Official Report: Minutes of Evidence

The Chairperson (Dr Aiken): The Committee will now have its second oral evidence session of today on its consideration of the options for an independent fiscal council. I welcome, via StarLeaf, David Phillips. Hi, David.

Mr David Phillips (Institute for Fiscal Studies): Good afternoon.

The Chairperson (Dr Aiken): Excellent. The session is being recorded by Hansard. David, I invite you to make your opening statement, please.

Mr Phillips: Yes. The work of the Institute for Fiscal Studies (IFS) tends to focus more on fiscal rules and the fiscal outlook as opposed to fiscal institutions, but many of the recommendations that we made prior to the setting-up of the Office for Budget Responsibility (OBR) on the independence of forecasting and other kinds of data that need to be provided to show the judgements and assumptions made and the credibility of those judgements ended up being taken up by the OBR, so we have some experience in the field.

In today's session, I am happy to talk about my and our views on issues to do with the operation and remit of a fiscal council. I was also asked whether I could provide some information about business rates and business rate reform. I can happily talk about that as well.

The Chairperson (Dr Aiken): Delighted.

Please continue.

Mr Phillips: I was waiting for some questions. I have not prepared a speech in advance. Usually, when I do these things for the Scottish Parliament, the Welsh Assembly etc, there is a list of questions. I had not been told to prepare a speech in advance. I am sorry if that was expected.

The Chairperson (Dr Aiken): That is fine. Are you ready for some questions?

Mr Phillips: Yes.

The Chairperson (Dr Aiken): What benefit would a new fiscal body bring to the governance of Northern Ireland Executive finances? Germane to that is this: Northern Ireland is a relatively small jurisdiction, with very limited tax-varying powers, so how could we assess and report on the sustainability of the Executive's finances? Should that be delivered by a sub-office of the Northern Ireland Audit Office (NIAO) or by a new research body? What do you think?

Mr Phillips: The greater the extent of demand-led spending and the greater the extent of revenues that are devolved to a territory, the greater the importance of an independent fiscal council. Those are typically the sorts of things for which, rather than working to a fixed budget in advance, the amount of spending will depend on, for example, the number of claimants of the benefits, while the revenues will depend on the performance of the economy and the tax bases. Those things are uncertain. In order to set Budgets, you need to produce forecasts, and you want those forecasts to be credible and independent. The first thing that I will say is that the resourcing that a fiscal council would need will very much depend on the range of powers and responsibilities that the Northern Ireland Assembly has for demand-led welfare spending and taxes.

As it stands, there are relatively few responsibilities in that area. Welfare spending is formally devolved, but the way in which it is devolved here is quite different from the Scottish model for its element of welfare. In Scotland, when welfare was devolved, the responsibility for funding that welfare was devolved as well, so, even if policy remains the same, the Scottish Government will bear the risks of costs going up or down more or less quickly than Governments in the rest of the UK. Northern Ireland funding for welfare is not like that. Northern Ireland bears the cost of policy deviations, but, for a common set of policies, the UK Treasury provides the amount that it costs to do that. There is no risk in the Budget if the number of claimants goes up or down in a different way from numbers the rest of the UK. Although there are quite a lot of devolved powers on the welfare side in principle, the funding responsibilities are not there.

The only real areas in which the fiscal councils in the UK — the OBR and the Scottish Fiscal Commission (SFC) — are involved are around the costs of any deviations from welfare policy and how those might vary according to claimant numbers. On the revenue side, there are the business rates and regional domestic rates. In a UK context at least, the traditional remits of those fiscal councils concern relatively small amounts. That would probably suggest having a relatively small and focused body for Northern Ireland.

The terms of reference are slightly different from, for example, the terms of reference for the Scottish Fiscal Commission. The terms of reference that are being set out for a fiscal council therefore relate very much not only to the annual assessment of the Executive's revenue streams but to sustainability and the implications of spending policy and efficiency measures. Getting to the bottom of what that will mean exactly could be important. For example, at the moment, the Scottish Fiscal Commission does not do any sort of longer-term assessment of fiscal sustainability, spending outlook or revenues outlook. It is very much a medium-term forecaster. The OBR does a long-term fiscal sustainability report that looks at the outlook for health spending, pension spending, revenues and so on. I think that it goes out 50 years. It is not clear to me whether that suggests having something along those lines for the elements that are the responsibility of the Northern Ireland Executive, such as health and social care, education and potentially any devolved taxes, or whether the thinking is about doing something in the shorter term more about the medium-term sustainability of the plans. The discussion about the effectiveness of long-term efficiency measures suggests to me something shorter than the 50-year horizon that the OBR has when it looks at long-term sustainability.

It is an interesting question about the extent to which an independent institution that is involved in forecasting and long-term fiscal sustainability analysis should also be assessing the effectiveness of efficiency measures or whether that is something more for an audit office to do. For example, the National Audit Office (NAO) might do in the context of UK Departments.

The Chairperson (Dr Aiken): OK. Thanks.

Mr Catney: I will stay on that point. The Budget process here often runs late owing to delays. People can then blame that on the Westminster Government. In the event of the Westminster Government or the Executive failing to provide Budget information, might an independent fiscal council produce a kind of shadow Budget that informs the Executive of decisions that are outstanding?

Mr Phillips: That is a really interesting question. This year's Scottish Budget was published at the very end of January, which was before the UK Budget, although after the autumn statement and the 2020 spending round. The SFC made its own forecasts of Scottish revenues, even though those would potentially be affected by decisions on the income tax personal allowance, for example. It said, "We are going to assume that the income tax personal allowance follows standard indexation policy". In the end, it was announced that, for subsequent years, there would be a freeze, but the SFC said that it could subsequently update its forecasts to take account of that.

Potentially more significant than what the Scottish Fiscal Commission did was what the Scottish Government did. When the Scottish Government set their Budget, they assumed that there would be £500 million of additional Barnett formula consequentials announced in the Budget for 2021-22. When they set their departmental budgets, they built those in, despite their not being yet announced. That would have been a problem if the Barnett consequentials had not then been announced, because of the balanced Budget rules within which devolved Governments operate. The combination of what the SFC did in its forecast based on default assumptions, which can be updated subsequently, and what the Scottish Government did by making their assumptions, which were then assessed by the SFC as to their reasonableness or not, could provide a guide to how the Northern Ireland Executive and any Northern Irish fiscal council could operate in such circumstances. I should add that one thing that will help deal with those circumstances is the financial flexibilities around borrowing. For example, if tax forecasts or block grant adjustment forecasts were to change once a Budget is set, there would be the ability to borrow to offset those rather than having to make in-year cuts. That could be also an important aspect of dealing with such situations.

Mr Catney: Thanks.

Mr O'Toole: Thanks for coming to give us evidence, David. You talked about fiscal forecasting. Can I ask about your views on economic forecasting? The OBR has a role under the Budget Responsibility and National Audit Act 2011 to do an economic and fiscal outlook (EFO), which is an economic forecast and a fiscal forecast. Do you think that it is important for economic forecasting to be integrated with fiscal forecasting?

Mr Phillips: To the extent that economic forecasts are determinants of the fiscal forecast, you will need to use some economic forecasts to determine what you expect to happen to, at the moment, business rates and domestic rates, as well as potentially further tax devolution or welfare devolution. You therefore need to have economic forecasts. The question of whether you need to have separate forecasts is an interesting one. I will explain in a second what you need to do on both the economic and fiscal side of things, but one option would be to use the OBR's economic forecasts and plug them into Northern Ireland's fiscal forecasts. That would mean that a common set of assumptions was being used around the performance of the economy.

On the one hand, the view might be that Northern Ireland-specific forecasts for the economy as well as for public finances could take account of specific developments in the Northern Irish economy that the OBR perhaps would not have done. More bespoke models could be built to take account of the economic variables that are most important for the taxes that are devolved. The OBR would not spend much time forecasting some of the smaller taxes. That might push you towards wanting to have separate forecasts. You need to have forecasts, and having your own might mean that they are more bespoke and pick out more specific things.

On the other hand, one of the things that we see in Scotland is that it is not just when there are differences in economic performance between Scotland and the rest of the UK that you can get divergences in revenues and a need for reconciliations, if, say, the Scottish Government get a bit more money than expected or pay back some money that had previously been forecast that they were going to receive. It is not just differences in performance that has driven that but differences in the economic and fiscal forecasts. If you have two forecasters making different assumptions about what will happen to the economy and to revenues, that adds another element of uncertainty where things may differ. In general, you would expect that, if the OBR forecasts a strong economy, other forecasters probably will, including Northern Irish forecasters. They will not all perfectly align, however. Having two different forecasters, with one doing the Northern Irish economic and fiscal forecasts and the OBR doing the UK ones that will matter for any block grant adjustments that operate in this new fiscal world, can result in more volatility.

That would mean a need for more fiscal flexibility around borrowing, reserves or other things in order to address that.

There is a bit of a trade-off here. Do you think that having a bespoke forecast for Northern Ireland would be significantly better and that it is therefore worth taking on that additional risk of volatility, or do you think that it would be better to minimise the risk of volatility, even if the forecasts are a little bit out as a result of having one common set of forecasts?

Mr O'Toole: I am not sure that I completely follow you on the trade-off. Are you saying that our new fiscal council having stand-alone responsibility for producing a bespoke economic forecast for Northern Ireland would build in a degree of volatility, if it were mandated to —.

Mr Phillips: Yes. Let —.

Mr O'Toole: Sorry. Go on ahead.

Mr Phillips: Let me explain it. Let us assume that, if there is further fiscal devolution, as you know, it is likely that some additional tax revenues will be devolved to Northern Ireland, and there would then be a block grant adjustment. Some money would be taken off the block grant to reflect the tax revenue that had been devolved. In Scotland, for example, the Scottish Fiscal Commission does the forecasts for the tax revenues that are devolved, but it is the OBR forecasts that matter for determining the block grant adjustments and how they change over time.

The economy might evolve in exactly the same way in Scotland as it does in the rest of the UK but if, at the start of the year, the SFC is potentially a bit more bullish about the economy than the OBR, the SFC will forecast stronger growth in revenues, while the OBR will forecast that the equivalent revenues in the rest of the UK will not grow as strongly. The block grant adjustment will therefore not grow as much. Initially, in that scenario, the Scottish Government say, "The tax revenues are going to go up more than the block grant adjustment. We have got this extra money to spend". Come the end of the year, when they have finally calculated things, it turns out that, no, there was no different performance, only different expectations and different forecasts. Later on, that money that they thought that they had is an artefact of one forecast being more bullish than the other. Down the line, they will have to pay that money back to the UK Government, because, in the end, revenues did not grow more quickly in Scotland.

Mr O'Toole: OK. Although it is —.

Mr Phillips: When you have two different forecasters, you have a greater likelihood of different judgements being made than you do if you have one forecaster doing everything. Having two forecasters make different judgements on different sides of the Budget — the revenue side and the block grant adjustment side — adds a bit more noise and volatility, and you need borrowing powers and reserved powers in order to deal with that. If you are happy to do that, having the more bespoke forecasts might be considered to be a good thing. If, however, the idea of potentially having those errors and, down the line, having to pay money back is concerning, you could go down the Welsh route, which is to have the OBR do both sides of the forecasting: the revenue economic forecasts and the block grant forecasts. Scotland and Wales chose different routes.

Mr O'Toole: You may be aware that, for block grant adjustments, we have one of the most egregious examples in the UK. Owing to a forecast done, I presume, by the OBR a decade ago, we are paying out £2·5 million a year for a theoretically reduced long-haul air passenger duty rate. If you are not aware of it, David, please go away and do some work on it, because it is one of the most pathetic examples of forecasting around a bit of fiscal devolution.

There are various examples in the UK of fiscal advisory bodies. The other broad model, which does not exist in the UK, is the Congressional Budget Office (CBO). It has a much more expansive role as a highly specialised, non-partisan, expert body. At a time when US politics is hyper-partisan and very divided, that is a treasured position. I am interested in getting your thoughts on the CBO model and how it compares, favourably or otherwise, with the various UK models.

Mr Phillips: The Organisation for Economic Co-operation and Development (OECD) probably talked about this in its evidence session. There are two broad models of independent fiscal institution. There is the fiscal council model, such as the OBR, the Scottish Fiscal Commission and many others. Then there is the Parliamentary Budget Office (PBO) model, which tends to have a somewhat different role, focusing less on forecasting, long-run sustainability and costing of government policy and more on supporting the scrutiny and policy development function of Parliaments.

There is nothing stopping you from having a —

Mr O'Toole: Hybrid.

Mr Phillips: — hybrid model or having both of those options. When you are setting up a new body, having a relatively narrow remit, which is clearly apolitical, allows that body to establish itself and gain credibility, especially in a set-up where there is potentially constitutional or political friction around significant areas of policy. Over time, as that credibility is gained, you can think about adding functions that expanding its scope, and, although avoiding policy development and comment on particular policies, you can start to get closer to the more contentious political aspects. For example, the CBO and the Centraal Planbureau (CPB) in the Netherlands look at the policy platforms of different parties and the impacts that those have on distribution, employment and so on. Those things are inherently more controversial, even if they are done in a very impartial, independent and rigorous way.

My view is that there is nothing stopping any part of the UK, or any country at all, from having a hybrid model or having both sorts of models in place. I would start off, however, by being limited to areas in which you can build credibility and reputation. You can then think about expanding over time.

Mr O'Toole: Following on from that, I have one final point about whether you expand that into a Parliamentary Budget Office- or CBO-style thing. I am sure that the high-quality political parties in Northern Ireland would be more than happy to have their extremely high quality manifestos subjected to independent fiscal scrutiny. I am sure that they would add up pretty convincingly [Laughter.]

The Chairperson (Dr Aiken): Sorry. Those were just guffaws in the corner there.

Mr O'Toole: I was going to ask about independence. You talked about the constitutional issue. That is not the constitutional issue that we habitually worry about here in Northern Ireland but the constitutional issue of the institutional friction between a devolved Finance Ministry, which will always have an interest, and the UK-level Finance Ministry. If part of the role of the fiscal council will be to create a repository of independently trusted analysis and information that will smooth and regularise certain types of processes between a devolved Finance Ministry and the UK-level Finance Ministry that doles out the block grant, what should the legal character and independence of that organisation be? There is clearly a risk on both sides. On the one hand, it could look as though it is marking homework on behalf of the UK Government or that it is there to be an overseer of devolved institutions that should be independent; on the other hand, it could look as though it is simply ballast for the devolved Finance Ministry when it is making its argument.

I am interested in your view of how you best ensure that it has independence from both sides, as it were. Primary legislation is obviously one part of that, but I would be interested to hear what you think. Do you think, for example, that the Scottish and Welsh systems have that? Is there any sense that the OBR's doing forecasting — having a fiscal advisory role — in Wales, for example, compromises its independence from Whitehall? Your thoughts on that would be interesting.

Mr Phillips: On that last point, my sense is that the Welsh Government took a practical decision about what they felt would not only be cost-effective but also limit some of the fiscal risks that arise from having two sets of forecasts that might differ so that you need some sort of reconciliation, down the line, between them. It was to save resources — not necessarily financial resources but, potentially, concerns about human resources. There is also this volatility point. A decision was taken in Wales that, at least for a period, the OBR would be a suitable forecaster. For a short while, they were doing internal forecasts which had then been vetted by Bangor University. That was another model that they considered: government forecasts, but with a peer review. Some independent fiscal institutions actually operate on that basis. Rather than producing the official forecasts, they operate as a peer reviewer of government forecasts. That is another model.

In Scotland, the SFC has done a remarkably good job in a very contentious area, but that is partly because of its narrow remit. The remit is extended to look at medium-term forecasts on what will happen to revenues and spending on these demand-led welfare measures, what are the policy costings of all this, and — a pretty soft touch — is the Government's policy on capital and resource borrowing sustainable. If it had been a broader remit, also considering things like the effectiveness of long-term efficiency measures or the medium- to long-term sustainability of the Budget, that would have become more politically contentious. You could see that becoming a battleground between the Opposition parties and the Government around effectiveness issues. You could see the sustainability issue becoming a battleground between the Scottish Government and the UK Government around issues such as how we are going to address the long-term rising costs of health, social care and other issues and, "Oh, look. They're not giving us enough money to pay for these things". Whilst that sort of work is really important, if I were starting off trying to get credibility and independence for a new body, I am not sure that I would immediately jump into those somewhat more controversial areas.

I have answered part of your question. On the second bit about who it reports to and how it avoids being seen as a stooge of either party, part of that is starting off with a focused remit which is as apolitical as possible to try to build that credibility. Part of it is making sure that the body is —. I do not think that the body can be directly accountable to two Parliaments, but it should interact with two Parliaments. It should be willing and able to give evidence to both the UK Treasury Select Committee and other subject Committees and the Northern Ireland Assembly Committees. There should be agreements between the fiscal council and not only Northern Irish institutions, in terms of accessing data, but also UK institutions, where it needs data from UK Departments, like HMRC and so forth. There will need to be memorandums of understanding and so on for a lot of UK Departments as well as for the Northern Ireland Assembly. I also caution that many of the things that you want done here on the effectiveness of long-term efficiency measures and the medium-term sustainability of the Northern Ireland Budget — I am not sure that a fiscal council is the right body to look at those matters, at least in the first instance. Instead, the Northern Ireland Audit Office or equivalent, or the Northern Ireland Executive themselves, could look at those. The Scottish Government have a financial strategy that is moving in the direction of medium-term sustainability and fiscal risks. That becomes somewhat more political, so perhaps having some of this done by the Government is the correct —.

The Chairperson (Dr Aiken): Thanks, David. I am conscious that time is running on and we have two other people to come in.

Mr Allister: I have a practical question, if you can give us a brief answer. The last year has been unusual, but the various Treasury handouts and the COVID Barnett consequentials resulted in splurges of expenditure, particularly towards the end of the financial year. The imperative seemed to be to get the money spent, without too much scrutiny of the return. What difference would a fiscal council have made?

Mr Phillips: In that context, the fiscal council would have no role in determining the UK Government's funding for the Northern Ireland Executive, so it would not affect —.

Mr Allister: No, I am interested in the Executive's spending of the money, not the gifting of it to them.

Mr Phillips: Right. In general, a fiscal council would not comment on the use of funding for particular policies, as that starts to impinge on policy and gets it into more political grounds. The useful role of a fiscal council is less in spending control or in giving a critique of spending; that is more a role for an audit-office-type body or a Parliament, actually. A fiscal council can play a useful role in making sure that there is a public understanding of what is happening in the funding situation. One thing that the Scottish Fiscal Commission did very well was to put out information about how much funding the Scottish Government had, when it was received, how much was being carried forward into the next year and how it was being allocated. A fiscal council can serve a role as an information body, both for the public and for the Parliaments, but that does not really address that particular concern.

I will add that your concern is probably lesser in the devolved countries, given that the Treasury gave additional flexibility to the devolved Governments to carry forward funding that did not exist in England. For example, health spending had a substantially smaller splurge in the devolved regions, particularly towards the end of the year, because some of the Budget could be carried forward, and that could not happen in England, where it was lost. The short answer is that a fiscal council would not solve that problem, but it could play an information role.

Mr Allister: OK. Thank you.

Mr McHugh: Just quickly, I know that it appeared to be funny earlier on, but would it not actually be a good thing if the fiscal council had a role in the costing of party political manifestos and the like? Parties would be forced to use accurate information when talking about the policies that they intend to implement in the event of being elected.

Mr Phillips: There can be real benefits from that approach. You are right; it is important that parties are subject to scrutiny on their plans, both in terms of overall affordability and in terms of the impacts that they say they will achieve on distribution or the economy and so on and so forth. I can definitely see the benefits of that. The prime example of that is the CPB in the Netherlands. Since the 1980s, they have been —. It is all on a voluntary basis, I should say. Basically all the parties now submit their manifestos to CPB, which produces a report that looks at the impact on revenue, spending, jobs, distribution and so on and so forth.

There can be some drawbacks to that approach. The first one comes back to the point that, when an institution is new and working in a potentially divisive landscape, politically or otherwise, actually getting into that area can sap some of its political capital. That can be a risk. The second issue is that there needs to be a very, very clear understanding of the uncertainties that are involved in some of this. I work at the IFS, which does some of this type of analysis of UK policies, and I am not sure that we always get across, as well as we could, the uncertainties about some of this stuff. Using the CPB model — the Dutch model — a manifesto will come out, and the CPB will analyse it and put a figure of 20,000 jobs on it. Another manifesto might have 25,000 jobs. However, in economic analysis, the range of uncertainty with those figures is quite a lot greater than the 5,000 difference in jobs. The point is that, if a body went down the route of doing comprehensive analyses of party manifestos, it would need to be clear about the uncertainties and the limits of its knowledge. Maybe it could start off by looking at a smaller set of things, like costs and basic distribution analysis, and then, over time, as it embedded and built its expertise and understanding about these issues, start to do more complicated analyses, making sure that it explains the uncertainties. There can be — as an economist, maybe I should not say this — almost too much focus on some of the quantitative estimates when actually some of this stuff is the subject of quite wide uncertainty and confidence intervals. That is something to bear in mind.

The Chairperson (Dr Aiken): Thanks very much for your evidence, David. Thank you for coming along; it was informative and very useful.

Mr Phillips: Thanks very much.