Official Report: Minutes of Evidence

The Chairperson (Mr McKay): I welcome Alan Trench to the Committee. It is good to see you again, Alan.

Professor Alan Trench: It is good to see you again, Chairman.

The Chairperson (Mr McKay): I welcome Professor Gerald Holtham to the Committee via Skype. You are very welcome, professor. Can you hear us OK?

Professor Gerald Holtham: I can hear you, just about.

The Chairperson (Mr McKay): Good. First, we will ask both witnesses to introduce themselves, outline any relevant biographical information and then to make an opening statement. We will start with you, Professor Trench, and then move to you, Professor Holtham.

Professor Trench: Thank you very much, Mr Chairman. I am, among other things, an honorary fellow of Edinburgh University and formerly a professor of politics at the University of Ulster. I have been working on aspects of devolution finance for about seven years now, and my experience in relation to it has included acting as specialist adviser to the House of Lords Select Committee on the Barnett Formula, which reported in 2009. I also worked on the Wales Bill and the implementation of the Silk commission's report on fiscal devolution for Wales in my capacity as specialist adviser to the House of Commons Select Committee on Welsh Affairs.

The Chairperson (Mr McKay): Thank you, Professor Trench. Professor Holtham, can you make some introductory remarks? Can you lower the camera slightly, as all we can see is your forehead?

Professor Holtham: OK. I am a professor at Cardiff University. I am chair of the Welsh Government's independent commission on devolution finance. I have been working on it since 2008. I have also given evidence to the Scottish Parliament on aspects of fiscal devolution there.

The Chairperson (Mr McKay): How is convergence affected by different influences, for example, population growth, additional public spending versus austerity, Whitehall spending priorities etc? Alan, if you would like to go first.

Professor Trench: Convergence is an arithmetical property of the Barnett formula. The Barnett formula allocates a version of spending changes that are made for England to a devolved Government, but it can never allocate to them more than 100% of what the changes are for England, even though the relative needs of all three devolved Governments are pretty clearly, as we know thanks to the Holtham commission's work, greater than those of England. So, convergence is something that is built into the system and ought to happen. It appears to have happened to a degree for Northern Ireland. It has happened to a more significant degree for Wales, and because Wales is now below its figure of relative need, it becomes a pressing political issue there. It does not appear to have occurred in relation to Scotland, and that is something of a puzzle.

One of the aspects of convergence is that it should occur more the more public spending increases in nominal terms. During the 2000s, it increased very significantly under the various Labour Governments. The best explanation for that appears to be that you use population figures for Barnett to establish comparability ratios that, by definition, are somewhat out of date because you use the mid-year figure usually from the year of the spending review. That is normally an estimate based on extrapolations that ultimately derived from the census. That figure will be a little out of date when the spending period starts and three years out of date by the time that spending review period ends.

Scotland's population has been declining significantly for a considerable time, certainly since the 1970s, if not earlier. That was, initially, at least, as a consequence of deindustrialisation. That trend continued through most of the 2000s. It appears to have reversed about 2007. The explanation appears to be that, even within that three-year spending review period, the decline was enough to cancel out the convergence effect that should otherwise have been taking place. Whether that is going to carry on happening in Scotland is a moot point, because Scotland's population is once again declining but at a significantly slower rate than it was up to about 2007. The rate of decline appears now, according to Office for National Statistics (ONS) forecasts, to be around 0·1% per year as opposed to 0·3% per year. So, convergence is quite unlikely to bite in Scotland. It appears to have bitten already to a material degree in Northern Ireland and bites quite painfully in Wales.

There is one other factor that is material in the context of whether convergence or, indeed, divergence happens, and it appears that, since the coalition Government came to office in Westminster in 2010, a material degree of divergence has started to emerge or re-emerge. That is because of the spending choices that have been made by the coalition Government in the way that they have delivered austerity so far. They have sheltered education and health spending in England, and those are, by some distance, the largest spending items for devolved Governments as well, certainly as far as the DEL element of the Northern Ireland budget is concerned and for the whole budget in Scotland and Wales. The sheltering of health and education has protected devolved Governments from some of the impacts of austerity in England, where the impact has been borne particularly by Departments such as Justice, the Home Office and Local Government, which account for much lower proportions of devolved spending than education and health.

Professor Holtham: The Barnett formula will give you some degree of divergence at a time

on public spending

The same applies to the divergence you get when public spending is growing. There are several other points. The amount per head is the same in cash terms. If there is a cash cut, a smaller percentage of

baselining power than

Therefore, some divergence

The ratio of public spending in Wales to the rest of the UK has risen by percentage point since 2009.

So, while austerity

it is relatively

in the sense that you

The other thing, as I say, is that my understanding is that the Barnett formula is not correct for population changes

so far as the level that is concerned.

change in expenditure

is not well adjusted.

in Scotland

Barnett formula.

Mr McCallister: It is good to see you again, Alan and Professor Holtham. I know that the professor did not see the research session, but we were talking about some of the pressures that might be put on Barnett if you start to devolve some physical autonomy to London and even if you extend that into other larger cities across England, such as Birmingham, Manchester, Liverpool etc. All that leads to the entire debate on the constitutional set-up of the United Kingdom, including what powers and how much central control the Treasury needs. How will the Barnett formula cope with that? Given that it was designed in the late 1970s, how will it cope with the much changed constitutional set-up that the UK might have in the next few years?

Professor Holtham: Experience shows that the Treasury will try to make the minimum bid possible. It will look for

The thing about devolution in England is that there has not been any real talk of devolving tax powers. They have been given the ability to borrow, and they may be given a little more latitude to change the council tax than they have had in the past.

There is no suggestion so far of devolving significant tax powers

I am not sure how far

To the extent that devolving powers to the devolved countries is concerned, the issue there is of how you value the tax

and what effect that has on the block grant. Those questions will come up

Whatever it is, you still have the issue when devolving tax rates, what price

How much am I going to reduce the block grant

I think that is a big issue, which

It will appear whether the Barnett formula is reformed or not.

Professor Trench: Let me start with the constitutional side of that question, and then I will come to talk about England. One of the things about the Barnett formula is that it was devised for a very different political situation, not merely a different economic situation. Actually, it was arguably devised for a rather similar economic situation. It was designed to deal with a great period of financial stress and the need to impose spending restraint very quickly across UK government. It was designed to deal with doing that in the context of a single UK Government in which all spending Ministers were bound by Cabinet collective responsibility and all formed part of the same Administration, dependent on a majority in one legislature in the House of Commons. Its use for devolution is a decision that I discovered was taken with remarkable speed and a remarkable lack of debate or discussion. That existing mechanism was simply applied to a situation where you were creating three new legislatures with their own distinct political composition, accountability and policy ideas. So, to use a system that is designed for that setting in the new setting has always struck me as being somewhat odd. The Barnett formula, because it delivers changes in spending according to decisions that are made about spending in England, contains within it an implicit assumption that devolved Governments will, broadly speaking, be happy with the overall pattern of public services in England. If you dramatically alter the pattern of spending in England, that feeds through the block grant.

We have already seen an example of that happening in relation to higher education. As part of the package that came when the coalition looked at the Browne report on university fees in England, the Government also slashed the teaching grant, which supports teaching in universities and which, formerly, was allocated by the Higher Education Funding Council for England to English universities to support all humanities and social science subjects. So, with the possible exception of mathematics, depending on how you decide to classify it, all those subjects suddenly lost government support. If you study English, history or politics in an English university, the costs of your higher education are entirely being paid for through your tuition fees. There is no longer any public funding going to support the teaching of those subjects in universities in contrast to medicine, mathematics or sciences. That cut in the teaching grant, in turn, fed through into devolved block grants because the teaching grant is allocated as part of the budget for the Department for Business, Innovation and Skills (BIS), and so it went through the Barnett mechanism as a reduction and was subject to the multiple that applies to spending changes for BIS.

Looking at the statement of funding policy, I see that the figure for that is 79·1% in Scotland, 78·4% for Wales and 79·2% for Northern Ireland. So, Northern Ireland will have had a population share of that cut multiplied by 79·2%. That has had a measurable effect, particularly in Wales, where funding is tightest. It becomes tighter the more money that is involved. If you were to dramatically change the funding basis of the NHS in England, this would start to have a very dramatic effect on what devolved Governments are able to do, if they are simply reliant on the block grant. Ultimately, a decision will need to be made. Are devolved Governments willing to accept, essentially, being able to do a version of what goes on in England or do they want to be in a position to make meaningfully different policy choices for themselves, not just at the margin in a fairly limited way but in a broader way?

I will answer the second part of your question about what happens in England. Barnett necessarily uses England as its reference point. That makes sense because England is by so much the largest part of the United Kingdom. The attitude that accompanied the introduction of Barnett in the 1970s and, indeed, its application to funding devolved Governments in the 1990s was that devolved Governments should get a share of what the UK Government are doing in relation to England. If you start fragmenting what goes on in England, that certainly creates difficulties, because you simply lose your reference point. Every system that operates any sort of grant funding arrangements needs a reference point. Usually, it is some sort of national average, either a tax capacity or a spending need. Because England is so big, England serves that function in the United Kingdom, but, if you were to go down the path of meaningful fiscal devolution in England, you would end up in some tangles.

I certainly agree with Professor Holtham that there is not much sign of that happening at the moment. Pitches have been made in relation to London for the devolution of stamp duty or for the assignment of its revenues, and those have been rejected by George Osborne. There have been suggestions that there might be further devolution of the non-domestic rate in England. There has been some movement on that but not a great deal. There has been a suggestion by the Commons Political and Constitutional Reform Committee that income tax revenues could be assigned to fund local government services in England. Again, that has not had any traction in Treasury. I am not sure that there will be a huge amount of progress in that sense, and, until there is, I suspect that this is a thing to think about rather than a thing to actively worry about.

Mr McCallister: I have a couple of points on that. On your point, Alan, if you see more policy divergence between what is happening in England and the devolved Administrations, for example, different decisions on tuition fees in Scotland and on water charging here and prescription charges, will pressure not come from English regions that might want that power over time? I accept that that maybe is not in a debate at the minute, but are we entering into a period now of a fairly fluid mix of what might change in the constitutional set-up of the UK?

Professor Trench: We are in a fluid period, but it is a very, very fluid period. Part of the issue in relation to England is that it is unclear what England wants. It is unclear what it wants in terms of devolution by way of area, and it is unclear what it wants by way of powers. There is a lot of talk about devolution within England as well as to England through the mechanism of Westminster and English votes for English laws, but it is not very coherent at the moment. So, there is a debate, but it is quite hard to say where this debate ends up. The timescale in which it is likely to end up is even harder to say.

Mr McCallister: I have a quick one for Professor Holtham. In Wales, they have had problems with funding due to the convergence. We have probably been reluctant to open up the Barnett debate here much because we have maybe been doing reasonably well out of it.

The coalition Administration in London protecting health and education has had an impact on us. We are falling behind in health because social care forms part of our Department of Health, Social Services and Public Safety, whereas it is separate in England. Therefore, there are some issues around our health and social care funding that are starting to show signs of pressure, but, generally, we have been shielded. Now that the Welsh are not doing as well, is there a major push to open up the Barnett formula for debate and move to a needs-based judgement?

Professor Holtham: As Alan was saying,

Wales to look at the Barnett formula or at least there has not been for six or seven years now, but, as you said, Northern Ireland has not done too badly. We are maybe only a few years away from being in the same position once public funding starts to grow. Certainly, in the past and at the moment, you were and are probably doing OK. The Scots, of course, would suffer if a needs-based adjustment were introduced to Barnett. Therefore, I think that the pressure for change is pretty

and there has been strong resistance to change from Scotland.

consideration,

that was made at the time of the Scottish referendum, Westminster

is quite sensitive to that and are not showing any readiness to adapt at the present time. If I may, I would like to disagree with Professor Trench on one point. We usually see eye to eye

but there is one area on which I disagree: if we take a step back, there are two ways to make an adjustment to the public spending capabilities of different areas. You either say

we will adjust your revenue to the same level as the rest of the kingdom. If that is what you did, which, by the way, is what is done in most federations, your expenditure per head would be the same as England's, because you would be getting the same revenue per head. Therefore, a revenue adjustment would not be

for the devolved territories as the current Barnett-based adjustment.

The Barnett-based adjustment is based not on revenue but on expenditure. Although it is not fully needs-adjusted at the moment, it is derived from history, where there was a higher level of public expenditure in the devolved territories, because they needed it. The population was more sparse. They were more expensive to service. They were poorer, so there were more social problems that had to be tackled by public spending. Therefore, it is only an expenditure-based adjustment that will give you the kind of resources that you currently have. If you move to a revenue-based adjustment, you will be worse off. How can you do an expenditure-based adjustment except by

those expenditures that are devolved.

Professor Trench says that the trouble is that that ties you into the expenditure pattern in England, because the grant is derived from the expenditure that it is making in areas that are devolved to you. In my opinion, that objection is answered by the fact that the grant is fully functional. It does not matter where it comes from. You can spend it on whatever you like. Therefore, if it decides to abolish the National Health Service, which would hugely reduce your grant, if you wish then to spend it on health or education, you would be able to do it. The real difficulty that Northern Ireland is in, just as Wales is, is that it receives such

a large grant. If England decides to cut public expenditure, that will have a disproportionate impact on Wales and Northern Ireland, but I do not think that there is any alternative. You cannot ask the English to provide public services to the devolved territories at a higher level than that which they are prepared to vote for themselves. The problem is that, because they pay the piper, they will call the tune.

If England elects a Government who are going to slash public expenditure and you do not wish to do it, you cannot then say, "Give us a subsidy at a higher level than you are prepared to support your own people at". The answer would be that, if you want to have more public spending, you can raise your own tax. The real problem we have is that Wales and Northern Ireland are in receipt of a very large subsidy. That is the problem. He who pays the piper calls the tune, and there is

Professor Trench is right in saying that it is down to the way the consequentials are derived from Barnett.

Your problem is the size of the subsidy and the

provides a partial answer to that. Therefore, unless you want to move to a revenue-based adjustment, which would make you worse off, there is really no

practical alternative to what is being suggested. The Scots have a practical alternative. They are not in receipt of a very large subsidy, if you count the North Sea oil as belonging

to them, so you can assign more of the tax revenues that they collect to them, and they would not have to have very much

. Unfortunately — I speak as a Welshman, so it is unfortunate — that is not true of Wales and Northern Ireland. We are in receipt of a large net subsidy. That is our problem.

Professor Trench: If I may come back on a couple of points —

Mr McCallister: We have caused a split.

Professor Trench: I entirely agree that, if you are a recipient region, as Northern Ireland and Wales are, you cannot expect to receive an overall higher quality of public services than the unit that is effectively paying for that, which is also, in this case, the reference point. As it were, the English norm is necessarily going to be the norm. There are things that you might be able to do about that if you have fiscal devolution at the margin, but it depends on the degree to which you have fiscal devolution and the taxes to which it applies.

There are at least a couple of federal systems that operate systems of equalisation that combine both revenue equalisation and spending equalisation. Australia and Germany both do that — the more centralised of the federal systems. Of course, the difference is that they are symmetrical. All the states and territories of Australia have the same spending responsibilities. All the Länder of Germany have the same spending responsibilities. Therefore, you work out the reference point for that based on a national average. You cannot do that in the same way in the United Kingdom, because only Scotland, Wales and Northern Ireland have devolution. That is why we end up necessarily using England as the reference point.

Ms Boyle: You are very welcome, Professor Trench and Professor Holtham. My question is to Professor Trench, because I can hear you a wee bit better. Apologies to Professor Holtham.

The current discussion around fiscal powers and, indeed, the role of Barnett should begin from the perspective of what we are trying to achieve here, which is ensuring that our Executive are fully able to create jobs and tackle inequality. What specific changes or additional powers would you identify as being central to achieving that?

Professor Trench: Devolved Governments already have quite a significant number of levers in their hands. In particular, they have many of the tools related to the active labour market. Indeed, through the nature of welfare devolution here, it would be possible to take that even further forward. You are not tied into the structure of Jobcentre Plus in the way that Governments are in Great Britain. You also, of course, have things such as vocational education already in the hands of devolved Governments.

One of the issues that is widely discussed here as a way of trying to stimulate the economy is the idea of the devolution of corporation tax and the use of a lower rate of that tax as a means of trying to spur economic activity. As you may know, I am rather sceptical about the value of devolving corporation tax. It may well be a case for Northern Ireland of beware of what you wish for. The devolution of corporation tax would put you within the ambit of some very restrictive rules that would necessarily be quite tightly enforced, notably by the European Commission on the issue of state aid. That would mean that you would have to establish the amount of corporation tax that was attributable to Northern Ireland; make a cut in the block grant that related to that amount of revenue; ensure that that cut in the block grant was adjusted in subsequent years by a mechanism that satisfied state-aid tests; and ensure that the cut was sufficiently robust, and that the methodology for adjusting the cut was sufficiently robust to do that. I suspect that the Commission would keep quite a close eye on what happened here. Then, assuming that a cut in the rate of corporation tax were made once you had obtained its devolution, you would have to apply an actual cut in spending to Northern Ireland Budgets. You would have to make a very tangible and rather substantial cut in immediate public spending, which would clearly be uncomfortable. It is very obvious from the experiences that Northern Ireland has had in the past few years, despite being comparatively generously funded, that that would be a difficult and uncomfortable experience. All that you, as politicians, would really be able to offer your voters is the assurance that, at some point in the future, that would spur enough economic activity that it would reduce the demand for public services and increase the tax revenues available to pay for them. However, it would be a case of a cut today in the hope of a benefit tomorrow. That is going to be quite tough to do.

Ms Boyle: Thank you, Professor Trench. Professor Holtham, do you want to add anything to that?

Professor Holtham: One area to look at is the borrowing powers of the devolved authority. I imagine that, in Northern Ireland, it is the same as elsewhere, in that those large borrowings, if you wish to undertake them, would come out of an excessive share of your annual capital budget. If you do not have the capability, without borrowing powers, to spread those expenditures over a long period, you are

in Northern Ireland or not. If there are, you are hampered. An agreement would allow you to accumulate a certain amount of debt on your annual income at some sort of reasonable level and help you at least to contemplate a larger budget. It would also partly get around the problem that Professor Trench just mentioned. If you were able to finance your capital by borrowing

you could fill the gap left by cutting your revenue

block grant.

from capital to recurrent spending or, indeed, the other way. You would have the freedom of the use of the block grant

, you would have borrowing powers and could do more. Of course,

you would have the power

to do more things. A general point that I would make is that Treasury is afraid of tax

, which is why our capital grant

. That is a legitimate concern.

the ability to devolve the higher rates of income tax. You do not want all your millionaires going to live in one place

. The answer to that is not a blanket ban

. The answer to this is

.

we may be ready to see EU funds go there. We may be ready to send regional development funds there

. There is a limited amount of

Isle of Man or something like that. Therefore, you need

. For any tax, if you are prepared to set what latitude

If those are specified within reasonably tight limits, the objections often go away. The trouble that you have with corporation tax is that the Treasury already knows what you want to do with it. It knows you want it at the same rate as the Republic and, therefore, can work out the cost

It is going to try and sting you with the bill

. You can devolve these things, but

.

Ms Boyle: Thank you both. That is very interesting.

The Chairperson (Mr McKay): Given the lack of transparent and independent public finance data, which leads to these situations in which the Treasury is making all sorts of estimates on the devolution of further powers, how can that be addressed to provide an assurance to the devolved Administrations that any devolution of further fiscal powers is done on the basis of a fair and proportionate adjustment to the block grant?

Professor Holtham: I think that that is a very difficult issue, and one that the commission that I chair looked at in some detail. The bad news that we came up with was that it is very difficult to find a general formula that you would want to apply to all taxes that would work. The truth is that different taxes have different characteristics, and the way in which you want to structure the deduction from the block grant differs from tax to tax. Nobody likes that, and we do not like it ourselves, but that does seem to be the case. For example, on income tax, we devised a measure for how you treat the deduction from the block grant, but the Treasury initially said that it was too complicated. We persuaded the Scots that that was the measure that they should have, and the Scots said that, if they got income tax devolved, they wanted the deduction to the block grant to be done according to the

formula. The Treasury finally rolled over and agreed to that. When it was discussing with your civil servants how it would treat deductions from the block grant for any corporation tax that is devolved, it said that it would use the Holtham formula for how it would treat the block grant. That would be disastrous in the case of corporation tax.

That is a case in point of one size simply not fitting all. What made sense for income tax made no sense for corporation tax. Therefore,

tell you what, I think, the answer is. The broad answer is that you have to look at the structure of the tax. Is it a tax-base problem? Is your tax base going to be growing faster or slower than the tax base in the UK as a whole? Things such as that need to be considered. You need to look at the structure of each tax and project how it will behave in the future before you can derive a reasonable formula for how you reduce the block grant. In year 1, it is easy. In year 1, you say, "Last year, this tax brought in x amount, so we will knock x off the block grant". You cannot keep doing that; otherwise, we have not devolved the tax, because any changes in the tax receipts are immediately offset by the block grant. If you think about that for a minute, you realise that there has not been any devolution, because your resources are always the same

. You cannot do that going forward. There has to be some other way of indexing the reduction in the block grant

. Unfortunately, as I said, the sensible way in which to do that for each tax depends on things such as how fast the tax rate is growing in your country and how fast it is growing elsewhere in the UK. That will be different for different taxes. My advice is this: resist any one-size-fits-all formula and insist on looking at each tax one at a time.

Professor Trench: Forgive me, Chairman, but were you asking about the impact of changes in the block grant from tax devolution or from further devolved spending functions?

The Chairperson (Mr McKay): Further devolution.

Professor Trench: Further tax devolution. There is a set of issues, which Mr Pidgeon drew attention to in his presentation, raised by the Silk Commission in Wales, where the devolution of further functions is a very active issue and where ensuring that those functions come across with adequate funding is also a substantial issue. Indeed, there have been some issues in Scotland, where devolution of rail franchising did not come with adequate funding. There is an issue about further devolution of substantive functions as well, if that ever happens in Northern Ireland.

Professor Holtham: Professor Trench on that. It is the case that, given austerity, the British Government will happily give you functions without the money. Again, one has to look at each function

Professor Trench: Indeed. Of course, as part of the devolution of council tax benefit in Great Britain, it deliberately transferred 10% less than the last year's level of funding that was needed to deal with that. We have already seen that move from being something that happens in practice to becoming an actual policy.

Talking about tax devolution, I will say without a moment's hesitation that Professor Holtham is the expert here. The work that his commission did is the landmark work that has been referenced in Scotland when dealing with those issues. I entirely agree that, on the one hand, a single approach does not work, but, on the other hand, you want to have something that is as reliable and predictable as possible as your mechanism for making the adjustment to the block grant in subsequent years, because what you desire from the point of view of making fiscal devolution work is to make that cut in the block grant as firm a Budget constraint as possible. It can never be a hard Budget constraint, but there are degrees of firmness, and you want to be at the firmer rather than the softer end to ensure that it creates a meaningfully transparent and durable system.

That said, I start to have some quite serious worries about what would happen eight or 10 years down the road if you were to apply the sort of approach that Professor Holtham has outlined, which appears to be the one being rolled out in Scotland, for income tax at least. You are going to end up with a calculation of what the block grant would have been and a mechanism for making a reduction from that block grant that will apply a multiplier to the initial pound sum, which, in the case of income tax, will be the overall growth in the UK tax base, and that may or may not relate to the growth of the size of the devolved tax base. It may well become increasingly hard to identify what the growth of the devolved tax base would have been absent tax devolution, not least because there will be behavioural changes. If you have a different tax rate, people will do things differently, so working out what they would have done if they had not done something else is going to become increasingly difficult. It is a bit like the exercise that now goes on, certainly in England and Wales, if you build a new property and have to have it banded for council tax purposes. You have to work out what the value of that property would have been if it had existed in 1991. That is a rather farcical thing to do in 2014.

I suspect that, after a number of years, there will be some real problems around the increasingly notional character of several of the calculations. I would rather forestall those and put a more rational system in place earlier on. I would certainly say that there will be a need for a review within a number of years to look hard at those issues again.

Professor Holtham: May I just make a point, Chair? Professor Trench has put his finger on a real problem. It is a problem to which I do not think there is a complete solution. If you devolve tax powers in such a way that the devolved Government have responsibility and take the consequences for those actions, you cannot at the same time ensure that that is affecting things outside of its control. There is just no way to do that. Therefore, if you say that we are putting some weight on devolution being genuine

the consequences of their actions, you cannot fully protect them from outside influences that might make them worse or better

for reasons beyond their control.

full protection against things that you cannot. I fully admit that my own solution is a sticking-plaster solution. As Professor Trench said, after a certain number of years, be it 10 or 15, you have to recalibrate. You have to go back and start again. OK, let us say we devolve income tax to start with in Northern Ireland. We index to the UK tax base, and, lo and behold

Maybe they have made mistakes. Either way, you cannot unscramble that. We have to recalibrate.

My argument is that 10 or 15 years is long enough to incentivise

I do not see any alternative to that. I am very ready to accept something better, but I am unable to think of anything.

Mr Cree: Good afternoon, gentlemen. One of the difficulties that I see is the fact that we have not been able to achieve convergence over the years. I would welcome views from both of you on an idea of interest-free medium-term loans from central government to allow the regional economies to grow. For example, it could meet the costs of corporation tax and APD in the short term and, therefore, allow the revenues to build up to make the whole thing worthwhile. That would create employment and more taxation etc.

Professor Holtham: I think that greater borrowing powers

borrow from the UK Government. That is certainly one way of doing it. You can issue your own bonds on the market and do it that way. I am in favour of the idea. One of the problems at the moment is that, if you do not have any tax powers, suppose you borrow money and suppose you

achieve convergence on GDP per head in line with the rest of the UK. While you are getting a block grant and you do not have any tax revenue, Northern Ireland is better off but the Northern Irish Government is subject to the block grant, and you are not able

the advantages of investment you have undertaken. How do you repay the loan? Even if it is interest-free, you have to pay it back. You are not getting any more revenue to do it. It does seem to me that, with the power to borrow, you are accepting some tax devolution because

achieve more convergence, you then get more revenue enabling you to repay your loan. If you do not get more revenue, how will you repay your loan without using resources? So, I think borrowing and some tax devolution have to go together.

Mr Cree: Thank you for that. I was thinking, for example, of foreign direct investment (FDI). If we had the lower rate of corporation tax, we could attract that and gain the revenue from it in due course.

Professor Holtham: whatever taxes you get, you need tax devolution if you are going to borrow, because otherwise it is difficult to repay or squeeze it for other spending

Corporation tax is one example. There could be others as well. I think that you have to have the two things together.

Professor Trench: I do not have very much to add to that, except I would, wearing my lawyerly hat, raise some serious questions about whether state aid rules would be breached if you were given interest-free loans by the UK Government for particular purposes, particularly if those involved supporting tax cutting. To me, that smacks of regional selective assistance in a way that is contrary to state aid rules. I think that you would have some very real legal problems there.

Mr Cree: So the Osborne example is not good enough, then?

Professor Trench: The Osborne example?

Mr Cree: Yes; getting a special loan for a year to soften the blow of —

Professor Trench: That is in relation to general revenues, not in relation to supporting a particular attempt to stimulate the economy. It is when you are trying to affect private enterprise, which appears to be the objective of what you have in mind, that you are going to breach state aid rules. You said medium-term loans. I suspect that the Commission would take a rather different view about emergency short-term and medium-term funding that might turn out to be longer than medium-term funding.

Professor Holtham: medium-term funding

borrow at less than 2%, so you don't have to run into the state aid problem by insisting the loan is tax free, just take it at 1·5%. If you cannot make that work, maybe you should not be doing it.

The Chairperson (Mr McKay): Given the radically changing politically landscape across Britain and in Scotland, and the possibility of a federal Scotland with "devo max", more powers for Wales and the devolution of corporation tax and perhaps other powers to Belfast, if that happens, is that the time for the formula to be recalibrated?

Professor Trench: I was intrigued when listening to Mr Pigeon's presentation earlier. He talked about the wave of criticism there has been for quite a long time regarding the Barnett formula. The Barnett formula has been regarded as dysfunctional since certainly the mid-1980s and been denounced in various terms and by various people since then. Yet it remains remarkably persistent. That is due to a combination of politics and administrative convenience. The Barnett formula, without a doubt, offers some very meaningful administrative virtues; it is worth bearing those in mind. From the point of view of devolved Governments and Finance Ministers, it provides stable and predictable funding for three-year periods and, indeed, on longer terms than that. That is very valuable; it insulates you from the risks of volatility of tax flows and the need to work out how you are going to finance public spending given the uncertainty of those tax flows. That is a significant advantage.

From the Treasury's point of view, it becomes a very simple accounting exercise to work out what devolved Governments get and the additional costs of any changes in spending for England. The Treasury will know that that is going to stick roughly 15% on a wholly devolved function. Therefore, the Treasury works in 85-pence pounds. Every 85 pence you spend extra is going to cost you an extra pound. The Barnett formula has those mechanisms.

I suspect that the politics now will mean that it will be very hard to change, however, because of the nature of the vow that was made in the last days of the Scottish referendum campaign to protect the Barnett formula in relation to Scotland. I think that people in Scotland would be wise to be careful what they wish for, because preserving the Barnett formula, given its various administrative shortcomings and some of the other problems it presents, is not necessarily an appropriate way to build a further form of significant, enhanced devolution for Scotland — which will not, one might add, be "devo max", which normally brings with it the assumption of full fiscal autonomy. That is certainly not going to happen as part of what is currently under way. While there are very good reasons to question it, equally, it shows little signs of dying just yet, as it has shown little sign of dying for about 20 years.

The Chairperson (Mr McKay): Would you agree, Professor Holtham?

Professor Holtham: I do. I think there is a little bit of wiggle room if they want to leave, because, thinking about the Barnett formula, it is just a spreadsheet.

The point is, you need to change the Barnett formula

this sort of weird discussion going on about whether we have the Barnett formula or something fantastically complicated with a million variables. I reject that as a false choice. If we put four or five variables into it, the Barnett formula will be just as simple to administrate as it is at the moment or pretty much. Then 80% or 90%

You can make a rather

adjustment, sort out 90% of the problem and leave you with a system very much like Barnett but

If people had political will, it would be possible.

The Chairperson (Mr McKay): In addition to the more well-known case of formula bypass relating to the 2012 Olympics, what other specific cases have there been of difficulties experienced by the devolved Governments with the operation or administration of Barnett?

Professor Trench: I am not sure that I would necessarily categorise the 2012 Olympics as formula bypass, because part of what was going on with the 2012 Olympics was deciding whether and how the formula applied at all. Deciding that it does not apply to something is not quite the same as bypassing it and saying that it does apply but you are going to disregard it in that instance.

I do not know how familiar members of the Committee are with what happened around the 2012 Olympics, but, originally, I gather, the idea had been that there were going to be three funding streams for the Olympics, identified as programmes of the Department for Culture, Media and Sport. One was to fund the Team GB and sporting side of what was going to happen around the Olympics, the second was to fund the construction of the facilities at the Olympic Park and the third was to fund the wider-area regeneration.

Originally, the idea had been that the first two would be designated as for the benefit of the UK as a whole, and therefore they would not trigger consequential payments under the block grant, but the third would be for the benefit of England only — because, obviously, regeneration relates very specifically to the area it happens in — and that would trigger a Barnett consequential. That was, by some distance, the largest element of the funding. However, when the statement of funding policy was being framed in the wake of the 2007 spending review, the decision was taken to merge those three funding streams into one and to treat the whole stream as being for the benefit of the UK as a whole. That meant that all of the spending on regeneration in Stratford — there were very extensive works — did not trigger Barnett consequentials. That is the first stage of the argument that broke out there.

The second stage was what happened when the devolved Administrations got together to object to that.

Ultimately, that led to the first meeting of the recently established dispute resolution panel, which was created under the revision to the memorandum of understanding between the UK Government and the devolved Administrations in 2010. Essentially, the Treasury said no, that it had made its decision and that it would not budge. So, even though there was a very strong argument that that had taken quite a lot of spending out of devolved Governments' pockets and that it was completely inappropriate — they exercised all the pressure and made the strongest argument they could — they still could not persuade the Treasury through the disputes resolution panel, which, for these purposes, was Francis Maude, the Cabinet Office Minister, a UK Government Minister who had not been involved in the process before. He had no authority to impose a solution and simply had to accept the Treasury's conclusion. So, effectively, the devolved Governments got nowhere with that process, and that is a problem that will remain built into the present structure of resolving disputes, particularly finance disputes.

To answer your question a little more directly, I mentioned higher education tuition fees and the precedent that they may set. There is scope for the UK Government to, in effect, disapply the Barnett formula by moving the headings of spending from public spending to private spending. When you move the burden of spending directly to the citizen, and it is not borne by tax revenues generally, it effectively takes something outside the Barnett formula. Over time, that can lead to quite a significant reshaping of public services.

Beyond that, there are not many direct instances in recent years of Barnett being bypassed in that sense. However, there have been a number of cases in which there has been a reformulation, a debateable formulation or an argument about the formulation of the statement of funding policy. For example, Transport for London (TFL) was originally classed in 1999 as being for the benefit of the whole of the United Kingdom. Therefore, spending on TFL did not trigger consequentials. It was recategorised in, I think, the 2001 spending review, which happened to coincide with the point at which there were large amounts of capital spending on tube improvements, originally through not-very-well-structured public-private partnerships. So, devolved Governments got a benefit of that additional spending on local transport in London. Something similar has happened in relation to Crossrail. At the moment, HS2 is classified as an overall Department for Transport area of spending and, therefore, also triggers quite a significant consequential. When there is another spending review, it will be interesting to see whether HS2 is categorised in a different way, because, of course, only £2 billion or so has been spent on that so far. There will be very substantial spending if the project goes ahead in the next Parliament, and how that is categorised will potentially trigger very significant consequentials for devolved Governments.

Professor Holtham: The problem is that there is no statutory basis for the formula, and therefore, there is nothing to stop the Government moving the goalposts as and when they see fit. I will tell you a little story. It is a fairly trivial instance, but it is illuminating.

the Royal Botanic Gardens at Kew is regarded as of national significance and, therefore, is excluded from having Barnett consequentials. However, the National Botanic Garden of Wales falls entirely to the cost of the Welsh Government. How the hell is Kew regarded as being of national significance, when the National Botanic Garden of Wales is not? That is unfair. For once, the Treasury gave way and put in a small consequential for expenditure on Kew gardens. It was very small.

he happened to look at the accounts a couple of days later, and it was back out again. Treasury put it in and out as it feels like it. There is nothing to stop it doing that.

initial assessment of what consequentials are. In my opinion, that could be done by the Office for National Statistics (ONS), the Office for Budget Responsibility (OBR) or some other body that is nominally independent, and the Treasury could take the spreadsheet.

At least let us look at the political scene

as a technical thing

I would think some statutory base for

would be appropriate. It is a catch-as-catch-can

Professor Trench: If I might add a couple of points to that. I entirely agree with Professor Holtham that there needs to be some independent body that makes a judgement about what ought to trigger consequentials, even if that is then subject to review by the Treasury. I am not sure that the ONS would either welcome doing that or be the appropriate body to deal with it. Perhaps, the National Audit Office might be a better body, although, of course, the Comptroller and Auditor General is an officer of the UK Parliament, so that might raise a slightly different question. The other suggestion is one that Professor Holtham and I have recommended in various settings, and that is the idea of some sort of independent finance commission. I think that both of us would agree that the UK needs that sort of body and that that would be an appropriate job for that body to do.

The second safeguard that I think should be introduced and that would be very well worth further consideration would deal with the question of who approves the statement of funding policy. At the moment, it is simply a statement of Treasury policy. It is issued internally by the Treasury and is not a formal command paper that is formally laid before the UK Parliament. While there is obviously consultation with devolved Governments about it, there is no direct mechanism for devolved Governments to endorse or approve it before it takes effect in a spending review. Perhaps that is a reflection of the fact that the spending review process is a process within one Government. Equally, Northern Ireland would not expect that its budgetary process would be subject to that sort of oversight by another Government. Nonetheless, there needs to be some further thought. This document is no longer one that is purely for one Government, but it is potentially of constitutional significance.

One suggestion that I have heard made before by the Scottish Finance Secretary, John Swinney, is that devolved Finance Ministers should be responsible for approving it. At the moment, to the extent that there is political approval on behalf of devolved Governments, it comes from the territorial Secretaries of State, who are, of course, Ministers in the UK Government and are bound by UK collective responsibility. I suspect that the shift to having devolved Ministers endorse the detail of that would make a significant difference to how it is framed and the process of settling it and agreeing what goes into it.

The Chairperson (Mr McKay): The question that pops into my head is this: what was in place before Barnett, and did it work?

Professor Trench: There was something called the Goschen formula, which was named for a 19th-century Conservative Chancellor of the Exchequer. Essentially, that applied a population proportion of public spending in England to Scotland, Wales and, as was, Ireland, when that was appropriate.

The Chairperson (Mr McKay): Do members have any other questions?

Professor Holtham: The formula is very much devolution

accumulated more funds in their areas. At the early stage, expenditure was driven by the same geographical formula as in England, so in Wales we have adopted

needs-based formula.

which the spending ministries in England now regard differently

The Chairperson (Mr McKay): Professors Holtham and Trench, thank you very much. We may follow up on some of our queries in writing. I have found the session to be fascinating and a good start to our report, and I think that other members would agree. Thank you very much.

Professor Trench: I am pleased to have been able to help.

Professor Holtham: Thank you.