Good morning and welcome to the 14th meeting in 2015 of the Scottish Parliament’s Finance Committee. I ask everyone present to turn off any mobile phones or other electronic devices, please.
Our first item of business is to continue to take evidence for our inquiry into Scotland’s fiscal framework, and we will hear from two sets of witnesses today. I welcome to the meeting the following Scottish Government officials: Mairi Spowage, head of national accounts, and Simon Fuller, deputy director, office of the chief economic adviser.
Members have received a briefing paper from our witnesses so we will go straight to questions from the committee. Normal procedure is that I will ask some questions, then open out the session to colleagues around the table.
The first sentence in the Scottish Government’s written submission says:
“Economic statistics are an essential source of information for policy makers and government, parliament and the media, and academics and researchers.”
I am sure that everyone here will agree with that. However, numerous witnesses to the committee in recent weeks and, indeed, over past years have expressed concerns that we do not have enough high-quality data. How would you remedy that?
Thank you for giving us the opportunity to talk about the economic statistics that we produce. There are challenges in producing comprehensive economic statistics for Scotland. The national accounts, which my area of responsibility covers, are indicators that describe the Scottish economy—things such as gross domestic product and the flow of goods and services around the economy.
Historically, not many indicators about the Scottish economy were produced. For many years, services and production have been indexed, but it was only post-devolution that people started to bring those things together in a formal GDP measure. Seven years ago, the Scottish national accounts project was established to try to fill in some of the gaps in the national accounts data that are available for Scotland. That has come a long way, although there are still areas for further work.
We now publish a new bulletin that tries to bring together all the different data sources, to tell us a lot more about the Scottish economy. We have a lot more indicators, such as savings ratios and estimates of imports and exports and capital investment in Scotland. We publish detailed information about specific sectors of the economy, such as manufacturing exports and retail sales. Elsewhere in the Scottish Government a lot of information on the labour market is published, as well as statistics on income and poverty and business.
Compared with other parts of the United Kingdom, Scotland is pretty well served with economic statistics. The Scottish national accounts project’s aim was to produce a range of indicators, so that we would have a full set of national accounts for Scotland. By no means are we there yet; there are certainly a lot of gaps.
In a lot of cases it is not straightforward to do this for Scotland. We are not able to collect financial information from firms, because their accounts do not break down their UK operations. It is very difficult for large firms to provide a Scottish figure for their operations. It is a particular problem for trade, because firms do not record statistics in the required way. For example, exports from Scotland to the rest of the UK are particularly challenging to measure, and measuring imports from the rest of the UK to Scotland is even more challenging.
We are continually looking to improve the coverage of economic statistics. It is important to stress that we do not set our priorities by ourselves; we have a lot of engagement with external users. I welcome this engagement to get your thoughts on things that we should prioritise. We are lucky to have an active external user group, which is made up of academics—indeed, it includes some of the witnesses that the committee has had to talk about this subject.
Recently we had a meeting at which we said that we would prioritise improving data on imports to Scotland and on capital investment in Scotland, looking at the links between the offshore and onshore economies and looking at the timeliness of our key outputs, to see whether they can be improved for users.
We are focusing on those things, but it is a long journey. We will always have challenges with the data not being as good as it would be for the UK, because businesses just do not structure their accounts in the necessary way. That is the challenge that we face.
Surely there are some areas in which there can be improvements. For example, the Scottish Parliament information centre has said:
“a number of important pieces of economic data are either incomplete or missing.”
You have touched on some of those gaps. SPICe also says:
“This missing data limits our understanding of how Scotland’s economy works.”
You have touched on the Scottish national accounts project in your submission, but there are other areas that you do not touch on, such as tourism. SPICe says that the indicators are
“based on a relatively small sample of data from the International Passenger Survey to estimate tourism expenditure in Scotland and assumptions are made on the value of tourism from the rest of the UK.”
Surely you must be able to improve the data collection in such areas, given that tourism is the biggest employer in the private sector
I want to draw a distinction between estimating tourists’ expenditure and estimating the activity of tourism-related industries. We have very good information about, for example, the accommodation and food services industries, as well as other industries that are key for tourism. Therefore, the activities of tourism-related industries are well understood.
We are quite often asked a question about the tourism industry, but it is not defined in that way in the national accounts. Tourism is important for various industries. The accommodation and food services are two of those industries—the hotels, restaurants and so on. As an example, our business statistics colleagues produce information on the growth sectors in the Scottish economy. They try to take together tourism-related industries to produce an estimate for tourism.
The indicator to which the SPICe paper refers is our estimate of tourism expenditure by tourists, which forms part of our exports estimate. As I said, we have a very good understanding of the activity in tourism-related industries. It is the estimate of tourism expenditure that we have said is only of fair quality, because the sample that the Office for National Statistics uses to estimate that is fairly small.
Let us go on to the “Government Expenditure and Revenue Scotland” publication, which, of course, comes out to great excitement every March. How can its accuracy be improved? For example, in areas such as defence, there is a UK assignation, with £3 billion assigned as the amount spent on defence in Scotland. However, the spend is more like £2 billion, so the figures make it look as though £1 billion more is spent here than is the case. Surely, given that it is in the Scottish Government’s control, there must be ways to improve the accuracy of the information in GERS.
We get asked that question a lot. GERS shows the expenditure for the benefit of the people of Scotland. After a lot of consultation with users, particularly during the document’s major review in 2007-08, that is the definition that is used to assign spending to Scotland in GERS. With a sector such as defence, spending is at the UK level, which the Treasury classes as non-identifiable spending in its analysis. Basically, the spending cannot be assigned to any part of the UK, because a lot of it is overseas and so on and the expenditure is seen as being for the benefit of the whole of the UK. The GERS approach is to take a population share of that spending. That is the approach that is used for defence spending, which is also the approach that is used for areas such as public sector debt interest.
So GERS is a combination of many different measures and there is not a consistent format or formula. You are mixing apples with oranges—actual economic output and expenditure with assigned expenditure and, in the case of defence, theoretical expenditure.
Given that activity by the Foreign and Commonwealth Office, for example, can often be overseas, it is seen as being for the equal benefit of the whole of the UK, so that is how it is assigned in GERS. Defence spending is also in that category.
The approach that is taken in GERS is based on expenditure for the benefit of the people in Scotland. For example, some spending on museums and so on in Scotland can be seen as being for the benefit of the people in Scotland. However, because we get a number of visitors from the rest of the UK, it is also for the benefit of the wider UK. We adjust the spending similarly. It goes both ways, but it is the principle of “for the benefit” that applies, which we explain in our publication.
We are always happy to take feedback on whether we could improve our methodology. It is helpful to get the feedback that you feel that it is an inconsistent approach.
I think that it is. When GERS comes out, we argue about the extent of the Scottish deficit or the lack thereof, whereas in fact the deficit is assumed as opposed to being real because no Government would spend the amount of money that is allocated in GERS. It does not seem substantive in terms of the detail. If we are going to have such a document, surely the figures should be more robust, rather than being extrapolations that are not real?
You are more or less saying that it would not matter if zero was spent on defence in Scotland, because Scotland gets a benefit, arguably, from the amount that the UK spends on defence—certain aspects of that would be arguable. It does not make any odds whether £1 billion, £2 billion or £3 billion is spent in Scotland, given that the figure that is used is a percentage of the UK spend. Therefore, the GERS expenditure cannot be seen to be accurate.
I see you narrowing your eyes there, wondering what I am talking about, but what I am trying to say is that, by your explanation, whether £1 billion, £2 billion or £3 billion is spent on defence in Scotland it has the same impact on the GERS figure, because that figures is calculated as Scotland’s share of defence spending. Is that the case?
In a GERS sense, yes.
Okay, fair enough.
Let us move on to something else. As further powers are devolved, is it appropriate that economic data, such as GERS and GDP, continue to be produced by the Scottish Government? Should it not be produced independently of the Government, by a body such as the ONS, which publishes equivalent data at UK level?
Statistics such as GERS and our GDP figures are produced by statisticians in the Scottish Government and they are regularly assessed by the UK Statistics Authority. GERS in particular is assessed fairly regularly and the most recent assessment was finished just before publication in 2014. The assessment from the UK Statistics Authority is to check that we are producing the documents according to sound methods, consulting users and meeting their needs, that the statistics are reliable and that they are managed objectively and impartially in the public interest. The national statistics badge is there to reassure users that the figures are of a certain quality and have been managed impartially and objectively.
As statisticians in the Government, we work under the code of practice for official statistics, which sets very strict guidelines about how we produce our statistics and how and when we share them with colleagues and ministers.
I have a couple of brief points before I open up the session to my colleagues around the table. In its recent review of GDP, Fiscal Affairs Scotland suggests that a Scottish consumer prices index would be worth collecting in order to improve the economic data. How do you feel about that?
09:45
The Fiscal Affairs Scotland analysis was looking at the overall level of price movements that was implied by our real-terms GDP growth and our current price cash value of GDP. There are differences in how those things move relative to each other for Scotland and the UK. That measure is called the implied deflator; it shows overall movements in aggregate prices for the Scottish and UK economies. Those are different for our sets of measures, and that is what Fiscal Affairs Scotland was analysing.
The very fact that they are different shows that they are capturing the different movements and prices. Basically, we build up a very detailed series level, using either what is called a volume measure—for example, for electricity supply, we would use the production of electricity in gigawatt hours—or a deflated turnover measure, using the most appropriate deflator, which can be a UK deflator. It is a very detailed series level.
When all that is built up for Scotland, the aggregate movement in prices is different from that for the UK, because of the different industrial mix. The very fact that those movements are different is captured in the way in which we build our statistics. We follow international guidance, which means that, if we do it at a detailed enough series level, UK prices can be appropriate to use for Scotland. Different price movements that are shown by our series are reflective of the different industrial structures.
The real question is whether the detailed price indices that we are using for each series are the most appropriate ones. We look for the most appropriate price index to use on a case-by-case basis. For example, a subset of education is things such as driving schools. We would never use a UK education deflator to turn that from current to constant prices, because it is completely dominated by English tuition fee prices, so that would be an inappropriate price index to use. We use a deflator that is more appropriate for the personal spending on things such as driving schools. We look for the most appropriate deflator.
Do we have a set of Scottish price indices to choose from? No, we do not, because we do not collect the information that would be required to produce those. Doing so would incur a huge cost. However, given that we are doing it at such a detailed—over 100—series level, and the guidance from Eurostat is to use at least around 30 or 40 industries, we are confident that it is capturing correctly the different price movements in the Scottish and UK economies.
Thank you for the detailed answer.
My final point is about the proposal to assign half the VAT raised in Scotland. It is unclear how that will be calculated. Can you give us a view of your thinking on that?
It is a challenge to measure that for Scotland, as it can be for a lot of things. The approach that is taken for GERS is reasonably broad brush, although it can be seen as fit for purpose for a statistical publication such as GERS. However, using that approach to assign budgets is a completely different matter.
At the moment, the VAT estimate is built up of a few different sections, but the majority of it is based on VAT being paid by the final consumer—by people going into shops and buying things. It uses a measure of household consumption. There is a survey of households that asks in a very detailed manner about the things that they buy, which is used to estimate the share of VAT that should be assigned to Scotland for the household sector. At the moment the survey is of 500 households a year for Scotland, which is, in statistical terms, a very small survey. Therefore the sample is pooled over three years to get more stable results. There are a number of issues with that. The responsiveness of the survey to change will be reduced and the sampling variability is likely to be quite large. Obviously, those are not desirable things if this is being used to determine the budget.
This is definitely an issue that needs to be looked at. We are starting a programme of work with our colleagues at HM Revenue and Customs to examine ways of improving the measurement of VAT, and they are just starting an audit of the data sources that can be used in order to measure it better and are considering whether there needs to be an increase in the sampling for this survey or whether there are other ways to measure it more precisely.
I am tempted to ask further questions on that issue, but I want to open up the session to colleagues.
Good morning. In an earlier answer, you mentioned that the initial objective of SNAP was to get a full set of national accounts, and you said that we are not there yet. Obviously, the situation has evolved slightly since 2008 but, as far as the initial objectives are concerned, do you have any sense of when we will get there?
It depends on what we class as the end of the project. If you are asking about the things that we do not have, at the moment there are measures in our accounts that are not of gold-standard quality—things such as measures of imports and so on—but the beauty of the framework is that it allows us to estimate the things that we do not have primary data for, because of the way that we are able to use data around supply and demand in the economy to balance everything.
We do not have sectoral accounts for all the different types of corporations in Scotland, the Government sector and the rest of the world, and getting them would involve a big development programme. Particularly challenging is the issue of the balance of payments with the rest of the world, which involves the need to measure flows of income in and out of Scotland and transfers in and out of Scotland. To be honest, although we can develop estimates of those things, they will only ever be modelled and estimated, because we do not have that data about what is flowing across the border or the information about what is raised in Scotland.
I can give the committee more detail in a follow-up letter about all the items that are missing from the set. If you are asking whether we are halfway there, I would not say that we are. We do not have most of the things that are in a full set of national accounts. Part of that is about data and part of that is about the number of people that we have working on it.
SNAP was set up in 2008. Since then, there has been the Calman commission and the taxes that have been devolved through that, and the Smith commission, through which further taxes will be devolved. How does SNAP have to evolve and what changes are going to be prioritised in order to take into account the taxes that have been and will be devolved?
The estimation of various taxes, such as VAT, becomes more important if they are likely to be devolved, and we will look at how they have been estimated in the past and determine whether there need to be improvements. Obviously, the situation with regard to taxes that are going to be collected in Scotland is somewhat easier, because no estimation is needed. However, we need to think about what we are publishing with regard to those statistics and whether our users are getting enough information about the revenues that are being collected in the context of all the other taxes that are raised in Scotland, which we are estimating. We will need to think about how we publish revenue information and how to make it more accessible for users.
There is a lot of interest in the trade aspect and how much we can improve that. There is also a lot of interest in the links between the onshore and the offshore economy. Those are the things that we are prioritising this year. Simon Fuller might want to add to that.
I would like to add a couple of things. Obviously, the Scotland Act 2012 and subsequently the Smith commission have very much guided the priorities for what we look at within our team. For example, on improving the estimates of taxes, over the past few years we have put a lot of work into improving the estimates of air passenger duty, because that is a tax that is coming to Scotland and there is great interest in it.
The same is happening with income tax, responsibility for which is coming to Scotland in a staged manner through the Scotland Act 2012 and the Smith commission. We already have a good estimate of the overall level of income tax in Scotland, so we have been putting a lot of our resources into understanding the detail. For example, we know that total income tax is £11 billion, but we want to know how that is distributed across the income distribution and how much is paid by basic, higher and top rate taxpayers.
A lot of our focus has been on getting much more data and working with HMRC to improve our wider analytical capacity to inform the broader analysis that will be required when these new taxes are devolved.
Thank you.
I have a question about GERS. It comes out every March and there are revisions to the previous year’s GERS. On this year’s GERS, the revisions seemed to be more significant and more marked than in previous years. Is that correct? Is there a particular reason for that or could it continue in the future?
It is correct that the revisions were greater than one would normally expect. The main reason was the transition to the European system of national and regional accounts 2010, which led to a number of changes in treatment. Unfortunately, that had a particularly significant impact on 2012-13—the most recent published year—because of the inclusion of things such as the purchase of the Royal Mail pension scheme, which was a large one-off that changed the figures quite a lot. There were a number of different changes that we detail in the document.
I would not expect that level of revision every year. It occurred only because there was a major change to the system of accounts, which we had to incorporate so that our statistics were still comparable with those of the UK.
So the changes feed though in one year and should not have an impact on future years.
That is correct.
The changes are similar to those that occurred at the UK level and across other countries. The revisions in Scotland were comparable.
Thank you.
I move on to a couple of specific taxes. In your submission, you talk about your modelling for the land and buildings transaction tax. Can you tell us a little more about that model and whether you plan to make any changes to it, based on the comments of the Scottish Fiscal Commission?
The model for the land and buildings transaction tax is built up from detailed information on virtually every housing transaction in Scotland over recent years. With LBTT, we are very lucky that we know the detail on a per transaction basis—we know selling prices, dates of sale, locations within Scotland and so on. The model takes the historical data on levels and distributions of transactions in Scotland and seeks to project that, based on trends over recent years and how we expect those to evolve in the future. For example, if transactions just now were well below the long-term average, it might well be reasonable to expect a gradual uptake to return to long-run levels.
We take that information and combine it with wider intelligence and modelling of what is happening in the Scottish economy. All else being equal, if we expected the Scottish economy to grow strongly over the coming years based on what our wider economic models suggested, we would look at the historical relationship between GDP growth and housing transactions and prices and incorporate that into our model. We would then model how we expected transactions and prices to evolve in the future. From the expected bands and rates, we would derive what we could expect the aggregate level of tax receipts to be.
I have two points to add. The first relates to what you said about the Scottish Fiscal Commission. Last year was the first year in which we produced the forecasts, and we had a number of discussions with the commission about that and received valuable feedback. The model that we are using is subject to almost continuous improvement and evolution, partly to incorporate the commission’s changes but also because, as the economy changes, we need to focus on different parts of the economy and make sure that we are picking them up accurately.
My final point is to flag up the fact that, of all the taxes, land and buildings transaction tax—stamp duty land tax, as it was in the UK—is one of the more volatile, so it always presents challenges. The UK level halved during the recession. Despite that, we have a detailed modelling framework that is underpinned by a lot of raw data.
10:00
So you are making improvements. It was year 1, and the idea is to keep improving.
A specific point that the Scottish Fiscal Commission made was that there was no behavioural costing at all. The Cabinet Secretary for Finance, Constitution and Economy suggested that that was true.
You have static costing, which looks at the numbers alone. Behavioural costing asks what the likely impact is and how likely it is that the numbers will transpire. Are you planning to incorporate some behavioural element into the model for the next set of projections?
That is definitely something that we are looking at. In some ways, the behavioural part is much more challenging. That is partly because LBTT represented quite a significant diversion from the previous system. There was a move away from a slab tax to a more progressive tax. As we get more data and a better understanding of how consumers and households have evolved and responded to the tax, we will be in a better position to incorporate the behavioural changes. It takes time to build up the data and to be able to say with sufficient confidence, “This is what we can expect behavioural change to be, based on past experience.”
You say in your paper that the GDP figures are your most high-profile quarterly output. I know that you do not publish formal projections, but does your department have an internal model to allow you to compare your internal projections with what actually happens, or do you simply not touch it?
There are two points that I would make about that. First, in our paper we discussed some of the modelling that we are currently developing. We are in the process of developing a macroeconomic forecasting model for Scotland. Although the model is complete, we are still in the testing phase, so we cannot provide usable outputs yet.
Secondly, when we look at such forecasts, especially a quarter ahead, we would not normally use a big, detailed, macroeconomic model. We would probably use a data-driven model. We would take all the data sets for the next quarter that we already had, which might be from the labour market or from business surveys, for example. We know that, in the past, those have broadly correlated with GDP growth in a certain way, so we can infer from that where, in broad terms, we can expect GDP to go. That helps to inform what we are seeing in terms of the statistics and the actual GDP numbers. We very much use that to guide our process and to support our wider intuition.
That is helpful. There is work being done on that and it is experimental at the moment, so you are not in a position to publish. Is it the Scottish Government’s intention to publish GDP forecasts?
I would not like to prejudge that at this stage. It very much depends on the current programme of work and how the model appears to be performing. This is the first time that we have had such a model for Scotland, so it is impossible to say at this stage quite how it will be able to be used in future and how usable it will be for that specific purpose.
Perhaps we can look further at a few issues that have already been mentioned. The comment has been made that we have far fewer statistics and figures than other countries—including the UK as a whole—have, but that we do quite well compared with other regions or sub-states or whatever we want to call them. Are there good examples out there? Are there other sub-states—California, Bavaria or Catalonia, for example—that are doing better?
Not that I am aware of. We have done some work on that and it is quite tricky to find sub-states that are using supply and use tables, which are the framework on which we build all our estimates. A number of other countries estimate things such as economic growth at a lower level.
Does the picture vary a lot among countries? Do big countries such as the United States have fabulous data while little ones such as Luxembourg have much poorer data, or does it not work that way?
Across the European Union, all countries are required to produce the same sort of things and to give them to Eurostat. Exactly how they do it, how much resource they put into it and the quality of the statistics vary. Larger countries are likely to have a much bigger all-singing, all-dancing set of national accounts that is of higher quality. For example, the UK produces what it needs to do for Eurostat, but it also does many other things. For example, it needs to produce its GDP figures within 60 days, but it also produces a figure 25 days after the end of the quarter. In doing so, it is one of the fastest countries in the world. It does not need to do that for Eurostat and most other European countries do not do that, but the UK does because it is that much bigger and the number of people who work on the national accounts is much higher.
Is accuracy sacrificed because of the speed?
Yes. The amount of data that has been got in for the quarter at that point is quite limited—it is about 40 per cent of the data that will eventually be used to finalise the estimate. For the quarterly economic statistics, in particular, it is necessary to strike a balance between accuracy and timeliness and to decide what level of revision we are prepared to accept, because the figures will be revised going forward.
We have done some work with, for example, the Irish national accounts people, because I wanted to get a perspective from a smaller country with a national accounts unit that was closer to the size of ours. Ireland’s unit has about 40 people compared with our 10, but it is a little bit more comparable. It produces one GDP estimate per quarter compared with the UK’s three and it produces the estimate 70 days after the end of the quarter. It has negotiated that timescale with Eurostat, because that is the best that it can do. That is a more realistic target for us to aim for. I am not saying that that is what we will do, but it gives us hope that we could achieve such an aim—it does not seem quite so impossible.
So there is a range of experience out there.
Yes. It is very helpful to talk to other countries in Europe. We have a lot of engagement with the ONS, which is very helpful and gives us a lot of data—most of the data that we use to produce our national accounts comes from the ONS. However, it is good to get other perspectives and to find out about other ways of doing things.
You could obviously do more if you had 40 staff instead of 10.
Yes.
The other aspect that has been mentioned is that some things are legally required by the state. Who decides what is required? Is that totally reserved to UK level? Can the UK Government say, for example, to what level organisations have to break down their VAT figures?
No. It is all driven by Eurostat—by the EU, in other words. For example, all member states in the EU have to produce regional accounts, because the EU asks for them. The regional accounts are modelled estimates of UK figures for different areas of the UK. The figures go down to quite a detailed level—equivalent to local authority level—but they are very different from the statistics that we produce for Scotland, because we build them from the bottom up. There is a requirement to produce regional information. For example, there is the NUTS—nomenclature of territorial units for statistics—II level information. That sub-Scotland level information—there are four areas in Scotland—is what is used to determine European structural funding. That is how the EU uses that information.
The Central Statistics Office in Ireland estimates that about 90 per cent of what it does is done because it must do it, as it is required to do so by EU law. The ONS estimates that about 80 to 90 per cent of its work is required by EU law. The vast majority of the work of the national statistics institutes in the EU is driven by what Eurostat insists they must do.
There has been a lot of discussion about how company profits are moved around countries. It is doubtful whether we know what the profit, sales or VAT levels of some companies are in the UK, let alone within Scotland. Who tells an international company such as Diageo how to split up its accounts? Does the EU instruct companies on how to do that?
That is a slightly wider discussion, because there are distinctions between economic statistics, how those statistics are produced and how we disseminate them. The issue of tax payments and where the tax liabilities of a company’s operations fall is slightly separate and would normally be agreed between finance ministries. Members of the Organisation for Economic Co-operation and Development would perhaps agree common methodologies and accounting practices for how profits should be assigned, and such things are agreed at the European level as well. However, the agreement is not so much about the collection of statistics and data; it is very much an accounting and finance ministry matter—almost a revenue collection issue—rather than a pure statistical issue in the sense of the information being used to produce economic statistics.
Do you rely on some of those accounting figures for GDP and things?
A lot of our statistics are based on the business surveys that the ONS carries out. A number of companies put in those surveys similar information to that which is in their company accounts. That presents a challenge in estimating income flows, for example. Such information is relevant to estimating gross national income, because we are talking about where income is flowing. After stuff is produced and profits are made, where does the income flow to? It can be more challenging to measure that. Ireland has a lot of issues with that.
Even though Ireland is a sovereign state, it finds that issue difficult.
It is challenging.
I take it that companies or organisations have a legal requirement to produce the data. Is the power reserved to the UK to tell companies to produce data, if it wants to?
Yes. Everything is collected under the Statistics of Trade Act 1947. Companies have a legal obligation to fill in the surveys.
I am a little concerned about some of the things that we are told about in the SPICe paper. The imports figure, which you mentioned, seems to be incredibly vague. We are told that
“the remaining difference is assumed to be imports”.
Is the UK—indeed, any state—clear about what its imports are, or is that vague as well?
There is very good data from HMRC on goods—stuff that moves across the border—and we can exploit some of that data. We do not get access to a lot of the data from HMRC, but we can use a lot of data on international imports and we are looking to exploit that. More challenging for us is getting data on imports that come from the rest of the UK, because there is no official border.
Services are a tricky area, even for countries such as the UK. The UK conducts surveys such as the survey of international trade in services to get a sense of the services that are flowing across the border, which are less tangible than goods leaving or entering the country.
Even the UK and other countries are struggling with that.
Yes—with services, certainly, and that is the vast majority of the economy. The issue is challenging.
The definitions of imports and exports in the national accounts have changed since the introduction of the new system of accounts. For goods, there is supposed to be a change in ownership—it is not just about their flowing across the border—and that complicates everything, because it increases the challenge of measuring such things. That change has not really been implemented across Europe yet, because countries cannot figure out how to do it.
Would it be a problem if a company—Volkswagen or something—was moving stuff between branches?
Precisely. If a company sent a car to Scotland to be sprayed and then sent back again, there would be no change in ownership. Officially, under the new national accounts, that would not be an import and an export; it would be just an export of a service to the country. That complicates everything massively.
Another area that seems vague in the SPICe paper is capital investment. The paper states that the public sector data is pretty reasonable but that the private sector data is a bit more difficult. The assumption seems to be that profits equal capital investment.
10:15
That is not our strongest set of data. There are definitely issues with it, and one of our priorities for this year is to look at it.
Capital investment can be lumpy, and there is no requirement on companies to tell the ONS where they are investing money in the UK. There has been a lot of pressure from us and from regional accounting people in the ONS for more information to be collected for really large bits of capital investment by asking people, “Where is the building that you are building? What is the location of this capital investment?” If that approach is approved, it will improve our estimates quite a lot. We are also looking at whether we can exploit more the data that the ONS collects from Scottish companies.
We have to prioritise, because we could do so many things, but capital investment is one aspect that we need to look at this year. We can improve things even with the data that we have, but we will do even better if the more detailed data is collected from companies about big bits of investment. If we can get the big stuff right, we will be doing pretty well.
The final issue that I will touch on is VAT. We have discussed whether, if we get a share of VAT, it will be just what the customer pays in the shop at the end or whether it will also be based on the point at which the factory increases value, even though the end sale might be elsewhere, because VAT is based on each step of the process. Would it be more difficult to get data in the latter scenario, because it would come from different places? Is it easier to get data or statistics from the simpler approach, which is based just on sales in shops and so on?
Having consumption as the basis is definitely the simplest way to approach the VAT that the household sector pays. We also have the exempt sector and Government. Government is obviously simpler, but it can be challenging to make proper estimates for the exempt sector, such as for financial services and charities. There are a lot of challenges, but we will work closely with HMRC on this.
That is very helpful.
Mark McDonald has a supplementary question.
Yes—I have a brief supplementary to ask before every question that I had written down gets taken away. [Laughter.]
On capital investment, you spoke about the difficulty of determining what is being developed and where. I presume that there are ways of collecting that information. For example, local authorities issue building warrants for premises that are being developed and business rates bills for newly developed properties. Are you looking to collect information from that? Rateable value will give you some indication of a development’s value. Are you looking into collecting that data?
Yes. We are starting an audit of all the different data sources that we could use to improve our estimates, so we will be looking into the subject.
Over the past couple of years, we have become aware—I certainly have—of a lack of information at the Treasury and HMRC because they have never really needed to show the difference or produce figures specifically for Scotland. Are they working towards that more and more? Is that getting easier? Are you part of that?
We have been working quite closely across the devolved Administrations and with the Treasury, HMRC, the ONS and the Cabinet Office. In the statistical environment, we have been thinking about the data that might be required as a result of the devolution of further powers to all areas of the UK.
The focus is on looking at solutions that can be future proofed, so they might not be for just one area of the UK. If we are looking at improving information, let us think about how it could be used across the UK. We are thinking about things for the English regions as well.
We had a big discussion about that and a group was formed to discuss, on the basis of current proposals, the work programmes that might need to go ahead. That seems to be high up the list of what the Treasury, HMRC and the ONS are thinking about for data. In particular, the ONS seems to be prioritising the development of statistics for different areas of the country. I would like those bodies to collect lots more data, which would give us more finely grained estimates about activity in Scotland. I am not so focused on the publishing of statistics, but other users would want that.
The process has been positive and has raised a number of issues. We need to see what will happen next, once the further powers are firmer.
You talked about the information that is required to go to Europe. Could you produce that as an exercise to show where any gaps would be and so on, as if we had to make a return for Scotland independently?
Yes—that would probably be how I would approach that. I could do that for the committee, if that would be helpful.
In relation to the Smith commission, there are a number of areas of our economy where we are responsible for parts such as part of income tax—the Scottish rate of income tax—and we have the potential to have 50 per cent of our VAT back. When considering what that really means, one imagines all the complications at the back of that.
In your opinion, would it be easier, or would we get much greater information. if we were responsible for collecting our own VAT? I am aware of the complication with Europe, and VAT has to be collected. On a regional basis, however, would it be easier to collect VAT alone, to get the figures here and then to deliver them to HMRC, rather than try to do something fairly complicated?
Both approaches have pros and cons. Under the approach that you just outlined, one potential challenge is that, in some cases, it might well be easier to collect or identify the taxes paid at UK level, if all the companies were in effect operating UK-wide. For other companies, it might well be easier to do things with a Scottish sub-sample.
To be honest, we have not thought about the question in any great detail, but I can see pros and cons to both approaches. The issue is not clear cut, especially given that control of the tax will remain reserved.
You said that one of the difficulties that you had with some of the stats about businesses was establishing where they did business. Would the approach that I described clarify that much more, if there had to be a return for a business that was in Scotland?
If firms had to do that, it would improve the statistics, but there would be downsides, with a burden on business and the need to separate things out, as well as the administration costs of collecting in Scotland. From a purely statistical point of view, I would love all businesses to say what they do in Scotland—that would be marvellous—but is that benefit enough to justify the cost to business, the potential collection costs and so on? It is a question of balance.
The convener mentioned the importance of tourism and how figures for it are calculated. I think that I am right in saying that the figures that are calculated to show how many people work in tourism go across a range of businesses, which may all sit in other sectors. For example, a percentage of the retail sector is taken into tourism. It is important to keep those things separate for the purpose of getting a picture from across Scotland of what things look like. “Tourism” is a kind of umbrella term.
Yes. The term cuts across a number of industries, which is why, in national accounts terms, it is not an industrial classification—that involves the types of goods and services that an industry provides, rather than who consumes them.
On the other hand, analysis such as the growth sector analysis that my business colleagues produce can consider all the sectors of the economy that tend to be dominated by tourist consumers, and it can bring them together to produce estimates as you describe. Those are different products for different purposes.
During the referendum campaign, a number of claims were made about the amount of business that goes from north of the border to south of the border but, from everything that you have said, it seems that those claims were vague, because such statistics do not exist. Is that right?
Which statistics do you mean?
Mairi Spowage mentioned that it is easier to know what our exports are to the rest of Europe or the rest of world than it is to know what they are to England. My point is that, although some facts and figures were produced during the referendum campaign, I wonder how easy it is to be exact about the size of our market in England as a percentage of our exports.
We are in a better position for exports than we are for imports. We do a survey of businesses called the global connections survey to estimate exports to the rest of the world, to individual country destinations and to the rest of the UK. That is a key data source for us, because it shows the market that exists for Scottish goods in the rest of the UK. Even though that is not a statutory survey—we do not collect that information under the Statistics of Trade Act 1947—we get a good response from businesses, which is quite heartening, and we bring in a number of other data sources to improve the estimates.
We are all right on exports—we get good data, although we would always want more—but the situation is more challenging on imports. That is why we use the framework of the national accounts to estimate that information.
That leads me to my question, which is on import data. You have talked generally about the importance of having enough data to inform your work. In response to Mr Mason’s questions on import data you said that, although a lot of information on import data is available to you from HMRC, a lot of information is not available to you. Why is that data not available to you? Is it because it has not been collected by HMRC or is there a problem in accessing it? Why should that be the case? I presume that, at all levels of Government, importance is attached to ensuring that you have all the data that you need.
Yes. In the past, we have had difficulty in getting access to some data from HMRC for a number of reasons. Some of those reasons have been to do with legislative issues. The ONS gets access to data called the intrastat and the extrastat, which are a record of the information that businesses have to give about their goods that cross borders. The intrastat data is on goods that cross borders within the EU, and the extrastat data is on goods that go outwith the EU. That is a key source of information for the UK’s balance of payments, but we do not currently have access to that information, nor do the other devolved Administrations.
What reason is given for that?
I would have to check exactly what we were told the last time we asked.
It sounds as though it might not be a terribly good reason.
To get access to HMRC data, the business case has to be very strong. The national statistical institute needs to give balance of payments data to Eurostat, so it has a very strong case for getting that information.
Given the importance that is attached to developing a fiscal framework for Scotland, I think that we need to return to that issue. It is essential that the Scottish Government has access to the appropriate level of data.
I absolutely agree.
A UK bill on data sharing was initiated in the most recent UK Parliament and it is still in its early stages. The idea of that is to allow HMRC to share more data with the ONS and other departments. I hope that that route will enable more data sharing to happen and that the barriers that HMRC feels exist to the sharing of data with other UK Government departments will be removed.
10:30
It is surprising that legislation is required. We will need to pursue the issue with both Governments, and with the UK Government in particular.
I have one final question. In your submission, you mention
“Further development of our Oil and Gas Statistics Publication”.
What progress are you making on that? What level of detail do you presume to go into for that publication, and how regularly will it be published?
Our publication is not yet a quarterly bulletin because it is still in the early stages of development, but it contains information on production and indicative sales revenues for different types of products in the North Sea. It is a quarterly series that shows all the different products and the sales revenue that might be generated.
We also produce indicative trade flows. Because our principal GDP measures, trade flows and so on are based on onshore Scotland, we add the offshore component to show overall trade balances. Our plan is for the series to become a more established statistical publication.
It will be published quarterly on a full basis.
Yes.
Most areas have been covered, and I have found what you have said very interesting. I have three follow-up questions, the first of which may have been partly covered by your previous answer.
The new priorities that you have listed include capital investment and imports, which you have covered, and the relationship between the onshore and offshore economies. I am interested in that last priority. I take it that there is more to it than what you described in your previous answer.
Yes, that would be an additional piece of work that would involve looking more at our supply use framework. It would cover all the different products and industries in the economy and all the little links with the offshore economy. That would give us a better understanding of the supply chain and what the offshore economy is buying as well as the impact of changes in the offshore economy on the onshore supply chain. It would be a more detailed piece of work, and we will look at that in developing our next set of supply use tables. Our users are quite interested in information on the total contribution of the offshore economy to the onshore economy.
My second point—again, perhaps I am missing something—is that you keep saying that GERS is for the benefit of the people of Scotland.
Yes—the expenditure in GERS. That is not necessarily what is spent in Scotland, but what is spent for the benefit of the people of Scotland. If money has been spent abroad on a UK reserved matter, the population share of that may be taken because it is seen to benefit all citizens of the UK.
Is that a justification for the defence issue? From that point of view, if you argue that all UK defence expenditure is for the benefit of UK citizens, it does not matter what is spent in Scotland.
That is from a public finance point of view, but for GDP and employment the question is about what is produced in Scotland and employment in Scotland.
Would that be a brake on trying to find the answers to some of these questions? It would change the rationale for doing it. One could argue that, to know the benefit to the people of Scotland, we do not need to find out how much is actually spent in Scotland. That is a contentious remark, but somebody could make that argument.
There is a distinction in terms of what we are trying to achieve with the data. When we produce the GERS publication for public finances, we try very much to be consistent, where possible, with the wider international standards for public finances. That is tricky, because the elements of where spending takes place and where it is for are not covered by that framework. Nevertheless, we try to be as consistent as possible.
If you were interested in, for example, how much defence spending occurred in Scotland—which is itself an interesting question—that would be more about the economic impact of the spend. It would be a wider economic question, and we would factor into our GDP estimates the defence spending that occurs in Scotland, for example, because, in producing that data, we are interested in what occurs specifically in Scotland.
Mairi Spowage will correct me if I am wrong, but I believe that the defence part would be specifically the defence spend that takes place in Scotland and the wider economic activity that is associated with it. There is a slight distinction—it really depends on the purpose of the exercise.
Okay. That is interesting. I will move to my third point. We are all interested in VAT just now because of the Smith commission proposals. It was quite alarming to see that the survey of 500 households used such a small sample. Nevertheless, on the basis of that survey—which may have been around for a year or two, I do not know—first, what comparisons can you make between Scotland and England in household expenditure and therefore in VAT? Secondly, how has that picture varied over the past few years?
Over the past few years, the Scottish share of UK VAT has been fairly stable. The share that is taken at the UK level has not moved that much.
I do not have the England figures or the four nations figures in front of me, but I can send that information to the committee. HMRC produces figures for the four parts of the UK, so I can certainly show you some figures on that.
Okay—thank you.
We appear to have exhausted questions from members of the committee. I have just one or two questions to follow up some of the questions that have been asked about HMRC data.
Will you have direct access to HMRC data on the Scottish rate of income tax? You talked about some of the sources and about data coming through the ONS. Why would that information come to you through a third party instead of directly?
We are still discussing with HMRC precisely which data on income tax we can get access to. I am not personally involved in that directly, but I understand that there are particular restrictions on HMRC’s ability to share data on individual taxpayers.
When data is anonymised it is much easier to share, whereas when it contains identifying information and suchlike there are, once again, legislative difficulties with HMRC’s ability to share it not just with us but with the Treasury or with anyone else. My understanding is that the Treasury would not be able to get access to the raw income tax data collected by HMRC for the same reasons, which are to do with confidentiality more than anything else. We have on-going discussions with HMRC about how we can get anonymised data.
When you get that data, will the Scottish Fiscal Commission have access to it?
I assume so. The discussions are on-going and may include issues around how the data can be shared more widely, so I would not want to prejudge anything. Our starting point is that we would give the Scottish Fiscal Commission access to all the data that it needed.
It would be good to get clarification of that, if possible.
Those are all the questions from me—I do not want to return to the subject of GERS, although I am sorely tempted. I am thinking about all those great mandarins in Whitehall who are benefiting Scotland in terms of our GERS figures, and of the money that is spent there, but we will not go into that.
Do you have any further points that you want to bring to the committee’s attention?
No, I do not think so. Thank you.
No.
I thank you for answering our questions, and I thank colleagues for asking them. We will have a five-minute suspension to allow for a change of witnesses.
10:38 Meeting suspended.
10:45 On resuming—
I am sorry for the slight delay, which was due to our previous session overrunning, such was the committee’s interest in the topic.
We will continue to take evidence on Scotland’s fiscal framework and I welcome to the meeting our next panel of witnesses: Alan Bermingham of the Chartered Institute of Public Finance and Accountancy, Charlotte Barbour of the Institute of Chartered Accountants of Scotland and Patrick Stevens of the Chartered Institute of Taxation.
We have received your excellent written submissions, so we will move straight to questions. I will open with a few introductory and exploratory questions and then my colleagues will drill down to greater depth or ask completely separate questions, as they see fit.
Let us start with the Institute of Chartered Accountants of Scotland.
That is me.
We may as well start with you, as you are seated in the middle. Do not look so shocked—you knew that you would be asked something at some point. [Laughter.]
As I said, all the submissions are excellent. I will start with the response to the first question that was put to ICAS, but before I do that, I make it clear that if other witnesses want to comment on the issue that I am raising, they should indicate that they wish to do so. I am not just asking about your own specific submissions; rather, I want the session to be more interactive.
ICAS states:
“Ensuring fairness and effective mechanical measures involves significant analytical and statistical input. Fairness may require more detailed calibration of different balancing elements in the Barnett formula adjustments, but the greater the analytical detail and complexity of arithmetical tax adjustments the more this is likely to reduce transparency. A balance needs to be found between these conflicting aims.”
The $64,000 question is how we achieve that balance.
The short answer is that I am not sure. One reason why we said that and have not given you an answer is that we do not think that there is a clear-cut answer.
VAT came up in the discussion that you had with the previous witnesses, and one of the difficulties with VAT is that there is not a full range of evidence. I accept that, if you really wanted to do something, you could do it. You could pinpoint Scottish VAT if you put your minds to it, although there would be quite a lot of work for civil servants or the business community. When it comes to borders, VAT has its intricacies as it is. There would be a lot of work to be done somewhere and somehow if you wanted to identify Scottish VAT. I realise that that is where one wants to get to, but if the work is awfully detailed and it then goes through the no-detriment Barnett calculations, we will begin to lose sight of where we are going.
I do not know quite where we strike the balance between the two. If we take 8 per cent of the economy, that is just a variation on the Barnett formula.
I saw you nodding, Mr Stevens. Do you have anything to add?
I suggest that there are so many calculations that one could do—each of which will follow bits of logic that economists, which we are not, come up with—that the bottom line is that nobody will be able to get into their heads what has been achieved. Part of the process needs to be to explain to people how the Barnett formula is arrived at, as well as the adjustments to it—and there are so many choices for adjustments—that are set out in the command paper. I presume that what we most want to be able to do is to explain why the result that we have reached is fair, but “fair” is a difficult word.
In conversations that we have had about adjustments, the starting point seems to be that not everyone fully understands how Barnett works in the first place.
Yes. We have had a great deal of evidence to that effect, as I am sure you can imagine.
I am sorry. I will not say that again.
Mr Bermingham, would you like to make a comment from CIPFA’s perspective?
Yes. The Smith commission report seems to allude to a simplified route, by which I mean that there will be an initial adjustment to the block grant representing the tax forgone in the UK, and then some form of indexation after that amount, which would then be fed into the Barnett formula.
There is great difficulty in measuring in this area, but a simplistic approach could be taken, as with the income tax proposal in the Smith recommendations, which suggests that it should be indexed by the general growth in taxation or whatever in the UK. There is plenty of evidence from OECD analyses about how tax-sharing arrangements work in other countries, so there are opportunities to consider that kind of simplified approach. It is probably an area in which the Scottish Fiscal Commission or the Joint Exchequer Committee could agree with the Treasury how the approach will work.
I will switch between your submissions as we go on.
Mr Stevens, your submission comments on the no-detriment principle and says:
“there needs to be agreement about what constitutes ‘no detriment’.”
What do you feel should constitute no detriment?
It seems to me that the first year’s adjustment will be relatively straightforward. I am not suggesting that it will not entail lots of work and calculations and so on, but it simply involves moving lumps of money from one side of the border to the other—I am sorry if that is a bit simplistic. In the years after that, the questions of how you should index or adjust or move in accordance with the economies of each piece of land, how they should interact with one another and how those issues will get you to a position of no detriment will be extremely difficult.
I am going to concentrate on tax, because I am a tax person. The changes in tax take in Scotland will not simply be a result of changes to rates and thresholds; they will be a result of changes in the economy and people’s behaviour. How you will split those out in order to stay with the principle of no detriment will be a challenge, and it will increase the longer the situation goes on.
Mr Bermingham, do you have a view on the issue?
I look at the issue of no detriment in relation to two areas. One is the tax side and what gets shared through departmental expenditure limits through the Barnett formula. The other concerns what is outside the Barnett calculation, such as any devolved welfare spending, which is annually managed expenditure and is demand led. That will be a more difficult area to index.
I agree with my colleague that the initial adjustments will be fine. With regard to the Barnett formula, there is a need to come up with an appropriate indexation measure, particularly in relation to tax and so on. There are relatively simplified tax-sharing measures out there that you can avail yourselves of. The issue concerning AME is more difficult. If the growth in welfare spend is greater in Scotland than in the rest of the UK, how will you index that, given that it is not driven by the Barnett formula? The only thing that I can say is that you would need to look at what primarily drives welfare spend and agree a measure with the Treasury as the way to index that.
What constitutes no detriment for ICAS?
ICAS does not have a view on what exactly constitutes no detriment, but it has concerns about how it will eventually be defined. In year 1, it could be quite high level and pitched in broad terms. However, I have heard quite a few conversations where it has been taken a lot further. For example, if there was a radical change to personal allowances in the UK, that would have a knock-on effect on how much would be collected in the Scottish rate of income tax. Would that be considered to be no detriment or, because the Scottish rate of income tax is a partially devolved tax, would it be considered that there had been a joint decision to go there? I do not know whether that would constitute no detriment. That would have to be agreed as part of the overall package.
My understanding of no detriment is that the overall position in the UK, including Scotland, will not change and that only the share between Westminster and Scotland will change. If that is taken in the context of the Barnett formula being retained for the vast bulk of spend, there is a difficulty. The Barnett formula does not work with the devolved powers that you are getting, let alone if they were further increased and Scotland was to get the corporation tax abilities that Northern Ireland has and other such powers. There seems to be pressure to move in that way. In those circumstances, the Barnett formula would become less and less relevant. The way in which it works would become obsolete, and it would not allow the accountability that you want to the people of Scotland for tax raising and the resulting spending from those taxes.
Sticking with Mr Bermingham’s submission, I note that there is a section called “The Practical Impact of Additional Fiscal Powers”. Paragraph 2.3 says:
“The majority of Scotland’s income will still be in the form of the block grant, with the accompanying restrictions and limitations on financial planning and management, such as no end of year flexibility and the lack of ability to decide reserves.”
Given the thrust of your argument, should the restrictions be loosened? If so, how?
The restrictions should be loosened. Scotland has the ability to build up a fund—essentially, it is what the Treasury calls a rainy day fund—but it is quite small. Years ago, the public sector was always accused of rushing to spend its budget in March, and a lack of end-year flexibility encourages that regime. If Scotland had more power over reserves, with the ability to designate reserves and things like that, it would be able to encourage longer-term planning.
The thrust of my comment is that, if Scotland is looking to develop a fiscal framework, it should be aimed at medium to longer-term planning. A good fiscal lever would be the ability to designate and hold your own reserves so that you could plan better for longer-term projects.
What scale of reserves are you talking about?
That is the question. We can look at local government as an example. It has a fair amount of freedom to hold reserves. Essentially, there is a balance between making a profit from council tax and holding an appropriate level of reserves. In a sense, the elected representatives in the local authority are accountable for that. That prudential-type measure, which is included in a prudential-type regime or framework, involves deciding what the right level of reserves is and what would be too much. I do not have a particular view on that. There is not much evidence at a central Government level to say what the right level is. I am based in Belfast, and at the local government level there, the edict is 5 per cent of the rates base. However, there is not a lot of empirical evidence to say that that is the appropriate level—it is just a guideline.
Does Ms Barbour or Mr Stevens have any comments to make?
No, thanks.
No.
11:00
I switch back to the ICAS submission, which has an interesting section on preventative spend—I am sure that colleagues will want to drill down into its detail. It says:
“preventative spending, like capital spending, is about investing in the future. We believe this provides clear justification for the extension of the Scottish Government’s revenue borrowing powers to fund preventative spend initiatives within prescribed limits.”
What would those limits be?
I am not going to commit myself to any limits. My colleague who wrote that section was not able to attend today. I can certainly get you more written evidence on that, or somebody could attend on a different date, if you like. I am a tax person; preventative spend is not my cup of tea.
I am sorry about that. I have a wee follow-up question on that. Perhaps you can give me some information in response; if not, perhaps we could get some follow-up information. Your submission says:
“The debt financing of charities by government is a potential model for providing working capital to charities for preventative spend initiatives.”
Can you tell us a wee bit more about that?
I will put that in the follow-up information.
That is fine.
Mr Stevens, your submission talks about the no-detriment principle encompassing tax competition. As a taxman, so to speak, do you believe that tax competition is a good or a bad thing? Where does it fit in to the proposals?
I do not have a view that it is good or bad. I strongly believe that the world is full of tax competition between countries. If you have spent your life advising companies that trade across borders, you have seen that tax competition is rife everywhere you look. It has been for a lot of years and it is growing all the time.
On the subject that we are discussing today, it is completely inevitable that there will be a growing element of tax competition north and south of the border, even if that is not the driver in any way for any changes to tax law. As soon as bits of tax law move out of synch with each other, it will create tax competition. A few years ago, when the question of a Scottish rate of income tax started to arise, the first thing that became clear was that it would create another bit of tax competition between countries.
Ms Barbour, you are a tax expert. How do you feel about tax competition?
I feel exactly the same as Patrick Stevens. I do not have views about whether it is good or bad; it just is. As long as we have different rates and regimes, there will always be distinctions. Part of what people find attractive about a tax regime is that it is lower, simpler or whatever. However, it is quite difficult, because we have to pick out tax competition from tax incentives and tax avoidance. If there is competition, one element will be avoiding the higher rate, will it not? That is another side to it. Those elements are all tied up together, and I am not sure how they can be separated from the no-detriment issue.
Mr Bermingham, do you have a view?
Only in the context of the no-detriment issue. That kind of tax competition will be very difficult to assess. The Treasury might believe that there is some kind of detriment or adverse impact for it, but that will be almost impossible to prove. It might be possible to take a view regarding tax on companies—corporation tax—but it will be almost impossible to measure for individuals. I think that the no-detriment issue is going to be the difficulty with tax competition.
We have heard a number of examples, such as air passenger duty. Are you going to use that in a tax competitive fashion to make Scottish airports more attractive, or will the north of England do that? Is that tax competition or is it a matter of no detriment that needs to be evened out?
Thank you for that.
I open up the session to colleagues round the table. The first questions will be from Mark McDonald.
We saw an example of tax competition very early on with devolution of land and buildings transaction tax. The Chancellor of the Exchequer, in his autumn statement, announced radical changes to stamp duty, which led to a change being made to the rates of LBTT that were applied by the Scottish Government. A precursor to that was the process here in Scotland—the Scottish Government consulted in advance on the rates that it was going to apply. The way that the UK budget system works is that the chancellor can just stand up and say, “I’m doing this,” and it happens at midnight. As tax devolution becomes more prominent and more prevalent, will we need to look at how the systems work to ensure that such processes do not lead to the rug being pulled out from under the feet of a devolved Administration in that same way?
Was that question for me?
It was for whoever wants to take it first.
I will start.
That is not particularly what I had in mind as tax competition. That example was, in a way—please excuse this—political competition between two chancellors. What I think of as tax competition is where the aim is to attract individuals, businesses and activities to one location or another. That is just a piece of conversation—
I presume that the principle still applies, however. The Scottish Government consults on rates, whereas at UK level there is the rabbit-out-of-the-hat approach.
I am sorry—I was going to answer your question. I think that it is going to be an interesting journey for each set of lawmakers, particularly if they get into the mindset that there is going to be tax competition, but I do not think that many people are thinking of the situation in that way at the moment. At first it will be more of an unintended consequence of people following policies that they think are right for their part of the UK. I am simply saying that it will turn into tax competition, because that is life.
I will be quite surprised if the UK method of announcing such changes alters hugely, mainly because it is not Scotland that the UK Government would normally have in mind—Scotland is not who UK ministers will be thinking they want to compete with in attracting, for example, businesses or high-income people into the rest of the UK. The fact that in Scotland things are worked out through a more consultative process will not, I think, impact greatly on them.
I will make no comment on the possibility that a Tory Government might not think of Scotland.
Please—I do not wish to go there.
I am merely being flippant. Ms Barbour, do you want to comment?
We will not go there.
I agree that the SDLT move was quite political, and I think that it probably had a different impact because LBTT was coming in as a new tax here; it was a first settling down. However, I do not think that the Scottish taxes can become radically differently from UK taxes. I am sure that there will always be a tendency to pull together, as there is with landfill tax just now.
In terms of process, tax practitioners frequently ask for further consultation at UK level. We have never had it on rates and bands and things like that, and I do not think that we will ever get it. It is a political lever and not something that we would actively lobby for or comment on. We did not comment on LBTT rates, for instance, because rates are not set by tax practitioners. We work with the legislation as it comes to us; we will come to you if we think that the legislation is difficult to work with.
We like the consultative processes here in Scotland; they have been helpful. It will be an interesting task—and one that sits with the Scottish Parliament—to determine whether the processes around your rates need to differ. I have heard plenty of commentary about the fact that you do not have an annual finance bill. Whether we want one is a moot point, but maybe the processes will need to change as part of the overall fiscal framework.
I agree with colleagues; I do not see the process changing, per se. In Northern Ireland, the main driver for corporation tax legislation was competition with the Republic of Ireland’s rate in order to attract foreign direct investment. That was the key driver. Using that analogy in respect of Scotland’s nearest neighbours, you might see some marginal adjustments to tax and slight differences appearing, but I do not see a hugely significant difference coming in the near future. I think that the process here and the process there will evolve, as my colleagues have said.
A concern was expressed during examination of the powers that are proposed by Smith and the draft clauses about the narrow control that the Scottish Government would have over tax rates. It has been suggested that were a future Scottish Government to decide to reintroduce a 50p top rate, individuals could, because the Scottish Government would not have control over all areas of taxation, apportion their income to bonuses or dividends that could not be touched by the Scottish Government under that regime. Is that a concern for you in terms of how effective future tax powers could be in Scotland?
That comes back to the point about tax competition. If there are significant differences, people will use what is available in tax competition and tax planning to do their best not to pay too much, however that is described. That is just the nature of people. Tax measures such as individual savings accounts can also be used in people’s favour.
The suggestion was that a loophole could be created in Scotland that did not exist in other parts of the UK, and that people could use it to lower their tax liability.
I am not sure that I would call it a loophole. It is what it is. There is the potential for directors of family-owned companies to make a perfectly legitimate business decision as to whether they pay themselves a salary and/or dividends and what proportion they use. It will be the case that dividends fall to UK tax and that salaries will fall to the Scottish rate.
I agree with Charlotte Barbour. I follow her point by stating the obvious and saying that it will make a difference only if the two sets of rates start to differ by a lot. If they do, a number of questions will arise, and I accept that the example that was given could raise some interesting questions. I agree with a lot of what has been said.
Does Mr Bermingham wish to comment?
I have nothing further to add to what my two colleagues have said.
The ICAS submission has an interesting page about VAT, which we have already discussed quite a bit today. You have pointed out the production-based and consumption-based options, and it appears that we are moving towards a consumption-based option, as we heard from the previous panel. I was interested in your comment about no detriment and VAT, because I had not thought of the two as being connected. I had thought of no detriment in other contexts, but you are saying that any decision on the calculation of the VAT assignment also needs to be informed by how the no-detriment principle will be applied. Could you open that up a bit so that I can understand its significance?
11:15
It depends on how you calculate VAT, and then it will wash through. The calculations will go into the Barnett adjustment. Flowing through from that, you then need to unpick what relates to the economy and what relates to Scottish measures in order to continue to balance the Barnett formula adjustments in the future.
I will have a think about that. Thank you.
I take the point that somebody else wrote the part of the submission about preventative spend. You will write to us on that. The main question in my mind in relation to the suggestion is the problem that we have grappled with in other contexts: how do we define preventative spending? I found it to be an interesting suggestion, and it is not one that I oppose. However, I imagine that that problem would be one of the objections from those who did not support the suggestion, so it would be interesting to get your thoughts on that.
I will get something back to the committee.
At the end of your submission, you seem to be questioning the independence of the Scottish Fiscal Commission. You say:
“its independence is not necessarily obvious if it is a part of the Scottish Administration, and when one of its proposed functions is to prepare such reports as Scottish Ministers may require.”
Do you think that its status ought to be changed?
There are perhaps two points there. First, I completely accept that the draft legislation is structured in a way that means that the Scottish Fiscal Commission would be independent of the Scottish Government, but it is quite a technical independence. If the commission is sitting in the middle of the Scottish Government, I am not sure that the wider public would see the distinction. The commission needs to be seen to be independent, if independence is what you want.
Thank you.
One of our other obsessions is the block grant adjustment, so I will ask the other two panel members about that. The Chartered Institute of Taxation—I have got it right this time—makes interesting comments about that in sections 5 and 6 of its submission. Section 5.3 mentions
“indexed deduction and levels deduction.”
I presume that we will get indexed deduction. Can you explain what you mean by “levels deduction”?
I am going to duck that question. We talked to various colleagues and people in linked organisations, and we have tried to point towards some helpful things to consider. However, I am a tax person and that question is deep in the field of economics. I will happily write a note on the subject, but I will duck the question at the moment.
Is that also the case in respect of the comment in section 6.3 of your submission? We accept that we will get indexed deduction, but you mention that one concern is that
“the tax/spend areas in Scotland and the rest of the UK might diverge for reasons unrelated to Scotland’s policies.”
Do you want to pass on that, as well?
I will pass, but let me come at you for a couple of sentences about something that we touched on earlier, if I may. The different elements that go into the change in tax take will be enormously diverse. Behavioural change will be one of them, to go back to the point about dividends that we were discussing earlier. There will be a whole load of behavioural change if rates become very different from each other. I am simply suggesting that exactly how we get to the block grant adjustments will need to be the subject of lots more negotiation now.
Thank you—that is helpful.
Mr Bermingham wants to come in on that.
Yes—it is on the point about the Barnett formula. Essentially, the Barnett formula is the mechanism for adjusting what are called consequentials. Where Westminster spend rises or falls, it is about how comparable that is—whether it is a devolved function or whatever—and then whatever percentage is devolved is multiplied by a population-based percentage to get a Barnett consequential.
Back in the 1970s, there was a fixed budget: since then the Barnett formula has been applied, with consequential adjustments. We seem to be moving to a situation in which there will be an additional measure, with whatever adjustment comes with the devolved powers that the Parliament will get, in terms of tax forgone or spending being transferred, and the associated savings. The Treasury will list how that adjustment will work in its statement of funding policy at budget time and autumn statement time. I assume that there will be a bit of an adjustment to the section that covers Scotland’s consequentials. The base figure will need to be taken and indexed according to whatever the index is; that will be the amount that Scotland will get in consequentials.
What shrouds Barnett in mystery is that it is done behind closed doors in the Treasury, so there is little input into what is applied in terms of devolution. The classic example is the Treasury saying that the Olympics benefited the whole UK, so there was no consequential adjustment for all the increased spending on the games, even though Scotland may well have spent money locally—perhaps on supporting teams that were based in Scotland, or something like that. That would have come out of your pocket rather than there being any Westminster funding for it. That is an example of how the system can work incorrectly, or not in Scotland’s favour.
There is a role for the Joint Exchequer Committee or perhaps the Scottish Fiscal Commission to be involved with the Treasury so that you can agree in advance of the funding statement being produced exactly what the indexation measures are and what the impact will be on the Scottish budget.
I take your point on that, but you seemed to imply earlier that the Barnett formula would wither away. I did not quite follow your logic on that. There are grey areas—there always have been, and that is not going to change—but surely there will still be notional consequentials, and then, in principle, whatever we are supposed to raise in tax will be deducted from that. I do not see how the Barnett formula will wither away until the time—if one arises—when we raise all the money that we spend.
I said “wither away”, but I suppose the institute’s position is that the Barnett formula is no longer fit for purpose, particularly as we move further down the devolution route, and so it should be withdrawn and replaced by something that is more needs based. It is probably wrong to say that it would wither away because there would obviously be something in its place. We are suggesting that that something should be more needs based, rather than being based on a consequential adjustment to something that was set up in the 1970s.
So the Barnett formula need not wither away, as a matter of fact. It could survive, but your view is that it should not.
Yes.
My first question follows up on the comments on VAT assignment in the submission from ICAS, which talks about assigning VAT by place of production or place of consumption. That choice has been put to the committee before. Do we have a simple choice between those two options? Is a hybrid option possible or is it simply the case that we have to decide on production or consumption?
A hybrid sounds painful. I think that both choices have their difficulties. VAT is what we call an indirect tax, as it is not levied directly on the economy, and it has endless stages in it. If it is levied on a consumption basis, it will not necessarily be levied on things that are made here.
The question is really difficult. I see the sense in having a hybrid because it might cut out the worst of both options, but it might add the worst of both. I do not find VAT assignment to be an easy thing to marry up to the Scottish economy.
Do others have views? Is it simply a choice between the two options or is there some other system that we have not thought about yet?
If there is another system, I have not thought of it, either. On the assumption—it is a big assumption—that we are all going to remain part of the EU, it seems to me that VAT is moving towards being consumption based, and is mooring there. That was the theory that I had always assumed we would get to.
You made an interesting comment earlier, Mr Bermingham. The block grant adjustment is not an easy subject, as we have seen over the past couple of years and the past couple of months in particular. You suggested that it would become even trickier.
So far, we have considered block grant adjustment only for DEL, not for AME. You touched on how you might have to think about it for AME. Will you expand on that a little? We have spoken a lot about a block grant adjustment for DEL, but we have never really discussed how it would be done for AME. Are there any other factors that we should be thinking about that you did not mention earlier?
The first point is that AME is not driven by the block grant process and the Barnett formula, which is purely DEL. I suppose that there could be a position where welfare spend is devolved and either accelerates or slows at a different rate from that of the UK. Thinking about the no-detriment principle and how things would be adjusted if the overall position was going to be financially the same, if your welfare spend was increasing at a significantly different rate from Westminster’s rate, you would need indexation, or whatever process was in place, in order to adjust AME spend to reflect that.
In my view, that is a more difficult measure than simply describing things on a tax basis—saying what amount of income tax the UK has forgone and saying that you will index it by the general uplift in the tax base in the UK, or something like that. Because the AME spend can be disproportionate, that makes the adjustment more difficult, and it is not possible to apply a flat index. That is why I suggested that you might need to consider what the drivers of welfare spend are, so as to create an index that fits the bill.
I have a question for everyone, but I will start with Mr Bermingham as he talked about this earlier. In your written submissions, you all touched on borrowing powers. Let us put preventative spend to one side for now—I accept that we might get some more details on that.
On borrowing for other purposes, or specifically for capital, do you have views on what the regime ought to be? I think that you spoke about rainy-day money under the existing arrangements. Obviously, you think that it should be more than that, but by how much? Should there be limits at all, or should there be an unlimited opportunity? In effect, the markets regulate that in some way. What sort of borrowing regime ought we to have in place?
I will come to the point about limits in a minute.
The Smith commission alluded to the need to have some kind of prudential borrowing measures to support your borrowing powers. The Treasury has put ceilings on your borrowing levels of £2.2 billion and so on. The matter needs to be viewed in the context of your framework supporting a market view. If you ever came to issue bonds or anything like that on the market, you would need some kind of credible fiscal framework and regime to support that. That would support your case, impact on the interest rates that you get and so on.
The prudential regime suggests that you should be agreeing on what an affordable borrowing limit is. In that sense, there is no dictated limit but you, as the elected representatives, would agree what an affordable borrowing limit is. Affordability is viewed in the context of the impact of the borrowing on the local tax base that you are raising. For example, would it be acceptable if 25 per cent of the local tax revenues that you were raising was being spent on financing borrowing?
You, as elected representatives, would have to come up with the affordable level to be put in place and agree and perhaps legislate for that affordable limit through the annual budget process. That would allow the market and others to see that that is the kind of debt to GDP level, or whatever the Scottish Government is aiming for, that is budgeted and put in place. That sort of regime fits in with having a fiscal framework. That is the appropriate way to go.
11:30
Do other panellists have views on that?
The only additional point that I would make is that, on the assumption that the Scottish Government borrowing was being underwritten by the whole of the UK, I suggest that there is not quite the same link between levels of borrowing and what the market thinks of it, because logically the market would look on it as being part of all UK debt. If that is going to be the case, it would therefore seem that the amounts of borrowing need to be agreed primarily between the two Governments. Based on many principles that we were talking about, I was saying that the market would not have the same effect in that case.
Yes. However, if you are still, shall we call it fiscally federal, under Westminster, obviously Westminster could have a backstop of imposing a limit on Scotland, should you really go to town in a way that it fundamentally disagrees with. I am talking about a fiscal framework that gets you towards a position that fits in with that fiscally federal approach, but would also work if you ever went to full fiscal autonomy—the regime would already be in place.
On the question of analysis and forecasting, ICAS says:
“the Scottish Government are developed to deliver the financial analysis necessary to support policy decisions. The Government needs to work out how much tax it will really raise”.
One of the four critical questions raised in the CIPFA submission is:
“What is the fiscal impact if further devolution takes place?”
If we are subject to making a decision and, as has been pointed out, Westminster makes a decision that overrides that—and given the difficulties with the block grant adjustment, in that we have already taken on landfill tax and 11 months later we do not know what the adjustment will be—is this all just wishful thinking that we cannot possibly implement?
No, I do not believe so. The thrust of what we are trying to get to is a fiscal framework that will fully support further devolution, even towards full fiscal autonomy. That is the type of fiscal framework that we envisage should be put in place.
In terms of measuring the impact of the things that Mr Mason mentions, it is important, as we suggest in our submission, that Scotland reviews its financial management in the sense of having a whole-of-Scotland position—a Scottish balance sheet and things like that—regularly available for scrutiny, in order to be able to see what the impact would be on tax raising or on the balance sheet if we took on extra borrowing or did this, that and the other. You need to bring in a financial management regime that suits that.
Looking forward is obviously about prediction; it is a guideline and is not an exact science. So, yes, things will impact on Scotland, but that is why we have alluded to trying to make an assessment of the risks. We are talking about a medium to longer-term financial plan, which will also feed into any discussions that you have with the Treasury, so that there is an opportunity to pre-empt a situation in which the Treasury might want to do something that would have a fundamental adverse effect on your plan. If you knew that up front, you would have the ability to say, “Look, this is the way our plan works, it’s based on these assumptions, and if you did that, it would have an impact that would be detrimental.”
It is not an exact science, but I cannot see how the Scottish Government can properly manage some of those powers without having that financial management framework in place.
So we should do all that and then, in small letters at the bottom of the page, it would say, “This is all subject to nothing nasty happening to us.”
The same principle applies in Westminster. I am talking about the impact of the Treasury doing something that you had not planned for, or for which you had not worked out the risk in advance.
However, you also say that
“The Fiscal Framework should have a legislative basis”.
Will you expand on that?
It is the same as the Treasury statement of funding policy. It is driven by Treasury’s legislative basis, which sets out that the Government must produce a budget and a charter for funding statement at autumn statement time and at budget time. That must set out certain rules, such as what its fiscal mandate is. If you put that in legislation, that makes it a legal requirement for you to update that fiscal framework regularly.
So it is just the structure that is in the legislation.
Yes. It is the structure that is in the legislation.
It does not say, “You’ve got to balance the budget over six years,” or something along those lines.
No. That would be included in whatever your funding policy statement is in your fiscal mandate. The legislation is really saying, “This is the structure and this is what you need to publish and produce.”
Right. It was not clear to me whether you thought that the Fiscal Commission should make forecasts or just review them. In paragraph 4.7, you say that
“the estimates can be checked, independently reviewed by the Scottish Fiscal Commission”
and, in paragraph 5.3, you say that
“the Commission should provide independent economic and fiscal forecasts”.
The Office for Budget Responsibility produces economic and fiscal forecasts, which the Government uses, but we are saying that the Government does not have to use them. You could go and do something else. You are taking that as advice—somebody is independently feeding those figures to you. Again, whatever your output is and whatever your decisions are, there is a role for the Fiscal Commission to review that and say, “Yes, that makes sense”, or “No, we think that those projections are wrong”, or whatever. You do not have to take the Fiscal Commission’s forecast as read. You might well interpret things differently and come to a different budget decision. However, you certainly have some input to inform that decision making in the first place.
But the Fiscal Commission is not making forecasts at the moment, is it?
No, not at the moment, but I would suggest that you need that kind of input. To me, that would be a role for the Fiscal Commission.
Do the other two witnesses have any views on that?
We thought that there was a lack of clarity about whether the Fiscal Commission would make forecasts or check them; maybe it will do something closer to auditing. It was not crystal clear from the draft legislation exactly what the Fiscal Commission’s functions would be.
We were also uncertain about how much the Fiscal Commission would be working with the Scottish rate of income tax. We wondered whether the statistics would come from HMRC and, if they came from HMRC, whether there would be duplication of work. If there was not duplication, we wondered whether there was scope to be going in contradictory directions. We thought that a bit of fine-tuning might be needed on areas in which they would be working together.
It seems to me that the OBR does a reasonably good job at the moment. I am not saying whether its forecasts are good, but they are respected and accepted as being about the best estimate available and largely independent of Government. That is a reasonable place to be and I had assumed that Scotland would like some sort of equivalent.
I wonder whether that relates to what was said earlier about the OBR having the appearance of being independent. In fact, HMRC told us that most of the OBR’s information comes out of HMRC. The OBR is not really that independent, but it appears to be. Is that the important thing?
The OBR does economic forecasts with the information that it gets from wherever, including HMRC. I am not sure—
I think that HMRC would argue that it basically does the work and the OBR just puts the icing on it, but perhaps you would not agree.
I would not accept that. The icing is sufficient for me to be impressed by its objectivity.
In the draft legislation for the Scottish Fiscal Commission, there is quite a bit that gives Scottish ministers the ability to exercise influence over appointments. They can put them forward and the reports need to go to Scottish ministers before they come to the Parliament. Some of that might not necessarily lead to an independent Fiscal Commission, although you might not want it to be independent; you might want it to be impartial. Scotland is quite small.
The OBR takes a wider view than just the Treasury figures. For example, the Institute for Fiscal Studies and similar organisations feed into the OBR. I accept what John Mason says, and the Treasury might take the same view, but the OBR is slightly more independent than you give it credit for.
Fair enough. I will stick with you, Mr Bermingham. Earlier, we talked about Scottish Government borrowing and UK Government borrowing but where does local government fit in? I know that that is a CIPFA area. Local government has been left out of the equation. If we look at Scottish borrowing, we cannot totally ignore the fact that local government borrowing is included and it looks as though local government will have more freedom than the Scottish Parliament will have. Does all that tie together?
In a sense, it does. As you get new powers and increased freedom of financial responsibility, it is important that you take a view on the whole-of-Scotland position, including that of local government. If you move towards fiscal autonomy, that is the sort of information that you will need to know. You cannot leave out local government and just let it run on and borrow whatever it can, because that would mean that local taxation would be completely unmanageable.
I am not saying that a regime needs to be put in place over local government; it is already there and it works reasonably well. However, local government needs to be taken account of, particularly if you were, for example, going to a market to issue bonds or something like that. It would be important for you to have an idea of the whole of Scotland’s finances.
Are you saying that the Parliament should get more involved in what local government borrows?
I am not saying that it should get involved, but it should have an overarching view of Scotland’s total debt.
Mr Stevens, paragraph 3.2 of your submission talks about the Barnett formula and uses the word “mechanical”, which the committee has considered previously. I quite like that word, because it suggests that the system that has been set in place just works. You also say that there needs to be a regular, perhaps annual, review of the formula. How does that tie in? I have a picture, on the one hand, of something that churns along for a few years and, on the other hand, of something that is reviewed every year.
For the avoidance of doubt, we also like the term “mechanical” for a series of reasons. We talked about this earlier today. I believe that, each year, Government of whatever variety ought to account to the people for whatever the bottom line is of the block grant. That takes account of the Barnett formula and the various adjustments that we have been talking about.
We spoke about how we need to simplify and how we cannot go to every theoretical place that we would like. That should apply not just to the adjustments, but to the formula itself. Only when that is all pulled together can you present to the outside the reason why the net amount of money that is received is the right amount. You will therefore inevitably have to re-examine what the Barnett formula says as a starting point. You cannot just have a black hole at the top and then explain in detail the adjustments that are being made, because there would still be a fairly black hole at the bottom in that case. The whole calculation needs to be looked at.
There are loads of mechanics to help us towards that, but if you just assume that a big chunk of it will be all right every year, that will not get you to the right answer.
Is it possible to look at things too often, like digging up a plant to see whether it is growing?
11:45
It is possible. I am trying to say, most of all, that I believe that each year the public need to be satisfied that a bottom-line correct amount of money has been received. How you become able to do that is probably a matter that the Government can best judge, because it will need to present that information to its population.
In paragraph 13.1 of your submission, you talk about Scotland having control of
“60% of its spending and 40% of its tax revenues”.
Is VAT included under what is controlled? With VAT, we would have the money but we would not have any control, in a sense.
I believe that that is so, and I think that the implication of what you are saying is fair.
I have a question about the no-detriment clause. It has concentrated the minds of many people who have given evidence to us. Charlotte Barbour and others have made reference to the air passenger duty and how that might work. Is it even possible to adhere to such a clause? Do you think that it is there just to temper what people might think to do, or is it simply for political reasons—to reassure everybody that, whatever the outcome, we are all going to be the same and live happily ever after?
Let me start; I am sure that my colleagues will have better answers than I.
On day 1—or, more to the point, in year 1—I think that the concept of no detriment is far easier to get one’s mind around. Add in the concept of tax competition that we talked about earlier—inevitably, that will arise—and it is a term that is meant to be a little bit provocative. Tax competition and no detriment, you could say, are mutually exclusive. Therefore, I think that you may have a point.
I would agree with that in its broadest terms. Clearly, if you are part of a wider picture, you do not want imbalances between the two parts. However, if we take it down to much finer detail and a much finer analysis—as in the earlier example of the personal allowance being significantly increased—is that no detriment, is it competition or is it political? I would expect something like that to wash through into no detriment, and therefore—we are all the same.
Also, if the UK Government deliberately had a lower top rate of tax, which persuaded a largish number of extremely high-income people to move to its jurisdiction, would that be no detriment? I am just trying to give examples.
There may be an element in the no-detriment rule by Treasury that is related to EU rules about not creating, within the UK as a whole, particular areas in which there is unfair competition. That may be part of the argument for a no-detriment rule, so I would not agree that it is there just for political purposes. I think that there is a rationale for it.
Coming up with a simple way of adjustment is the way forward. I think that there will be certain areas in which it will be almost impossible to measure and judge, because we assume, as we have said, that the adjustments and measures are put in place to achieve something such as stabilising the economy or economic growth rather than to create tax competition or something like that. If we stick with the original driver, there will be some areas that are very hard to measure and which perhaps should be left out of areas of indexation or anything like that.
On that basis—if you see that as basically a good thing—are there already examples of areas in which we suffer in Scotland and which could be corrected? I am thinking of the area of renewable energy, for example. Do you think that the no-detriment provision could correct a number of such issues?
Potentially, yes. It will not always be a negative thing. As I said when I used the example of welfare, disability allowance and so on, there will be examples of the arrangement impacting Scotland favourably.
That appears to have concluded questions from the committee, but I seek clarification on one issue.
Mr Bermingham, you said that Barnett should be replaced by a needs-based formula. Is that the collective view of CIPFA?
It is indeed. We have issued a paper on that.
What would the impact be on Scotland of that?
It is an interesting question. If I knew what the needs-based measures would be, I could tell you exactly, but as I do not, I cannot. However, I can allude to some historical evidence. As you are probably aware if you have considered this issue in the past, there is a suggestion that Wales would argue that it is underfunded, that Northern Ireland, which does not have an exact view on the issue, is overfunded and that Scotland is perhaps overfunded as well.
The only evidence that I have comes from two old studies from the Treasury. They are out of date—one goes back to the 1970s and one goes back to the 1990s—so I would issue that caveat. However, there is also a piece of work that I did for Northern Ireland’s Committee for Finance and Personnel in relation to its review of the Barnett formula, and there is another piece of work that was done by Professor Iain McLean that looked at some fairly crude measures of need. Those three studies roughly correlate with each other and suggest that Northern Ireland is not too far away from its needs level, Scotland is probably overfunded and Wales is definitely underfunded.
I cannot tell you what will happen if there is a move to a needs-based formula, as that is something that would need to be agreed on, but there would obviously need to be a huge transition to that. The only conclusion that I can draw is that there would be more of an adjustment in Scotland than in the other two areas, in terms of an adverse impact.
I would have thought that the biggest adverse impact of a needs-based system would be on London, given that it sucks so much money out of the rest of the UK and has so much spend in terms of the civil service, infrastructure and so on, but we will leave it at that.
Does anyone want to make any further points?
No.
No.
No.
I thank you for your evidence, which was helpful. I suspend the meeting until noon, to allow for a changeover of witnesses and a natural break for members.
11:53 Meeting suspended.Previous
Attendance