a headline figure such as the GDP figure than the market collectively has is to put a weight on those people that is probably greater than there deserves to be and to stretch their bounds of knowledge.

The real public good side of the forecast is the detail. The taxation breakdowns that the OBR provides in its forecasts are provided by no financial institution and very few other bodies. That allows people to do all the estimates and see the degree to which the economic forecasts feed into the fiscal rules and whether it looks as if they will be achieved. It is the detail, not the GDP number per se, that is the most important part.

Inside the forecast, it is what we call—rather unfortunately—unobservables that are important. They are things such as what the trend level of output is. That is crucial for determining whether the fiscal rules are being hit. Because it is unobservable, reasonable people can disagree. That is where the real value added in the forecast is; it is not in the headline GDP numbers. In the OBR’s early days, it was an extremely difficult forecasting period. If a forecast is away from the consensus, by human nature it can take a little time to get back to where the consensus is, so being away from the consensus is almost a disadvantage. I think that the OBR’s headline forecasts are now highly consensual. That is probably quite deliberate. It does not presume that it has that sort of knowledge; nor is that the public good side of what it delivers.

I think that a specific forecast can be a real issue for a Government—or, indeed, an Opposition. A couple of weeks ago, we saw the publicity that the IFS’s projection of Scotland’s performance 50 years into the future attracted. Such forecasts can be taken almost as gospel by commentators and the media and, for politicians, they reinforce specific views. I think that that is a danger, is it not? How do we get the balance right? We have to have an independent fiscal body that is credible and whose opinion people respect, but we do not want the amount of weight to be attached to its forecasts that we have seen happen in the recent past.

It is important to distinguish between forecasts and projections. I think that the IFS report was widely misunderstood in that it was a projection—in other words, it was based on things not changing, in policy terms. The committee discussed with the OECD people whether the new fiscal body should influence policy directly. I think that the IFS tried to say what might happen, under its best assumptions, if there was no policy intervention. That is quite different from saying, “In a couple of years’ time, the growth rate will be 1.8 per cent.” Different models address different kinds of issues. It is quite difficult to get that message across to those who are interested, and it is extremely difficult to get it across to the general public.

I add that it will be the responsibility of the independent fiscal body to be careful to conduct appropriate sensitivity analysis around any central projections to reflect alternative possible scenarios in the future. The IFS did that; whether it did enough of that is open to discussion. Uncertainty can be handled in a variety of ways through such approaches, and I think that it would be incumbent on the body to reflect that in one way or another.

David, in your submission, you talk about what a separate forecasting body would add. You say:

“The OBR may choose not to produce forecasts specifically for Scotland and so differences between the Scottish and UK economies may not be fully captured.”

Will you expand on that a wee bit and say how such a body would assist?

I also have a question for all the witnesses: do you believe that we should establish such a body?

I think that the OBR currently takes the opinion that the construction of a full-scale Scottish model is not feasible, given the data availability. The Scottish Government has had under way for some time a project called SNAP—the Scottish national accounts project. That is building up a quite extensive time-series database of aggregates such as consumption, investments and savings, in as good a way as possible given the data constraints that it is confronted with.

The National Institute of Economic and Social Research has a large number of macroeconomic models for countries around the world—Angus Armstrong and Katerina Lisenkova can speak about that much better than I can. It produces a UK model each quarter. In the next few months, with funding from the ESRC, using the data that is available from SNAP, the institute will produce the best macroeconomic model that it will be feasible to produce. That could offer quite a different perspective, because of the different methodology. It might produce different numbers from the ones that emerge from the approach of the OBR, which I do not think is that sophisticated in relation to the forecasts that are produced for tax revenues in Scotland.

On the question of whether we should establish an independent fiscal body, the answer is probably yes, regardless of the degree of fiscal autonomy. However, perhaps the nature and scale of the investment ought to vary with the degree of fiscal autonomy. Clearly, if you are going all the way to independence, it would be an absolute requirement. With the degree of fiscal autonomy that is currently enjoyed by the Scottish Government under the Scotland Act 2012, there is still a requirement for such a body to model tax revenues.

As David Bell suggested earlier, some of the wider functions that such a body could have could be extremely useful on issues of equality, integrating environmental impacts of policies with the economic modelling and so on. It is possible, under some modelling approaches, to integrate sectoral demands for energy, for example, because emissions vary dramatically across industries with the composition of output as well as the total output and, if you are going to consider them, you will need to have some kind of multi-sector modelling approach to capture them. I believe that a lot of value could be added by such an approach.

On the question of whether an independent fiscal body should be established in Scotland, I would say, on balance, yes. However, that raises a question of the costs and the benefits of that body. You will gain access to a certain type of information that is currently not available, at a level of detail that the OBR is not able to provide. However, you must make a difficult decision based on how much it will cost the taxpayer to get that level of disaggregation.

As David Bell just said, we have a wide range of different models that have been built for Scotland, and it is clear that the level of data that is available at the moment makes it possible to do these kinds of projections from various angles, looking at the short run, the long run, the distributional aspects, the behavioural responses and so on.

The work that NIESR has done to expand our global macroeconomic model to include Scotland showed me that the progress with regard to data availability in the past year has been dramatic. Data that was not available a year ago is available now. I suspect that that trend will not stop, so I want to praise the efforts of the Scottish Government’s statistical office, which is doing a great job. It is becoming easier to get the data for the kind of projection work that we are doing.

11:00

The decision on whether to have a fiscal council is a tough call. That is not a cop-out; I am trying to say that there is, under the current constitutional arrangements, a balanced budget. A fiscal council would exist to provide credibility for discretionary changes in aggregate fiscal judgments. That is the big economic issue, and people who have given evidence to the committee have explained how it works. A fiscal council is therefore not really necessary at present, and it would cost money. As we all know, everyone has budget constraints just now, so I would be wary of creating another body for which the remit is not clear.

There is also the interesting question of how a fiscal council in Scotland would interact with the OBR. If there were a different forecast, the implication would be that the forecast for the rest of the UK was the remainder, which may or may not be consistent. Credibility in terms of consistency and transparency is very important for such bodies. As I said, I do not think that one can out-forecast the markets’ forecasts, but one can provide real credibility and show exactly how the forecast is made up.

Unless there was clear and close engagement between both fiscal councils, the situation might start to erode some of the gains that have come from having the OBR. It is important to work out what the relationship between a Scottish fiscal council and the OBR would be. That is not to say that the requirements cannot be met, but it is an important question.

There are a lot of important non-forecast issues—which my colleague has mentioned—such as inequality and distribution. Fiscal councils were not set up in the first place to address such issues—that was not their rationale—but those questions are very important, and I do not think that the OBR, with its current resources, could be expected to address them for Scotland.

Perhaps there is a role for an evolutionary process in that regard. We could start with a fiscal council that has a slightly different mandate from normal fiscal councils, and if there were further constitutional change, the remit could be amended over time. Initially, the body would not look like a normal fiscal council. Given that there is basically a balanced budget, I do not see what would be the return for it, given the cost.

We have a list of 17 countries; the costs of such councils in most countries, including France and Ireland—but excluding the Netherlands, the council of which has a huge budget and covers a lot of ground—are around the €700,000 or €800,000 a year mark, although Slovenia’s council costs €100,000.

One reason why the Scottish Government has considered the idea is that for devolved taxes such as the landfill tax and the land and buildings transaction tax, the OBR was just extrapolating from UK figures that did not bear much resemblance to the figures that we had here. There was a concern that that would impact significantly on the budget that Scotland ended up with on a year-to-year basis.

David Bell mentioned the need, in establishing a fiscal body, to establish

“a critical mass of independent economic analysis”

and expertise

“in Scotland.”

I do not want to hog the meeting, because my colleagues want to come in, so I will just ask one more question on a subject that panel members have already raised. Paragraph 12 of the submission from the Scottish centre on constitutional change states:

“There is a strong argument that the fiscal body should be tasked with analysing the distributional and environmental impact of budgetary decisions.”

I am not aware of whether the bodies in the 17 countries on which we have information perform that task. Is it routinely carried out by those councils? Perhaps panel members can talk us through that.

Fiscal bodies carry out a variety of tasks in different countries. Some just focus on the macroeconomic issues. The Treasury does a distributional analysis along with, I think, a gender or equal opportunities analysis of the effects of its budget.

The reason why I highlighted those points was based on the view that such bodies are fairly idiosyncratic, so there is no blueprint into which they all fit. Having looked at many budgets, my feeling is that environmental impact, for example, is an important topic but it does not receive much attention. That important topic would perhaps fare better if it were the subject of external scrutiny rather than scrutiny from within Government. The classic example on distributional analysis is the Institute for Fiscal Studies, which is not an independent fiscal body but which everybody treats almost as though it is. Its post-budget analysis of who has gained and who has lost is treated with a great deal of seriousness—much more so than the UK Treasury’s analysis, which is done in much more detail and much more thoroughly.

There is no single blueprint. It seems to me that Parliament could pick up on each of the issues, because it has committees that are set up to look into them. That would spread the impact of an independent fiscal body beyond the Finance Committee and into other areas of the Parliament.

It is important to understand the full implications of the issues that the convener mentions, such as the landfill tax and stamp duty, but that is not really the problem that fiscal councils were originally thought to be trying to solve. That problem is the credibility issues that Governments have because they are seen always to promise something and then to offset it.

It is a bit unfortunate that people use the term “watchdog”, because it has a slightly negative connotation, but the OBR has played a valuable role. Such bodies can be powerful in enhancing the credibility of Government policy—whether or not people agree with it—by giving it independent validity. Those are macroeconomic judgments.

It is extremely important to examine the distributional consequences and tax-raising potential of individual taxes, but that requires a slightly different type of body from the classic fiscal council. That is why my answer to the question whether Scotland needs a fiscal council was not ambivalent, but set out the points on the one hand and on the other. There are good grounds for having such a body, but I am not sure what its remit would be. David Bell referred to the IFS, which raises important issues about the distributional consequences. Its independence gives it a credibility that a Government would, to be frank, find impossible to replicate. I agree that the reports should always be made to Parliament.

I will add one point that I wanted to make earlier. Proposals for things such as fiscal councils and independent central banks often build up a head of steam and are held up to be a panacea. When economists are unanimous about something, we should be slightly sceptical and think that they might have missed something. We saw what happened with independent central banks—the outcome was not so terrific and we nearly blew up the system. Likewise, with fiscal commissions, a lot of Government activity is moving off balance sheet. That is an inevitable consequence, because it is what humans do—they find ways round things. One needs to think about the broader governance structure as well as just the narrow remit of the council.

Referring to your starting point, convener, and to the concerns that have been expressed about the projected tax revenues, we need some way of assessing the likely system-wide economic impact of Scotland-specific taxes on revenues. Tax revenues depend on everything else; they depend on the impact on economic activity. If you are going to focus on Scotland-specific tax policies, you need Scotland-specific economic models in order to explore the issues. I do not see how you can proceed without doing that.

The issue has been that the OBR did not take Scottish Government policy into account. On the landfill tax—to take a simple example—the OBR was predicting £107 million of income three or four years from now, whereas the Scottish Government was saying that, because of the zero waste policy, it would be £40 million. That means that there could be a significant difference when settling the Scottish budget regarding the allocation of resources to Scotland. Therefore, more accurate forecasting is now being considered, which will take into account Scottish conditions and policies.

I do not want to hog the meeting for too long.

I have a comment about the results that the OBR produced that did not coincide with what the Scottish Government was saying. Any forecast or projection involves a learning process. Time needs to be allowed for the methodology to mature. The OBR had to carry out a new task, and it invested a certain amount of time and money in it. The next forecast will not necessarily produce a vast difference. Allowance must be made for development of methodology. To focus on the first forecast is not a very wise approach.

On the question of cost, you should consider whether the OBR will, in the future, be able to address the issues that you want the independent fiscal body in Scotland to address. If Scotland stays within the UK—we will find out soon—does that body necessarily have to be in Scotland?

The key to understanding whether the Scottish budget is going to undershoot or overshoot is the block grant adjustment and how it works in relation to the Scottish rate of income tax. If there is a no vote and a change in the Barnett formula so that, somehow or other, a different formulation is determined for allocating funding across the different parts of the UK, there is a strong case for having some capacity already in place in Scotland so that the implications of different funding arrangements—not necessarily forecasts—and of different methodologies that might replace Barnett can be explored before they are finalised.

Regardless of the referendum result, the issue of the block grant adjustment was of real concern for the Finance Committee with regard to devolved taxes.

Professor Bell’s written submission is pretty strident about

“the difficulty in producing accurate economic forecasts”.

The convener has already picked up on that point. Your submission goes on to suggest that

“given an OBR which produces forecasts that are accepted as independent forecasts of the UK economy, a low cost solution for the Scottish Parliament may be simply to adjust the OBR forecasts as required for the Scottish Government’s budget process.”

Robert Chote was pretty clear when he told us that the OBR cannot provide the data that he expects the Scottish Government would want. Indeed, there has been discussion about the implications thereof. I am therefore wondering how serious your suggestion is. I suppose that it could mean one of two things. Is the suggestion that the OBR adjust its forecast to suit our circumstances or is the suggestion that a new body be set up and all that it does is look at the OBR forecasts?

11:15

That picks up Katerina’s point that the issue is about how much effort the OBR will put into enhancing the quality of the Scottish forecast. That is difficult to predict. However, we are developing alternatives anyway. The Economic and Social Research Council is funding alternatives, so within the year, there will be alternative ways of projecting the three taxes and looking at their distributional impact. Those alternatives will be available without any extra expenditure on the part of the Parliament or the Scottish Government. The question is whether those forecasts are maintained.

How serious was the suggestion that you made? Did you throw it into the paper as a possibility rather than a serious prospect?

If one felt that the OBR would devote lots of resources and come up with forecasts on which one could not improve—that is not necessarily to say that those forecasts would be right—one might ask what the benefit is at the margin from doing additional work. We have suggested that there are benefits from doing additional work, but that is a judgment that the Government and Parliament will have to make.

I agree—but as I said, Robert Chote was pretty clear on that point. To be fair to him, he was making the point that the OBR is not established to carry out that work. It is not necessarily the case that the OBR is saying that it does not want to do it; rather, it is saying that it is just not its job.

You mentioned the on-going work on different approaches to economic modelling, which is interesting. I want to explore that a little in the context of the submission from the Scottish centre on constitutional change. It was interesting how that matter was set out in the paper. Perhaps I am reading too much into it, but I thought that you were drawing a contrast between how the OBR works and its reliance on HMRC data, and that you were positing the work that you are doing, and have just mentioned, as an alternative.

Robert Chote was clear; he said it is vital that any Scottish independent fiscal body have access to HMRC data. Do you agree or are you saying that the different modelling work that you are doing makes that less of an issue and that any body could be less reliant on HMRC data? If the alternative modelling is available, would it be important to such a body?

Let me quite nerdy for a little bit.

Just as long as you are not going to talk about the computable general equilibrium model too much.

Our modelling work looks at different effects of changes in, for example, jobseekers allowance in families in Scotland. We look at how changes in jobseekers allowance affect people in the bottom 10 or 20 per cent and so on. We call that a distributional impact. We use a data set that is available from the Department for Work and Pensions to work on that. The data start to fall down on higher incomes because they are not recorded all that accurately. We therefore use HMRC data for that. For example, the increased inequality that has happened in the past few years has been concentrated right at the top end of income distribution—the 99th and 100th percentiles—which contains the richest people and who have left the rest of income distribution behind. The data from our regular survey do not capture that effect, but we can find it if we go to the HMRC data, which are based on people’s tax returns. The OBR uses exactly the same data to analyse, say, the effects of the Scottish rate of income tax, but we are supplementing it with that other data set, which has much more information on the benefits that people get from the state and so on.

Broadly, we have access to the same data as the OBR, but at the moment, at least, we have the resources to put it together in a more comprehensive way that does not just focus specifically on the three taxes.

For the purposes of absolute clarity, are you saying that it is important that such a body has access to HMRC data but that it should look at it and use it in a different way from the OBR?

It should supplement the data. There are lots of other places where one can access data but, to return to what Katerina Lisenkova said, I note that it is a resource question. It depends how much resource the OBR is willing to put in. If it does not put in much, it will not have time to explore all the other data sets and use them to enhance whatever forecast or piece of information it is trying to provide for Scotland.

It is interesting that you state in the paper:

“Staff should be hired on the basis only of their professional competence.”

Call me naive, but I would have thought that that would go without saying. However, you said it. Are you aware of any issues that led you to feel that you needed to say that?

Perhaps I could have put that a little better, but I really think that the body should be at arm’s length from the Scottish Government. That should be pretty clear. I do not know whether Angus Armstrong wants to say anything about the way in which the OBR was established.

There was an interim OBR that was located within Her Majesty’s Treasury. Location is vital. It seems bizarre that human relations change so much depending on whether people are down the corridor or in a separate building, but they seem to. I believe that it helps if people are in a different building. It is much better if the people and the institution are in place, as far as they can be, before they are tasked with doing the work, which inevitably becomes political. There is no getting away from the fact that doing tax distribution consequences gets you in the newspapers.

It would be beneficial to have all the appointments done beforehand, and it would be an improvement on how the OBR was set up, which was in very different circumstances in the couple of months between the election in May 2010 and the emergency budget in July. It was an extremely fast process as there was not much time. You have time to take a more considered approach, so the appointment process, the location and other obvious but important issues can be taken more fully into account.

Thank you.

I would like to go back to Professor Bell’s comment on the impact of the budget, looking at the environment, inequality, employment and so on. So far, nobody has mentioned the national performance framework, on which we have also taken evidence. There have been pretty positive comments about the NPF, and I wonder how you see it fitting in with the work of an independent body of the type that we are talking about.

I must admit that consideration of that is an omission from the paper.

It seems to me that there is a danger that the national performance framework goes off into an area by itself and does not receive the attention that perhaps it deserves. I do not want to argue for an all-singing, all-dancing, do-everything type of body, but the national performance framework is a coherent way of understanding what the Scottish Government is trying to do, and not many Governments have that kind of structure. That being the case, there is certainly an argument for some way of analysing it.

The framework is not made up of only economic arguments and those that are not economic would clearly not be relevant here, but there are broad economic issues that are relevant. For example, should a fiscal body try to call the Government to account on aspects of the national performance framework in which performance was less than optimal, or should it inform committees and politicians and just put things out there, rather than getting dragged into the political process itself? It is nevertheless not difficult to imagine a fiscal body at least making a commentary on where economic performance is in relation to the Government’s stated objectives, and perhaps just leaving it at that and letting politicians take the debate further.

It seems to me that relating the budget to the national performance framework makes it much more interesting from a public and parliamentary point of view, because otherwise we are just talking about money, I might argue.

You could argue that.

I hope that, when we are writing the remit for that group of people, we can consider agreeing to make it different to the OBR’s remit, which seems to be pretty limited.

Absolutely. What we are talking about in the first instance, I guess, is the mechanical forecasts associated with the taxes that are currently devolved. I have suggested that, if you were thinking of preparation for other constitutional arrangements or other frameworks of fiscal distribution other than the Barnett formula, you might want to build in some capacity to analyse and comment on those. I mentioned in the paper the environmental, distributional and equalities impacts, all of which relate to different parts of the national performance framework, so you could use the national performance framework, in a sense, to link to some aspects of the independent fiscal body’s work. However, as Angus Armstrong said, it is more like constructing something that would be able to help the Parliament in the exercise of its duties, rather than what is conventionally thought of as a fiscal body that is there to restrain politicians from making decisions that could be fiscally dangerous. I am not making a judgment between the two, but that is essentially the argument.

I would like to hear again about the IFS report and the notion that we might predict or observe a nation’s wealth 50 years hence. A hundred years ago, who would have predicted two world wars and the discovery of oil? Other extraordinary things have happened that we would not have expected any public body or group of observers or professionals to predict just 50 years ago.

That is a good point. As I said, the projection is based on a number of important assumptions and goes forward a long way. The one thing that we can forecast a long way ahead is the likely age structure of the population, so that was one of the primary drivers of the IFS forecast. Why was it one of the primary drivers? It assumed that higher costs of healthcare and social care are associated with ageing.

11:30

The projection was also associated with long-term forecasts of North Sea oil revenues, which are contestable, and with an assumption that the productivity of the Scottish economy will continue to grow at an annual rate of 2.2 per cent right through that period. That is contingent on all kinds of assumptions: that technology will improve, that we will grow our capital stock and that each worker will become significantly more productive. All those assumptions are necessary to get the projection that the IFS produced and you could argue with every one of them. What the IFS has produced is the basis for a discussion, but we know at least that its forecast in relation to demography will probably be fulfilled.

I understand that no prediction that has been made of the Scottish population so far has been correct, so it is difficult to say that that prediction will be different. There are also so many other debates around how the population might change as to make the forecast less acceptable.

You are correct about that. A fundamental assumption of the population projections is that migration is somehow a fixed number that does not change, irrespective of what happens to the economy, but that is demonstrably not correct. Migration depends on what happens to the economy and may well be sensitive to the particular policies that we pursue.

Is that a criticism of the IFS? Only in so far as we might ask whether the achievement of higher economic growth by a set of policies could get us out of the bind that the IFS has identified. I think that the answer to that question is yes—it would be surprising if it were otherwise. Economic growth can help us to get out of the situation. There is no rule that says that, as the economy grows, we will continue to spend the same proportion of gross national product on Government expenditure.

David Bell is absolutely right that the projections are based on a set of assumptions that may not hold true. If policies to change things radically were available and worked, those projections would not be realised.

It is almost inevitable that the projections will not be correct, as they predict the outcome on the basis of current policies—they do not say that that will be the outcome. I assume that if the policies are not sustainable they will change, although I am not going to second guess how they will change. The IFS is answering a very narrow question; the trouble is that it becomes political.

That takes us back to the narrow remit that the OBR has. One of the reasons why the OBR has such a narrow remit is to protect it, as it gets pulled in all directions. It gets asked questions that would pull it straight into the political environment, where it would lose its credibility. Having such a narrow remit and a very narrow set of questions that are answered in a certain way will not be particularly informative, but it is not supposed to be; after all, they are answering the question, “Based on this or that assumption, will this target be hit or not?” That is not to say, of course, that those assumptions will not be kept or will not change.

Moreover, having a narrow remit means that omission is not taken to mean acceptance. You can say, “Our mandate is not to answer those specific questions, so we are not approving or disapproving anything.” That is more powerful than a more open mandate, with which people can say, “Well, they didn’t say this was bad for us, so we’ll take that as acceptance.” It is useful to have a narrow mandate for these things.

With regard to the IFS report and indeed any projection that looks 50 years ahead, I will be holding a seminar in Edinburgh on the effect of population ageing and what will happen 50 years from now, and we are coming to somewhat different conclusions to the IFS on this matter. That is not to say that the IFS’s forecast is wrong and ours is right; all I am saying is that if you take a different perspective you can get very different results. Any forecast that goes so far ahead should be the beginning not the end of the discussion. Longer-run forecasts in particular should not be put on a pedestal and seen as some point forecast in which one can say, “This is the number that we are going to have.” Instead, one should take the forecast and say, “This is one of the possibilities of what might happen if A, B, C, D, E, F and G are correct.” Any of those assumptions can be wrong, and the forecast itself starts the discussion rather than finishes it. As with any OBR or IFS forecast, you cannot pick a number and say, “This is what they say is going to happen”; that is simply not how either organisation intends its figures to be used.

Can that kind of prediction become quite dangerous and affect the reputation of, say, the people who will be selected as members of this advisory body?

It depends on how the prediction is used. A body’s reputation should be based on its efforts to improve its methodology, make its results as robust as possible and look at the wide spectrum of the sensitivity analysis of its results, not on whether it has managed to hit one particular number that the media picked up and put on the front page of the newspaper.

I have found the discussion over the last little while to be really interesting. In fact, what really interests me is the extent to which you can set up a body that is not politicised either in perception or in reality. Presumably, the intention is to set up a body that is neutral in perception and, as far as possible, in reality, and one of the interesting suggestions you have made is the exploration of alternative fiscal policies. In principle, it seems possible to do that in a relatively neutral way; for example, one of your groups has analysed the impact of the Scottish-specific rate of income tax, and I presume that it is possible to do that without taking a view on whether it is a good or bad thing.

That said, two comments in your submission led me to question just how possible it is to be neutral. First, on the national institute global econometric model—or NiGEM, as I believe it is called—you say:

“It is essentially New-Keynesian in its approach”.

Obviously that is contestable, but would such an approach result in completely different conclusions from any alternative approach?

You also say that fiscal institutions

“are also required for the much needed support for introducing policies which may be unpopular but necessary”,

which suggests that they might have a role in nudging Governments towards certain policy positions that otherwise they might not find very palatable. Given that, how possible is it for such a body to be politically neutral?

I will start, and Dr Armstrong can add to my answer. NiGEM is developed by the national institute, and I am working specifically on developing a Scottish model within that framework. When I said that the model is neo-Keynesian, I was making a methodological comment so that people who know about different types of model—those who are sufficiently geeky—would know to which particular class this model belongs and what type of assumptions it takes into account.

Just to prove that I was not talking about the political perspective of the team of people who are working on the model, I should say that I have, while working on the neo-Keynesian NiGEM, also been working on the neoclassical overlapping generations model, and I am completely ambivalent about which one is correct. I am saying that we have to take different perspectives as that will give us a broader picture of the range of possible outcomes.

Any model that looks sufficiently ahead must be built on assumptions; there is no way around that. We cannot abstract our personal beliefs in making those assumptions, as we are human, but we can at least attempt to be non-partisan in choosing specific assumptions to produce a specific outcome.

NiGEM has evolved since it was first developed in the 1980s, and covers 60 countries and regions around the world. Describing it as neo-Keynesian makes it sounds a little bit ideological, but it is actually very mainstream. Its real skill, or advantage, is that it is big. It is very conventional, but it has some interesting options in terms of solving the whole model for the UK in particular, which is the most important country for us in forecasting because most of our user base is here. The model is very detailed in terms of fiscal accounts, so we present a very detailed forecast. It is not like using a normal macro model, as we work line by line on fiscal accounts. That is the model’s real advantage.

To go back to the beginning, the model is not so much about point forecasts, which is where we start to get into trouble again. People want economists to behave as if they are some sort of magicians, but they are not. What economists can do is show you the breakdown and how it feeds through, which other people do not show you. The OBR does that very well. It gets a pretty consensual forecast and gives you all the details, so that everyone can understand and poke around with it, and see what might be right and what might be wrong, and where they agree or disagree or whatever.

One of the real strengths of a model such as NiGEM is that it is very traditional and very complete. We have been working on the Scottish version for a number of months. It started off as far less detailed than the UK model for obvious reasons: we do not have the clear fiscal data that is available for the UK.

The idea is that we start off with a model that we can use for simulations, and build in more and more detail as the data becomes available. You cannot assume that you will suddenly have a massive model that will allow you to do everything, because it will not. I do not think that any model could do that at present with the current data sets, although those are clearly improving.

I will pick up on something that the convener said. The idea of creating a critical mass of macroeconomic capability is very important, because there are a lot of Scottish-specific questions to answer about the macroeconomy, fiscal distribution and so on. That would be a very important addition to the body, whatever it looks like, and would put some weight behind it, I hope. It is important to have that in Scotland.

11:45

I want to return to the corporation tax example. I was involved with that with colleagues at the Fraser of Allander Institute. It was a computable general equilibrium model, but I promise not to dwell on any of its details. It is an example of how it is possible to consider with such models, once developed, the impact of policies that may not yet be available to the Scottish Government, but could conceivably be available under some alternative constitutional arrangement, which is very important. Once you have such models, you can at least begin to explore the possible consequences of moving from the Barnett formula to some other arrangement with a greater degree of fiscal autonomy. For example, we have looked at the possible impact of a Scottish-specific carbon tax. Such models add to the range of information that you have to inform decision-making processes.

I should mention that the corporation tax example was a very specific look at potentially adjusting corporation tax while maintaining the existing devolution settlement—a renegotiation of the tax, rather than a negotiation of the powers around it. Possible retaliation by the UK Government was not taken into account in that case. I should also say that that was a collaborative venture with economists in the Office of the Chief Economic Adviser, who have been discussing and developing with us their own CGE modelling capacity so that they can explore the possible impact of further constitutional change.

It seems that Malcolm Chisholm’s question whether the body will enter politically difficult territory comes right back to the question of what the body is for. Is it to inform and enhance our understanding of the operation of the Scottish economy or is it to ensure that the Scottish Government does not go too far off track? If it is the latter, it will get into the political fray, no doubt about it—it would be impossible for it to do otherwise. Part of the question is who will stand behind it if it chooses to do that. To some extent, that must depend on how it is constituted, who the members of its board are, with what kind of authority it comes to its judgments and who is there to back it.

That was really interesting. I am interested in the exploration of alternative policies, which you seem keen on. I suppose that the question might be at whose instruction that is done. That seems quite important. If the body was proactive in doing such things it could be seen as promoting its own policy agenda. I think that it would be desirable for the body to be available to opposition parties or the Parliament in general. That would be a useful service with which to explore alternative policies.

I very tentatively suggested that the social partners might be involved in the board, which would be unusual in the UK these days. That might be a way of ensuring that if the body was commissioned to look at fiscal alternatives, that would be done by including a wide range of social and economic perspectives.

Thanks very much.

The Scottish centre on constitutional change says:

“The possibility of the fiscal body being independent of both Parliament and the Executive should not be dismissed without question.”

Do panel members have a view on the pros and cons of following that model?

The examples that I know of in other countries have grown up like Topsy, to some extent, such as the German council of economic advisers—I do not know whether it is called that, but it is certainly a group of wise men. Their views on the prospects for the German economy are treated with great respect. That group was not specifically set up with a particular role and I wonder whether it is possible to set something up outside Parliament and the Scottish Government that would instantly command that level of respect. A body can have the most wonderful set of models, forecasts and projections, but if it does not have some reputation then the extent to which those things will be paid attention will be very limited. That would certainly be detrimental.

It entirely depends on the body’s remit. If the remit is in line with that of a typical fiscal council, the body would have to be statutory and probably accountable to the Parliament. The IFS has shown that if the body has more of an IFS remit, which is not like that of a normal fiscal council, it can gain that kind of credibility while not reporting to Government. The IFS does not do macroforecasts; it just does the distributional side.

The answer depends on the remit that is created for the body and its governance structure. As always, the detail of the governance structure is what really matters.

The Scottish centre on constitutional change also says:

“The governing board should be selected ... for a period of no less than four years, thus extending beyond the length of a Parliament.”

How many more years than four should the governing board be selected for? Four years is broadly the same length as a parliamentary session, although the current parliamentary session is five years.

There should be a way of rotating members of the panel and probably no one should sit for more than two full parliamentary sessions. I overheard the OECD witnesses talk about this issue. It is not all that easy to find people with the requisite expertise. You certainly need competent people on the board and you might consider having people from outside Scotland. A period of two parliamentary sessions seems a reasonable safeguard against domination by a particular view.

That concludes questions from committee members. Thanks very much for your evidence, which is much appreciated and will have a major influence on our report.

11:53 Meeting continued in private until 12:32.