Oral evidence: Proposals for further Fiscal and Economic Devolution to Scotland, HC 760Tuesday 28 October 2014

Oral evidence, 28 October 2014.

Witnesses

Paul Johnson, Director, Institute for Fiscal Studies, and David Phillips, Senior Research Economist, Institute for Fiscal Studies, gave evidence.

Chair

Thank you very much for coming to give evidence. It is very familiar for the Committee to have the IFS in front of them. Mr Johnson, do you think anything has been ruled out of the discussions on devolution that it would have been wise to include?

Paul Johnson

I am not entirely familiar with what has been ruled out, but I presume you are referring to some extent to the Barnett formula. There are clearly issues about the degree to which the Barnett formula redistributes between England the rest of the UK in principle, but there are also quite a lot of detailed issues with the way that the Barnett formula works, for example with respect to the extent to which changes in the relative importance of business rates can make a difference of hundreds of millions of pounds with respect to how much income Scotland receives. I am not sure exactly what some of these commitments mean in terms of the principles, or whether they refer to the detailed operation of things like the Barnett formula, but certainly the detailed operation feels like something that should be open to consideration at the very least.

Chair

You think Barnett, or at least the administration of Barnett, should be looked at in tandem with the fiscal process?

Paul Johnson

In the end that is a political decision about the amount of money that the rest of the UK wants to see going to Scotland and how they achieve it over time.

Chair

You are discussing administration rather than overall amounts?

Paul Johnson

Detailed elements of the way it works, which I think inadvertently make quite a big difference to the actual results of money that flows between the constituent nations.

Chair

Lord Strathclyde, among others, has suggested there needs to be some kind of quadripartite commission of all the nations to consider the long-term future of Barnett. Given what you have just said, is that the sort of approach that you think might be appropriate?

Paul Johnson

I think the important thing is probably to spend appropriate time thinking about getting it right. In the end we are still living with the consequences of the Goschen formula of 1888, made in haste at that time. I think it would be a shame if in 100 years’ time we were living with consequences of choices made very hastily now that turn out not to achieve what you want to over the long run. I think being pretty clear that we get things sorted in a way that works for a long period is going to be quite important.

Chair

Have you done any of the number-crunching on the effects that fiscal devolution will have on assuaging English concerns about Barnett?

Paul Johnson

There is not much number-crunching that is possible to do at the moment, given the lack of information about what might happen. We will be putting out something in a couple of weeks looking in some detail at the impact of the way that the Barnett formula works, but more broadly looking at the whole issue of fiscal devolution, there is no adequately-defined question for us to answer.

Chair

But are you looking at what type of fiscal devolution can do most to assuage English concerns over what is seen as excess spending?

Paul Johnson

We are not looking at answering the question like that, no.

Chair

It strikes me that it is one that the IFS could run some options on or some sensitivity analysis on.

David Phillips

Devolving taxation variance to Scotland does not address the level of funding that Scotland gets, because what you do is work out how much revenues are coming from that tax in the first year you devolve it—for instance with income tax under the Scotland Act about £4 billion, and you reduce the block grant by £4 billion a year. Then there is a question about how we uprate that block grant reduction in years going forward, and the way they are doing it for income tax under the Scotland Act is that that looks like it will be going up in line with what happens to income tax revenues in the rest of the United Kingdom. If Scottish revenues go up by more than that, the Scottish Government gets to gain money. If they go up by less, the Scottish Government loses out. What you will notice there is that the actual devolution of tax does not change. If Scottish tax revenues just perform in line with UK tax revenues, devolving taxes does not affect the Scottish Government’s budget at all. The process of devolving taxes to Scotland need not lead to Scotland getting less money than it gets now relative to England. In fact, if the Scottish economy does better, it could get more money than it does now relative to England.

Chair

You have answered both parts of the question. In a sense you have given the riposte to your own point because clearly it is possible to think of quite sensible ways of devising fiscal devolution, isn’t it, which will reduce the net contributions through Barnett?

David Phillips

You could devolve a tax that has a slowly growing tax base and, therefore, slowly growing revenues and decide to increase the block grant reduction more rapidly than that. For instance, if you had a tax like fuel duty, which in the long run is probably on the decline, but you decided to increase the block grant reduction over time in line with the GDP, say, that would lead to a reduction in the variance going to Scotland. Similarly, if you were to devolve oil revenues to Scotland and have a block grant reduction that does not take account of the fact that oil revenues are going to fall in the long run, that will also put a squeeze on the Scottish Government’s budget and reduce this relatively high level of spending in Scotland. That is not the way in which taxes are being devolved so far under the Scotland Act, and I think it would be quite difficult in terms of negotiations to get the Scottish Government to agree to a way of devolving taxes that leads to a long-term squeeze on its budget.

Chair

You describe it as a squeeze on its budget. It could equally be described as a more equitable distribution of spending.

David Phillips

That is for politicians to decide.

Paul Johnson

That is for politicians to work out, I think.

Jesse Norman

I am just enjoying the irony that “a squeeze on the budget” is not a political remark but “more equitable distribution” is a political remark, you are implying. As you just implied, Mr Phillips, the tax settlement and the funding settlement are logically distinct entities. It is possible to have one tax settlement and one funding settlement and there is no necessary relationship between the two. In devolving tax you are not, ipso facto, saying anything about the budgetary settlement. It is a further step then to assume that the distribution of cash from the block grant would be adjusted. Is that right?

David Phillips

How I would put it is that the choice is about how you adjust the block grant or the funding settlement when you devolve taxes, but the way in which you do adjust the block grant or the funding regime is very intimately linked to the aims and objectives or what you are trying to get out of tax devolution—for instance, what type of risks you want the Scottish Government to be bearing in its budget. Let us imagine that we devolve taxes in the way they have done so far for income tax under the Scotland Act. The way they have decided to make the block grant reduction is to reduce the block grant in years ahead in line with what happens to tax revenues in the rest of the UK. That means that if revenues go up by more in Scotland, they gain, and if they go up by less, they lose. That means Scotland is bearing what is called the relative risk of its tax take. If its tax take grows relative to England, it benefits. If it falls relative to England, it loses. Now, we can think of other ways in which Scotland bears more of the risk. Let us say that, rather than uprating this block grant adjustment in line with tax revenues in England, it was uprated in line with prices or GDP or just potential GDP—the long run rate of growth of the economy. In that case Scotland would bear additional risks due to the fluctuations in income tax over the economic cycle or the risk maybe income tax was on a long-term decline or long-term increase, depending on what is happening to the economy, fiscal drag or more tax avoidance. I think there needs to be careful consideration about how the block grant adjustment and the funding regimes interact with the tax powers, because that has big implications for just what risk it is bearing and what type of tax devolution and what type of responsibility you are giving to the Scottish Government.

Jesse Norman

That is extremely interesting. What you are saying is, yes, there is no formal link between the two; it is essentially a political decision as to how the two are tied together. There were a series of assumptions, either explicit or inexplicit or shared or not shared, as to the terms under which more devolution was promised and as to the way it would interact with the block grant. By implication, there is all kinds of potential for free rides or for unattractive, unexpected outcomes if you get the interplay between the two wrong, because you could be incentivising the wrong kind of behaviour or behaviour that might give someone a one-way bet on future outcomes. I hope I am not misrepresenting what you are saying. That seems to be what you are saying. Can you just give me a sense then as to whether you think any one of these alternative risk-sharing arrangements would be economically preferable to the United Kingdom as a whole?

David Phillips

I will address it thinking about Scotland and the UK’s position. The answer to that question depends on the extent to which taxes are devolved. Clearly, if you move a system of full tax devolution, where Scotland gets to keep all of its revenues and to decide all tax policy and fund its own spending, in that case you can see the case for Scotland bearing the full risk of its tax revenues. There is no sharing arrangement across the UK, so there should be no risk-sharing arrangements either.

Jesse Norman

And by implication, no block grant.

David Phillips

Potentially no block grant. Although, of course, at the moment with Scotland’s fiscal position as it is, where the tax onus in Scotland is below than that of Government spending in Scotland, as you move towards full fiscal devolution in Scotland you would have a bigger budget deficit in Scotland, which would be difficult to deal with. If, as seems more likely, there is some intermediate form of devolution where some taxes are devolved and some taxes are not, it seems to me that it is more sensible for Scotland to bear only the relative risks of its tax revenues doing less well or more well than the UK. There are several reasons for that. First, doing that gives Scotland the right incentives. If it grows the economy more quickly or makes its tax policies in a way that allows the economy to grow more quickly and tax revenues to grow more quickly, it keeps those because it keeps the relative improvement in its budgetary position in its own budget. However, it does not bear the risks of big recessions that affect the whole of the UK and it does not have to bear the risk that the UK Government makes a policy decision that affects tax revenues in Scotland and the rest of the UK.

Jesse Norman

But that might not be preferable from a UK perspective.

David Phillips

When you have a union, there is a question about how much sharing you think should be going on in the union. One position is that if there is still some element of fiscal union or still some element of equalisation of resources—the Barnett formula is still in place for instance—there should also be equalisation of risk to some extent. A shock that hits the whole UK, not just Scotland, should be borne by the UK Government, which has more scope to borrow on the markets.

Jesse Norman

But there might be a UK concern that the terms of the deal as between the balance of risk allocated and the level of the block grant was sufficiently favourable to Scotland as to be removing a budget constraint that might be beneficial for them, and indirectly for the UK as a whole. That is a possible position, is it not?

David Phillips

It is possible that UK politicians could decide that they think that Scotland should bear some of the risk.

Jesse Norman

I am conscious of the time, so let me just ask a question, if I may, of Mr Johnson. Wave a magic wand. You are sitting there advising the Strathclyde Commission on quadripartite allocation of block grant. Is there a case, based on considerations of equity and equality across the country—the UK as a whole—for significant changes between the different devolved administrations and England?

Paul Johnson

If you look at the way that it works at present, it is pretty clear that starting from a blank sheet and basing an allocation on the basis of population and relative deprivation, which are by far the two most important elements of any formula, but if you take account of other issues as well, you would probably—

Jesse Norman

Ageing, presumably.

Paul Johnson

Indeed, or sparsity or whatever, you would almost certainly end up with a different distribution from what you have at the moment, which might be somewhat more favourable, for example, to Wales and somewhat less favourable, for example, to Scotland. Where we are is based on historic population shares and historic agreements that do not bear much resemblance to the way that we allocate across local authorities within England.

Chair

That work has been done, has it not, for the Welsh?

David Phillips

The Holtham Commission reported in 2010 on this.

Jesse Norman

But is there any work that you at the IFS are planning to do to assist such a process of evaluation with a parallel to a Mirrlees-type review of the situation?

Paul Johnson

We are not planning that, no.

Jesse Norman

I think you should. I think it is an important question. There are issues of equity across the country that demand expert assistance and informed comment.

Paul Johnson

We would be pleased to do that with—

Jesse Norman

—suitable grant from us, yes.

Paul Johnson

—demand and suitable funding, indeed.

Jesse Norman

I would support that. Thank you, Mr Johnson.

Chair

We are back where we started, Mr Johnson. Would it not be a good idea if you have an initial go and see what the numbers might look like based on a reasonable assessment of need? You have just come out with what you think the main drivers of need should be.

Paul Johnson

It would certainly be an interesting thing to do.

Chair

Isn’t that what the IFS spends much of its life thinking about and writing about?

Paul Johnson

Absolutely.

Chair

Perhaps you would like to consider it, Mr Johnson.

Steve Baker

Turning to tax design, Mr Norman mentioned the Mirrlees review, which set out to design a tax system that minimised economic and administrative inefficiency, maximised simplicity and transparency and avoided arbitrary differentiation between different people and economic activities. Is a deviation from the Mirrlees review and its principles the necessary price to be paid for further fiscal devolution?

Paul Johnson

I think the key thing there is what the differences are between individuals that you want to take into account—clearly, being Scottish may well be one that you want to take into account, which we do not at present—and what are the principles by which you might think about devolving taxes? There are a lot of things you might think about, but I think there are three big ones. You have talked about administrative simplicity and complexity. That is clearly one. The second is associated with the extent to which behavioural change might result. One reason why we currently devolve taxes on property is that property is fixed. You are not going to get too much movement across the board as you are not going to create too much in the way of either complexity or behavioural change. Business rates and stamp duty and so on—perfectly sensible. The next level down, you might think consumption and income taxes. There is some risk there if you have big differences between the countries, but you can probably design that and many countries have designed different consumption and income tax regimes across states within a unitary state. You would probably put corporation tax and other capital taxes at the bottom of that list because the capacity to change behaviour or change apparent behaviour is more significant. The third thing I think is important is to do with the scale and the volatility of the tax that you are talking about. You started off asking about oil taxation and petroleum revenue tax. That fits pretty well on the first two. You could probably separate it from corporation tax and you could probably administer it separately. It is pretty clear where the oil is, but for a nation like Scotland, to have petroleum revenue tax devolved would create an enormous amount of complexity in terms of what we were talking about earlier, in terms of how you adjust the block grant. This is an extraordinarily volatile revenue stream and it is one that is probably falling over time. Scotland is very much smaller than the rest of the UK. The UK as a whole can cope with that volatility a lot easier than Scotland could. You would probably need some staggeringly complex mechanism for creating equalisation for compensation through the block grants. I think that issue about scale and volatility is the third important leg of how you might think about devolution of tax powers.

Steve Baker

Is there anything you would like to add, Mr Phillips?

David Phillips

No, I think that covered the issues there. One thing I would add is on corporation tax. As well as thinking about what taxes are devolved, an important issue to think about is the way in which the tax is devolved. This is not in terms of how it interacts with the block grant but the way in which the tax bases are argued between countries. For instance, on corporation tax you can think of the way it is traditionally done across countries, which is on things like transfer pricing, where companies determine prices for cross-country transactions; or you could do what is done in many countries, like the United States and potentially Germany as well, where they have a variation in corporation taxes across states. They use a formula that is based on things like how much salaries are paid in a different region and where the physical capital is. The idea is that is less easy to shift around than—

Steve Baker

With that very much in mind, you have given a very clear explanation of your hierarchy of concerns, I think it would be fair to say, with different taxes. To what extent do you think taxes can diverge in Scotland from the rest of the UK before there are important material effects?

Paul Johnson

We do not have terribly good evidence on that, of course. You would not be allowed to do this very much with consumption taxes but one would imagine that some small changes in consumption taxes would not be a great problem. Fuel duty would be an obvious one. You do have this advantage of a fairly sparsely populated border for things like fuel tax. You could imagine that being different and you can imagine there are quite good economic reasons for thinking it ought to be lower in Scotland, because there is less congestion, and if you are thinking about externalities that driving creates, you might think that there is a case for that. Basically I cannot imagine that varying the income tax by two or three pence would make an enormous amount of difference. If you had a variation of five pence at the top, you might create some change in actual or apparent domicile. We have very little in parochial evidence. My usual instinct in these things is that small changes will make not very much difference except for some groups who we know are particularly sensitive to these things.

David Phillips

Again, to go back to corporation tax, how much you could vary the tax rates by would depend on the arrangements which they are devolved by. You could probably have greater variation in corporation tax if it was done on a formula basis than if it was done on a transfer pricing basis just because of the scope for tax avoidance being less.

Steve Baker

Where around the world do you think the Government would be best advised to look to see the lessons in this regard: the United States, European Union, somewhere else?

Paul Johnson

In general?

Steve Baker

In general, yes, in comparing countries and the economic impact of different types—

David Phillips

Countries with federal systems of taxation. There are countries that have much more decentralised tax system than the UK—for instance Denmark has local income taxes; I think Sweden has that system as well, and Switzerland. There is lots of experience internationally you can look at to see how countries manage having different tax rates in different parts of the country and how they try to reduce the scope for avoidance and evasion.

Alok Sharma

Can I just turn to corporation tax and potentially devolution of corporation tax. It has been reported that in the autumn statement we may have powers set out for corporation tax rates in Northern Ireland to be devolved to the Assembly. We will see what happens, but how does that then fit in the context of devolving corporation tax to Scotland? If you do it for Northern Ireland, do you think it is then going to be impossible to refuse it for Scotland?

David Phillips

Ultimately that is a political question again. Northern Ireland says it has special circumstances with the border with the Republic of Ireland and the need to compete with the lower tax rate there. I know that the Scottish Government says that it wants corporation tax anyway and I am sure that if Northern Ireland gets it, it will definitely want it. Wales has also said that it does not think it should be devolved, but if it does—

Alok Sharma

The Scottish Government makes the argument that being in London and the south-east you have a competitive advantage, so to redress this and to increase investment in more start-ups they need to be able to vary their corporation tax rate. Is that a compelling argument for you?

Paul Johnson

It is a plausible argument. Again, we do not know the sensitivity of real investment to small changes in corporation tax. It is right to say that London and the south-east has some kind of competitive advantage to some kinds of business relative to Scotland. If you look at California, they have a very high level of taxes and a very complicated tax system, but everyone wants to go to Silicon Valley however rubbish the tax system. I would not say we are quite in the same position in London relative to Scotland, but there are cases for having different systems for that reason. We struggle, don’t we, in managing different corporation tax systems across different countries at the moment. You would clearly add a layer of complexity by devolving to Scotland and Northern Ireland. I do not know how this might happen in Northern Ireland, but there are ways it might happen that would make it very unattractive for the Northern Irish to make use of it.

Alok Sharma

Mr Johnson, could you just repeat that? You said it would be unattractive to use that?

Paul Johnson

There are ways you could do it that would make it unattractive, depending on how much additional revenue they could keep.

David Phillips

For instance, when Northern Ireland cuts its rate, you would need to reduce the amount of money that Northern Ireland gets. There are ways you can do that. One is it just bears the cost in Northern Ireland of the reduction of the tax rate. I am aware the Treasury has also been considering whether Northern Ireland should also bear the profit shifting that is taking place from the rest of the UK, so it bears some of the cost on the rest of the United Kingdom. If you expect a lot of the behavioural response to a lower corporation tax in Northern Ireland or Scotland to be shifts from the rest of the UK—

Alok Sharma

They could be given corporation tax devolved to them, but what you are arguing for is setting up a system where there is net benefit of doing it and that is not the argument that is made in either Northern Ireland or indeed Scotland. The whole point for them of devolution is to be more competitive, so have a 12% tax rate as opposed to 20%.

David Phillips

Yes. I am not saying I am in favour of this way of doing things. All I am saying is that a key issue is what costs—

Alok Sharma

You talked about the Treasury. Are you suggesting this is the way the Treasury is thinking currently?

David Phillips

I know it has been one of the things that has been discussed by people looking at the issue of: how do we adjust the block grant to Northern Ireland and do we need to take into account the fact that a lot of the extra revenue they are going to be getting if they manage to become more competitive will be coming from the rest of the UK?

Alok Sharma

Let us say that either Northern Ireland or Scotland get the opportunity to basically set their own corporation tax rate and they make them materially lower than that in the rest of the UK. In your view, is this going to encourage lots of corporates to headquarter there? Have you done any analysis on what the impact of this could be?

Paul Johnson

Putting a number on that is almost impossible because we simply do not have the experience on which to base it. We do not know that that clearly is a result of the southern Irish tax system. Some multinationals, some corporates, have headquartered there. How that would relate to what would happen in Northern Ireland or Scotland, for example, it is just not possible to say, but it is clear that that could happen.

Alok Sharma

If there was to be a change in terms of devolving corporation tax, how long do you think that process would take in terms of the transfer and setting up the systems and all the rest of it? Is this feasible within a few years, or is this something that is so complex that, in your view, it would take many years to implement?

David Phillips

I fear that is a question you might have better directed to the previous panel.

Alok Sharma

I understand that, but I am asking what your thought is.

David Phillips

The process for devolving taxes under the Scotland Act and the Wales Bill envisaged several years. I would imagine that the challenges for corporation tax would be at least as hard as for income tax, and I would say that they would want to give real consideration to the way to do it. Rather than going for the default option of transfer pricing, spend the time to think about how we can do this in a way that still gives incentives for the devolved countries to improve their economy but does not give huge opportunities for tax competition and avoidance across countries. I think it would take some time and it should take some time so it is done right.

Mr Ruffley

Mr Phillips, returning to the trenchant answer you gave to Mr Norman, you said that if the yield from income tax were to outperform the rest of the UK, then they could, all other things being equal, be winners, and Mr Norman suggested there might be some potential inequity in this. Just thinking for a moment about income tax—not something volatile like taxing North Sea oil, just income tax—what work has been done that would get round this problem of inequity, as Mr Norman hinted at, by clawing back some of the block grant in the event of the income tax yield in Scotland outperforming the average for the rest of the UK?

David Phillips

There is a number of ways to think about this. As with many things in economics, it is a question of, to some extent, equity versus efficiency. Scotland getting to keep its revenues if it outperforms the UK means it has more incentives to grow the economy. Of course, that could mean you have less revenue utilisation across the country. I have not thought in detail about the mechanisms by which you could allow this to happen for several years and then stop it but—

Mr Ruffley

That is behind my question. I am asking has any work, not necessarily by you, been done—

David Phillips

I am not aware of any work specifically on income tax being done on this. However, something similar is happening with business rates where they have localised business rates partially to local government in England; they are allowed to keep the increments in business rates for 10 years and then, after that, the calculations are reassessed and that gets taken out. The fact that we have done it is not quite right because at the very end of the 10 years, you have no incentives to postpone growth until after the 10 years is up, but there could be some lessons to be learned from what has been done with business rates. It might also be worthwhile looking at countries that do have quite explicit devolved tax rates but also revenue equalisation—I think Canada and Australia do that—to see what they do in that regard, to see whether they have ways in which some of the growth is kept but some of it is not. One option would be to say Scotland keeps a proportion of its relative growth but not all of it. That would still provide some incentive, a weaker incentive, but would also have some equalisation as well.

Mr Ruffley

But there is the risk that if you reduce the incentive too much, they will not set tax policies to grow their revenues. They will just think it is a zero sum game and what is the point?

David Phillips

Indeed.

Mr Ruffley

I understand that, but there is a halfway house where they would not. You mentioned business rates in this country, which is an imperfect analogy, but you think there might be an analogy there?

David Phillips

Yes, because the idea was they wanted to give incentives but did not want to allow an equalisation.

Mr Ruffley

To your knowledge, is any work being done on this along these lines in Government?

David Phillips

To my knowledge, there is no specific work looking at it, but you can spend some time thinking about them. As I said, an obvious solution would be, rather than keeping all of the relative growth, they keep half the relative growth or half the relative loss for that matter.

Mr Ruffley

Because it is more likely than some of the other options—bigger devolution of income tax—it is rather surprising that this does not seem to be a piece of work that is going on. Could I invite you to look at this in the same way that the Chairman has asked you to look at other things or at least be in contact with the Treasury about it?

David Phillips

As always, we will consider it in the light of our resource constraints.

Mr Ruffley

Thank you. That was all to do with the question of income tax. The widespread assumption from the majority of parties is that savings taxation and VAT probably will not change. Working on that basis—that there will not be devolution of VAT and there will not be devolution of tax on investment income—it has been suggested that a proportion of revenues from those taxes be assigned to the Scottish Government. Mr Johnson, how easy would it be to determine a fair share of that set of tax revenues?

Paul Johnson

As a rule, it would depend on what you were trying to achieve. It would not be in the least bit difficult to say Scotland should get x% of the VAT revenues, and offset the appropriate block grant as a result. That might not be a bad revenue to do it with, because VAT revenues are relatively non-volatile so it would not be quite so sensitive to some of the decisions that you would make about how equalisation would happen in the future. You could probably set that fixed for two or three years and then reassess—but you would still have this choice, as David was describing, about whether you were offering Scotland the relative risk or the absolute risk. If the Scottish VAT revenues continued to rise relative to the rest of the UK, would they get all of that, would they get some of it, as you were describing earlier, or would they face an absolute risk such that if their absolute level went down and the rest of the UK’s went down, their total revenues would go down more than proportionately to the change in the VAT? The choices remain the same. I think VAT has the big advantage of being one of the least volatile of the big taxes.

Mr Ruffley

Referring to Mr Phillips’s reply in relation to income tax, it would be possible, potentially, to do the same kind of exercise of clawback if the revenues from VAT and the revenues from taxation on investment income and related taxes outperformed the UK average?

Paul Johnson

Or underperformed, yes, absolutely. You can play whatever tune you want, in a sense. You can give Scotland all of the upside and downside or half of it, or whatever you want to achieve.

Mr Ruffley

Sure. We have talked about income tax. We talked about tax on investment income. We talked about VAT. I think we can all agree from your earlier answers that something like North Sea oil taxation would be potentially too volatile.

David Phillips

I think it would be the most difficult of all in many ways because of volatility.

Mr Ruffley

It would be the most difficult. What other taxes would be susceptible to the theoretical possibility of a clawback along the lines Mr Phillips has hinted at?

Paul Johnson

I think the big taxes are all rather like that. National insurance, abstracting from the issues around whether it is or is not a contributory tax, would be in some ways relatively straightforward, because it is only charged on earnings, so you would not have this issue about different treatment of earnings and investment income. The big three—VAT, income tax and national insurance—which of course are 70% of the total tax revenue anyway, I think would all be fairly susceptible to that same kind of issue about how much you wanted to equalise.

Mr Ruffley

Is this a possibility that you think could be put into practice? Obviously a lot of work, but it is doable, do you think?

Paul Johnson

“It” being choosing the degree of compensation or risk or whatever?

Mr Ruffley

Yes. I would say clawback of grant—but it does not pose, from your expert point of view, insuperable possibilities?

Paul Johnson

No. From an economic point of view, I do not think it creates any kind of insuperable issues. In a sense, it is one of the big decisions around devolution, which is essentially about how much you want Scotland or Wales or whatever to take all of the risk and all of the upside and all of the downside of what happens, or do you want to have more equalisation within that? Once you have decided how much of that you want, it would seem perfectly possible to create what you want to achieve.

David Phillips

The only thing I would say is you need to think carefully about the mechanism, because they have tried to do with it with business rates, but because there is this fixed 10-year cut-off, one of the issues about the cut-off is it delays new developments rather than have it—

Paul Johnson

You need to get that right.

David Phillips

You need to get it right because at the moment the incentives are delayed until after the cut-off, so you get it for 10 years rather than one or two years. You would need to consider whether you are creating similar issues for income tax or national insurance or VAT. You need to think about making sure you get the mechanism for clawback or for sharing or pooling correct so you do not get these distorted incentives.

Chair

Do you not think, though, Mr Johnson, with respect to North Sea oil revenues, that it is a matter for the Scottish Government to decide whether they want to absorb the high level of volatility associated with devolution?

Paul Johnson

Sure.

Chair

They do not have to say yes.

Paul Johnson

They need to understand that. The difficulty would arise in the negotiation about what that meant for the block grant, I think. For example, if you were to devolve income tax revenues and you were to say you will reduce the block grant by the current year’s amount of income tax revenue that you are assigning to Scotland, it matters quite a lot how you then adjust the block grant over time.

Chair

We are back to where we began our discussion, you and I, about half an hour ago.

Paul Johnson

Exactly. If you do that with petroleum revenue tax, then almost the only thing that matters is how you negotiate how much the block grant is, unless you get a Scottish Government saying, “We do not care. We are willing to see our block grant fall over time.”

Chair

But, as you have said, that is basically a political decision and a political discussion that needs to be had anyway.

Paul Johnson

Technically, it will be extraordinarily difficult to come to an outcome that will be acceptable on both sides. It will be so difficult because you are dealing with three things. You are dealing with a huge amount of volatility year by year, you are dealing with a tax that is going down over time and you are dealing with not knowing what the impact of actual policy change will be. Getting to a clear agreement about how each of those would change the block grant I just think would be a very difficult thing to achieve. I am sure it is not beyond the wit of man, but it would make sorting out corporation tax look pretty easy.

David Phillips

Because North Sea revenues are overwhelmingly from Scottish waters, there is not the natural counterpart. One can look at most taxes and say, “Well, let’s change the block grant reduction in line with what is happening to the tax revenues in the rest of the UK,” but for North Sea oil, because it is nearly all in Scotland, there is not really a tax base in the rest of the UK to see what is happening. By definition, you almost have to have one of these more complicated ways of doing things because we do not have this nice simple comparator to compare to.

Chair

To be frank, listening to both those answers amounts to saying yes to me, does it not? If the political will and interest in wanting to do this in Scotland exists— All the problems you were describing are disproportionately problems for Scotland rather than England and, therefore, volatility and the decline, for example, two of the three problems you expressed—

Paul Johnson

Sort of. You need to get agreement on it. Going back to Scottish business rates, getting the agreement wrong would be disproportionately a problem for Scotland. As it has turned out, it has turned out to be a problem for England. Getting the agreement over oil taxes wrong would be a disproportionately big hit on Scotland, but it is not as if there is no hit on England if they get it wrong from the other point of view.

Chair

I do not know why you are baulking at the question I put to you. The only reason I am prolonging this set of exchanges is because you baulked at the question I put to you three or four minutes ago, which is it is a matter for the Scottish Government to decide whether they want to take those risks and have that discussion.

Paul Johnson

It is not just them that gets the risk. The point is the risk is not just for them. You might find a way of getting a perfect agreement, in the sense that you end up in the same revenue situation as the rest of the UK and Scotland as you otherwise would have in five years’ time. I think that is almost impossible to achieve. Therefore, either Scotland wins or the rest of the UK wins. It is not just an issue for Scotland.

Chair

But that is in the nature of things with devolution generally, isn’t it?

Paul Johnson

It is, but it is in spades on this one. I think it is just very hard to achieve what you are looking for and you could inadvertently lose Scotland a large chunk of its budget or lose the rest of the UK—

Chair

I am not even sure it is in spades, but each of your arguments, it seems to me, is, if I may say so, somewhat tendentious, even there. If you look at the radical end of the Strathclyde proposals for devolution, what are they worth in fiscal terms?

Paul Johnson

Sorry, say that again?

Chair

I think I know the answer, but if you look at the radical end of the Strathclyde proposals in annual fiscal effect, do you have a rough idea what they are?

David Phillips

The percentage of Scottish expenditure funded by—

Chair

In cash terms, in billions.

David Phillips

In billions of pounds. At the extreme end, if you include allocating VAT to Scotland—I think one of the things they considered was apportioning VAT to Scotland—about 60% of the Scottish budget.

Chair

If you exclude VAT?

David Phillips

About £13 billion.

Chair

I thought it was £15 billion to £16 billion, but anyway, we are using pretty much the same figures. What are oil revenues?

Paul Johnson

What are they, £10 billion?

David Phillips

It depends on who you agree with in future forecasts.

Chair

You are the IFS. I am asking you your view.

David Phillips

The OBR forecast in 2016 there will be about £3 billion. The loss to the Scottish Government is £7 billion.

Chair

Now we are looking at the comparison of these numbers, £15-ish billion against £3 billion.

David Phillips

The point is about the volatility. That £15 billion can go up and down a bit.

Chair

Of course, but one is small and one is large.

Paul Johnson

I am not sure the point you are getting at.

Chair

I think it is fairly obvious.

Paul Johnson

Not to me, I am sorry.

Chair

I am going to move the questioning on. I think Stewart Hosie has understood the problem.

Stewart Hosie

Just a general observation. There is a slight frustration, listening to this discussion, about Scotland being talked about in the third person. Paul, at one point you said, “They, the Scottish Government, have to understand the risk.” I think we should have a working assumption that the people in the Scottish Government absolutely understand the risks and the potential opportunities from all of the options that are on the table. In a paper that the IFS produced earlier about the fiscal context of independence, they said that in the long run Scotland would face stark choices over public spending, partly as a result of diminished North Sea revenues. Mr Johnson, has the IFS included in its future forecasts all of the new fields announced in the last couple of weeks?

Paul Johnson

If they have been announced in the last couple of weeks, we would not have done. That work was based on OBR projections on oil revenues.

Stewart Hosie

I simply asked that question to make a rather serious point, that volatility cuts both ways.

Paul Johnson

Of course.

Stewart Hosie

As there are new fields and prolonged life then that needs to be taken into consideration just as much as any of the downsides in this.

David Phillips

I agree. I would also add, though, that what Paul said is true; our baseline forecasts were based on the OBR’s North Sea forecasts. However, we also used a variant forecast where oil revenues were around three times as high as forecast by the OBR and, even under those scenarios, Scotland’s fiscal position still looked weaker than the UK’s as a whole in the longer term. Although we have not taken into account expressly the new findings, we did have high and low forecasts to take into account this level of volatility that you rightly point out.

Stewart Hosie

I will come back to the fiscal position later, but in terms of this volatility argument, Mr Johnson, and particularly in relation to PRT, which you mentioned specifically earlier, the Revenue tell us that Scotland’s geographic share of PRT for ’13-‘14 was £800 million; even a substantial change in that, say £200 million, against a GDP of around £150 billion or a tax yield of around £60 billion in and on behalf of Scotland—the IFS are overplaying the volatility argument in relation to PRT, are they not?

Paul Johnson

If you look at oil taxes generally, it is hard to overplay the volatility. They have moved by 200% to 300% over relatively small numbers of years and they are a very much bigger part of the budget of Scotland than they are of the UK as a whole. There is no getting away from that fact.

Stewart Hosie

There is no question that they are more significant, absolutely not, but in terms of volatility and the way price changes, particularly from the forecast, it is equally true, is it not, that when prices fall we can point to—I think Alistair Darling as Chancellor was suggesting a revenue yield for North Sea oil of £6 billion, but the outturn was £11 billion. So it cuts both ways. Is it not the case that the proposals previously put forward for an oil stability fund to manage that volatility are in fact a perfectly viable approach to dealing with that issue?

Paul Johnson

The stability fund would require money to go in it in the first place. Therefore, you would require further taxes from somewhere else or further borrowing from somewhere else in order to set it up.

Stewart Hosie

I am simply making the point that, in principle, one can put tax yield away in the good times into a stability fund to be used in the bad times. That is a perfectly viable route, irrespective of how it is populated.

Paul Johnson

Or equivalently, at the moment, you could vary the amount of borrowing that you do. We are borrowing large amounts and you could borrow more.

Stewart Hosie

Indeed. On the economic position generally, the deficit, it is true last year that Scotland’s deficit was worse than the UK’s, mainly down to the largely tax-deductible £14 billion of investment in the North Sea, but it was substantially better the year before, on average, it was better for the previous five years and, on average, it had been better for the previous 30 years. It is not wrong to be continuing to make a case that would imply Scotland has a worse structural deficit than the UK when the last 30 years would tend to indicate that is not the case?

Paul Johnson

You are right: there was clearly a period of years in which, if you allocated all North Sea revenues to Scotland, the deficit was smaller, and we are going through a period of years now where it is larger. That is down to the volatility in the oil revenues. It seems to me the arithmetic here is extraordinarily straightforward. Excluding North Sea oil, spending per head in Scotland is 15% more than the rest of the UK, and tax per head is roughly the same, a little bit less. Over some years in the past the North Sea oil revenues have more than made that up. At the moment, and it looks like going forward, they will less than make that up. That is the arithmetic.

Stewart Hosie

Just on that point, let us be absolutely clear, because you were very clear about what was excluded, but of course if one includes the total tax yield from Scotland, including geographic share, it is also the case that tax yield per head over the last 33 years has been £1,300 a year higher than the UK average. Is that not also correct, just for the sake of clarity?

Paul Johnson

I am not familiar with that particular number, but it has certainly been the case that over a number of years, if you include North Sea revenues, they have been higher. Looking at the present and the best estimates we have of the future, even with, as David said, pretty optimistic views about tax revenues in the future, that is unlikely to be the case.

Stewart Hosie

It is just to get a balanced picture. When you said in that report that Scotland would face stark choices over public spending, that means—forgive me if I am paraphrasing this wrong—that there would either have to be tax increases or service cuts to fill what is euphemistically known as a black hole. Can I just confirm that the IFS assessment of these things for Scotland would never take into consideration the dynamic changes or the dynamic effect of policy changes, and it is basically policy as is with the numbers as you forecast them?

Paul Johnson

Sure. It is worth saying that this was all in the context of independence rather than devolution, but the numbers were a mechanical looking forward under current policy. Clearly, and as I think we stated, if you were able to do things that changed the growth of the economy, migration, tax revenues and so on, then other things change and, again, we did—

David Phillips

We did some scenarios.

Paul Johnson

We set some quite significant scenarios.

Stewart Hosie

That is extremely helpful. It is just good to get this on the record. What the argument is effectively saying is that the so-called black hole that is identified is not a picture of Scotland under independence or devo-max or any form of devolution. It is a picture of the Scottish economy under current UK policy.

Paul Johnson

It is the best estimate we have of where an independent Scotland would be under current policies.

John Thurso

I must just comment very briefly. One of the fun moments of a long and happy summer was on various referenda hustings and, for my sauce, a wonderful document that we know in Scotland as “GERS”, the “Government Expenditure Revenue in Scotland”, where there is a wonderful table of all the taxes per head in which I note that Scotland pays more tax in a considerable number of areas. We pay more tax on tobacco per head, we pay more tax on alcohol per head and we pay more tax on gambling per head, but in the area of income tax we pay less. I always wondered what that said about us, other than that we might possibly want to change policy going forward. Can I come back to what we are looking at today? One of the things, and I think you mentioned it earlier, is the disconnect between the money raised and the money spent. You made the point that the three big taxes—income tax, national insurance and VAT, at £153 billion, £107 billion and £106 billion—taken together, are something like 73% of the total tax take and corporation tax is 9%, roughly, at some £40 billion odd of the total UK tax take; and however you cut it, the proportions would be the same at the moment in Scotland. Having the power to vary, say, corporation tax, in terms of the amount of money raised, is going to be a relatively small amount of money. The importance of it is it brings in companies, hopefully, which then have more people and you, therefore, collect the taxes in the other areas. It is a bit of a loss leader really. That then begs the question: what are you spending the money on? It comes back to, if your objective is full fiscal responsibility—in other words broadly you are seeking to raise the money that you are going to spend on the things you are—but in parallel to what you are deciding to devolve in taxation, you have to be clear about what you are asking it to pay for, and the objective is to get as close as possible. I put that as a background to my first question. In the paper that Stewart Hosie referred to, you did a critique of the Scottish Government White Paper in which you said there were more giveaways than takeaways, the implication being that it was perhaps not as fiscally responsible as it might have been. Would you expect the outcome of a good devolution settlement to be greater fiscal responsibility and accountability in Scotland for itself?

Paul Johnson

Yes. I think a good devolution settlement would create the kind of discipline that would mean that if you are raising less revenue that has an impact on the amount that you spend.

John Thurso

In that case, how do we set about designing it so that, rather than having lots of adjustments to a block grant, which are, as you have pointed out, extremely complicated, you are starting from the opposite end, which is to say, “This is what you are spending on and these are the ways you can raise it and, because we are in a United Kingdom, there will be a smoothing arrangement at the federal level? That is completely the opposite to how we now look at it, which is we, central government, give you the money and you go play with it.

David Phillips

We have not yet looked at the arrangements you could have for a system somewhat like Australia or Canada use, where they have a formula that determines a spending need. The states have their tax revenues and then there is a top up from the centre to top up the assessed level, or in certain rich states like Alberta in Canada, for instance, they take money off them and give it to other states. There are certainly options of how you go about doing that. We have not looked at it. Now, I am not necessarily sure that it would be simpler or it would be less controversial. These funding formulas could be incredibly controversial to set up. That is one of the objections that has been made to a needs-based formula. I am not saying you should not do it, but—

John Thurso

Can I just give you a “for instance” and ask you to comment? Just to be absolutely clear, my personal view has always been home rule for Scotland and a federal United Kingdom, which will come as a great surprise to everybody around this table. The core of that, if you were looking at proper home rule, is that you have to be responsible for the spending decisions you make. If you are going to spend more money you have to raise it and that means you go to people at an election saying, “I will do more here but you guys are going to pay for it.” That is the trade-off for every responsible level of government from local to national always has to do. When we are talking about “there is a block grant and we will take a chunk off it and you can have some flexibility on it but if you raise more we will lower the grant” then you get into a situation where—

David Phillips

That is not how it works.

Paul Johnson

I do not think that is how it works or what is proposed.

John Thurso

It is the reason why nobody has ever used the 1998 powers. The 1998 powers are unworkable because, with the variation of the income tax, whatever you vary down comes off the block grant and whatever you vary up comes off the block grant. So what on earth is the point?

David Phillips

That is not the powers under the Scotland Act of 2012.

John Thurso

Exactly, that is different. It is untested as yet.

David Phillips

This point has been raised before. What I would say is that countries do have these incentives and these responsibilities under a system of the block grant, provided you set it up in the right way so that if they do increase their tax rates, they do get to keep the revenue. Again, it is all about making sure that the devolved tax powers interact with the block grant and the Barnett formula in the correct way. It can be gotten wrong. I am not aware of how it is done in the 1998 Act, but if it is as you say that seems to be a silly way of doing things. There are better ways of doing it and, given the incredibly tight timescale with which the Smith Commission is supposedly working to, it has to be taken as given the block grant way of doing things rather than moving towards a more favourable—

John Thurso

I think the Smith Commission is going to come up with the core principles and direction of travel. I do not think anybody is suggesting that a year later everything in Smith is going to have been enacted. A lot of the 2012 Act does not come into until 2016. It has taken a lot of work to get there. I think everybody on all sides of this argument recognises that the route from decision to implementation is going to run across quite a number of years if we are going to get it right. I think that is pretty generally agreed, is it not? The point is it is more important to get the principles right than it is to do the detail, which will come once you have established the principles. Is that not the right way round?

Paul Johnson

Yes.

John Thurso

Can I just move on very briefly to a couple of technical questions? If we are all agreed that what we are seeking is as full a fiscal responsibility as possible, where the vast bulk of the money raised and spent is done in Scotland so that the majority of the responsibility lies with the Scottish electorate and the Scottish Government, do you think that there is sufficient within either the OBR or the Scottish equivalent to enable there to be sufficient data for the Scottish Government to those decisions?

David Phillips

I know the OBR is now forecasting the revenue yields for the taxes devolved under the Scotland Act. If there are substantial new taxes devolved, I am sure they would need to not only make forecasts for those new taxes but also invest more heavily in their Scottish forecasting. I know the Scottish Government has a national accounts team that has been trying to build its own estimates of GDP in Scotland. Again, I would assume that, with more powers being devolved to Scotland on taxation, there is more need to forecast additional variables, so you would need to ramp up the forecasting capabilities. So far they are looking at historic estimates of GDP rather than forecasts in Scotland, so they would need to introduce the forecasting element.

John Thurso

You have quite a number of elements in there. You want good data. You want to know that what you think you have received you have received and what you think you have spent you have spent. It is fairly simple accounting stuff, ONS stuff, and then you have all the areas of data, which have referred to, that enable forecasts to be done and so forth. The Scottish Fiscal Commission that has been set up is based at Glasgow University—I think that is where it is domiciled—but it has an OBR-like—I think it will have to work in tandem. They will have to co-operate. Have you had a chance to look at the operation of that and pass any judgment on it as to—

David Phillips

We have not yet, no.

John Thurso

You have not been able to look at that? Okay. It might be worth your while doing so because, in the same way we always ask you about the OBR’s performance when you come before us each budget, I suspect that similar questions will be raised by other people in relation to what goes on in Scotland. I will not bother asking you if you think it is fit for purpose or well contrasted or anything because you obviously have not looked at it.

Chair

Thank you very much for coming to give evidence this afternoon. Very interesting and somewhat controversial in places. Thank you very much indeed. Oral evidence: Proposals for further Fiscal and Economic Devolution to Scotland, HC 760 5