Fiscal framework exposes the Scottish Government to some risks which it can’t influence

Graeme Roy, Director, Fraser of Allander Institute

Professor Graeme Roy spoke about the biggest reform to the financial powers of the Scottish Parliament since 1999 including changes that will not only be fundamental for how public services are funded in Scotland but also for our future economic prospects.

Presentation to DHI seminar: May 11 2017

Much of recent debate has understandably concentrated on big political questions such as Brexit, IndyRef2 and the UK General Election, so it’s easy to forget that we’re going through the biggest reform to the financial powers of the Scottish Parliament since 1999. Changes that will not only be fundamental for how public services are funded in Scotland but also for our future economic prospects.
So tonight is a great opportunity to give a quick recap of the new fiscal framework for Scotland and to discuss some of the features that make it such a unique framework, not only in the UK, but also internationally.
I’d also like to set out what we at the Fraser of Allander see as some of the key issues that we think will be important to watch over the coming years. And then I’ll hand over to Caroline who will give us all the solutions!
So just to recap, and as you know, the Scottish Parliament is currently going through a process of substantial financial devolution.

Powers that will give greater opportunities to do things differently in Scotland. But powers that will also substantially increase the risks and uncertainties underpinning future Scottish Budgets.
Of course, there is a debate about whether or not these powers are enough or if we could further, but you’ll forgive me from staying clear of that debate this evening!
Soon the Scottish Government and Local Government will oversee a budget of over £40 billion.
In other words, over 60% of public spending for the benefit of the people of Scotland will be controlled directly – or indirectly – by Holyrood.
The big reform has been to expand the proportion of the Scottish Budget from ‘own-source’ revenues.
The fact that the Scottish Budget was determined almost entirely by a block grant from Westminster that had very little to do with Scottish economic performance or provided few opportunities for Scottish administrations to alter the overall spending envelope that they faced, was seen as a key weakness of the original devolution settlement.
A ‘house-keeping’ budget was how Jim Mather the former Enterprise Minister described it.
Now around 40% of ‘devolved expenditures’ will be funded by revenues raised in Scotland – a figure that will rise to 50% once approximately half of VAT revenues are assigned to the Scottish budget.
As a result, Scotland’s budget will now depend crucially on three key elements –
What remains of the Westminster block grant – as determined by the Barnett Formula;
Future tax policy choices of the Scottish Government; and,
The relative performance of Scottish devolved tax revenues.
If I can park in your mind this final point – the ‘relative’ performance of Scottish devolved tax revenues – as this is absolutely crucial for how the framework operates in practice.

So here are the revenues being devolved, starting with Land and Buildings Transaction Tax a couple of years ago and moving through to VAT in 2019/20.
I should say that to provide some consistency across the numbers these are the GERS estimates of the revenues for 2015/16.
If you’ve been following the GERS debate on social media in recent weeks then I apologise if the use of some numbers based on estimation here and throughout my talk upsets you!!
By far the two biggest transfers in terms of revenues are income tax, which came into effect this year and VAT.
It is important to also note what is not being devolved. For example–
On income tax, the personal allowance and ‘definitions’ of income are reserved along with dividend and saving income tax receipts.
On VAT, there is no authority to vary rates or exemptions. It is essentially a transfer of revenues rather than any real policy power.
Other taxes with close links to income tax – e.g. national insurance, capital gains and corporation tax are also fully reserved. This is important as it opens up the prospect for devolved tax policy decisions to rub against UK-wide reserved taxes. An early example of this has been the interaction between the new lower threshold in Scotland for higher rate income tax and the (now higher) threshold for UK-wide National Insurance contributions which means that a small number of taxpayers in Scotland face a 52% marginal tax rate on earned income.
Finally, the overall fiscal stance is also reserved. If the Scottish Government wants to spend more money on public services than currently funded by Barnett on a long-term basis, it will have to raise the revenues to pay for them.

Alongside these new tax powers, the Scottish Government will be taking responsibility for substantial welfare powers – worth around £3 billion by the time they come fully on stream.
These powers can essentially be grouped into 3 areas.
Firstly, flexibility in how certain elements of Universal Credit are paid.
Secondly, the full devolution of a number of benefits themselves including:
Benefits tied to ill-health and disability – e.g. DLA and PIP; and,
Elements of the Regulated Social Fund – e.g. Sure Start, Winter Fuel Payments etc.
Thirdly, the ability to ‘top-up’ existing UK benefits and/or introduce new benefits, subject to meeting the costs of doing so – including any additional administrative costs.
Alongside these welfare powers come important responsibilities around the delivery of employability programmes.
These new powers represent a major opportunity for the Scottish Government and the Scottish Parliament, both to re-design elements of the benefits system to better reflect their vision for social justice in Scotland, and to take advantage of synergies with existing devolved responsibilities such as in health, education and skills.
They also pose a challenge. Experience has shown us – both in the UK and internationally – that delivering new welfare programmes isn’t easy.
Given that many of these benefits – particularly the Disability Living Allowance and so forth – are vital for many of the most vulnerable people and families in our society, it is absolutely crucial that we ensure delivery and continuity of payments from day 1.
At the same time, it’s important to acknowledge that many of these benefits are expensive. Entirely legitimate and well-intentioned ambitions to increase the value of payments and/or to broaden eligibility may carry expensive price tags which will need to be paid for.
These are the new powers.

So how will the actual Scottish Budget be determined from now on?
Recall that in the past, the Scottish Budget was largely determined by a block grant. This grant was, in turn, calculated by the Barnett formula – the mechanism which allocates Scotland a population share of comparable spending in England.
Rather than design a new system for the block grant, the Smith Commission decided instead that the existing block grant should be adjusted to account for the new devolved and assigned revenues.
In year 1, that’s pretty straightforward. You take away how much was raised in that year in devolved revenues and then add back in these same revenues raised from the devolved taxes in Scotland.
There is zero impact on the Budget simply from the transfer of power, and the principle of ‘no-detriment’ is met.
But what happens after year 1?
If you keep on taking away from the block grant how much tax revenue is raised each year, then there’s effectively no change to the Budget process – the block grant is cut by exactly the same amount as is then added back in!
So what was agreed was an indexation mechanism to grow this block grant adjustment – or BGA – year on year at a counterfactual rate.
The purpose of this indexation mechanism was to create a BGA that represents a counterfactual estimate of the tax revenues that the UK would be forgoing post-devolution.
You’ll recall from last year that there was difficulty in getting both governments to agree on the most appropriate indexation mechanism.
The mechanism that was finally agreed – at least up to 2021 – was for the BGA to increase in line with the growth in the equivalent tax revenues per head in the rest of the UK (rUK).
In the case of income tax, this means that if non-savings non-dividends income tax revenues per head grow by 2% in rUK, the BGA will grow by 2%.
And this gives the key feature which underpins this framework: that is, how the BGA interacts with devolved revenues.
If Scottish tax revenues per head can grow faster than the growth per head in the equivalent taxes in rUK, then the Scottish Budget is better off compared to what it would have been under the Barnett Formula.
This is because Scottish tax revenues will be larger than block grant adjustment.
But equally, if Scottish tax revenues per head grow more slowly than in rUK, then the Scottish Budget is worse off.
An equivalent mechanism works for welfare payments. So that’s the broad outline of the framework.
What I want to do now is to run through some key issues that will be crucial to how the budget process will work in practice over the next few years.
Firstly, it is clear that we are dealing with an exceptionally complex framework.
If you thought Barnett and the existing Scottish Budget process was complex, you ain’t seen nothing yet!
We now have a mix of block grant, devolved taxes, local taxes, shared taxes and assigned taxes. All supported by a complex framework for borrowing – for four different purposes no less, forecast error, cash flow, emergency economic shocks and capital investment – use of reserves and various mechanisms to reconcile forecasts with the revenues that end up actually being collected at the end of the day.
How scrutiny occurs in a landscape like that is clearly going to be a challenge. And the Budget Review Group are trying to find ways to deal with this new and complex world.
It’s also going to be an important test for intergovernmental relations between Whitehall and Holyrood. In the past, HM Treasury has always been the dominant player in fiscal relations with the Scottish Government, but the new system is designed to be a partnership of equals.
Ultimately, it’ll come down to how well the two governments work with one another……particularly during times of disagreement.
Alongside all of this, there will be important practical challenges to overcome both in terms of the costs and administration around starting-up and running these new powers not to mention keeping track of the complex arrangements for different types of capital, resource, DEL, and AME spending.
Throw in to that mix things like the non-domestic rates pool and a new Scotland Reserve to mop-up underspends, and you have an interesting cocktail!
And remember, the key element of this – the BGA – is temporary and scheduled to be reviewed in 2021.

The 2nd issue concerns the economic uncertainties, challenges and opportunities that underpin this new framework.
As I’ve mentioned, one of the key objectives of Smith was to bring greater risk and reward into the budget process.
By being responsible for more devolved and assigned taxes, how the economy performs will, from now on, have a much greater bearing on future Scottish budgets.
But as I mentioned, there’s a subtlety in that the way this framework has been designed. What is actually most important is not how the Scottish economy itself is performing per se, but how the Scottish economy is fairing relative to the rest of the UK.
Recall that key calculation – “the Scottish Budget will be better off under this new framework compared to Barnett if Scottish tax revenues per head grow more quickly than rUK tax revenues per head (which in turn determine the block grant adjustment).”
And here the current state of the Scottish economy presents a challenging backdrop.
The Scottish economy has been sluggish for the best part of two years now.
Output fell in the final quarter of 2016.
The Fiscal Framework prepared for ‘emergency’ situations with additional borrowing available under certain circumstances:
“A Scotland-specific economic shock is triggered when onshore Scottish GDP is below 1% in absolute terms on a rolling 4 quarter basis, and 1 percentage point below UK GDP growth over the same period.”
It’s interesting – and a reflection of the fragility of the economy in recent months – that with growth of just 0.4%, compared to 1.8% for the UK as a whole, Scotland’s current economic performance satisfies this criteria.
If this continues, the Scottish Government has the opportunity to borrow an additional £600m per annum.
Whether or not they chose to do that remains to be seen. What will be important is whether or not the Scottish Government believes that this year’s weak economic data will continue into 2017/18 and how much this divergence in overall economic performance will spill over into actual income tax revenues.
Setting aside any current challenges, what is interesting – and not that well appreciated – is just how significant the effects of different economic performance will be for public spending in Scotland under this new framework.
We’re now dealing with very large revenues, particularly around income tax and VAT.
A small variation in the growth rate of income tax revenues for example – given that they amount to over £11bn – could very easily work out at hundreds of millions of pounds over a very short period of time.

To provide some context, we can illustrate what happens if just one of the economic determinants of tax growth is changed.
Wages in Scotland and the UK tend to be correlated.
Over the last three to five years for example, median wages have grown by a difference of – on average – around 0.3% points per annum.
Let’s take that number and see how a variation of that sort could impact on future Scottish income tax revenues and therefore the Scottish Budget.
As the chart highlights, a variation of this magnitude – equivalent to Scottish median wages growing 2.5% vs. 2.2% in the UK or vice versa etc. – is worth around £50m in year 1.
If the gap was 1 percentage point the difference would be closer to £150m.
And of course, if this gap was to stay for longer than 1 year, the potential revenue shortfall or gain gets that much larger over time.
And this is just on one factor. If employment growth for example was faster or slower in Scotland than the rest of the UK, then the impact will be magnified.
So sustained economic performance is absolutely crucial.
With many factors out-with the Scottish Government’s control including the ongoing challenges in the oil and gas industry, an ageing population vis-à-vis the rest of the UK and uncertainties around Brexit etc. many believe that Scotland will do well to match UK economic performance over the next few years.

But of course, the Scottish Government now has a range of new powers over tax and spend to support public services and/or help to grow the economy.
I think it’s fair to say that we’ve yet to have a fully informed debate in Scotland about how best to use these powers, plus the choices and trade-offs involved.
What’s the optimal level of public spending we want to support and how do we raise enough revenues to pay for these ambitions?
Do we want people in Scotland to pay more tax than in the rest of the UK if it supports greater investment in public services and what might be the long-term economic consequences of doing so?
In thinking about the policy choices open to the Scottish Government, the unique nature of this framework throws up a number of important – but subtle – issues.
Firstly, when you change a policy it’s often designed to alter economic behaviours to obtain a desired outcome. For example, the government believe that by cutting air passenger duty – through the creation of a new air departure tax – this will stimulate the economy by boosting tourism and connectivity.
But outcomes are uncertain and on many occasions when deciding whether or not to ‘pull’ a policy lever the evidence in favour or against its use is often unclear.
There are also concerns that need to be borne in mind around ‘unintended consequences’ arising from changes to behaviour. The Scottish Government has for example, expressed an openness to consider – in principle – the introduction of a 50p tax rate on those earning over £150,000 in Scotland. However, they are concerned about the unintended consequences on the Scottish Budget if some of the relatively small number of taxpayers that this will impact upon choose to change their behaviour – for example, by moving to somewhere else in the UK.
Secondly, it’s not just economic behaviours that policymakers need to be mindful of. Changing one tax could lead to unintended consequences in terms of other taxes.
For example, an increase in income tax rates targeted at earned income, may encourage people to switch their income into dividends or to move from self-employment to incorporation all to gain a tax advantage. Taxpayers in Scotland would still be paying tax but they would be switching between tax revenues that flow to the Scottish Government (NS-ND income tax) and tax revenues that flow instead to the UK Government (e.g. dividend income, capital gains tax, corporation tax etc).
The net effect of such action would be a cut to the Scottish Budget but a boost to the UK Budget.
This isn’t to say that one policy is better than the other, but simply that policymakers need to be mindful of how their decisions will actually play out in practice. In the Scottish context, given that we don’t control all the tax levers they have to be even more careful.
Finally, and perhaps less well known, is that because of the different structures of income between Scotland and the rest of the UK, common UK-wide policies on income tax (remember this remains a shared rather than a fully devolved tax) can throw up some odd implications for the Scottish Budget.

This chart shows non-savings non-dividends income tax liabilities across the income distribution in Scotland and the rest of the UK.
Two things are worth noting –
• Firstly, Scotland collects a greater proportion of its income tax revenues from the lower end of the income distribution than rUK.
• Secondly, in the rest of the UK – in part driven by the London effect – very high earning individuals make up a much larger share of income tax liabilities than in Scotland.
What this means is that the same income tax policy in Scotland and the rUK could have different revenue impacts.
For example, consider the introduction of a 50p tax rate not just in Scotland but across the UK as a whole.
Let’s set aside behaviour impacts and focus simply on the mechanics of the framework.
A UK-wide 50p tax rate for Additional Rate taxpayers will increase the tax paid by the highest earners in Scotland. But because there are more additional rate taxpayers in England – and they are on average richer than in Scotland – rUK tax revenues will increase by more than in Scotland.
The end result is that the Block Grant Adjustment will be larger than the increase in Scottish revenues, leading to a reduction in the Scottish Budget. All this, despite Scotland increasing its tax rates!
Now should the UK Government choose to spend the revenues that it raises from a 50p tax rate on public spending that brings with it Barnett consequentials – e.g. in health or education – then Scotland’s Budget may still rise (albeit because of a transfer from rUK taxpayers to Scotland).
But even then because of this interaction between the Block Grant Adjustment and Scottish tax revenues, the increase will still be less than would have been the case prior to 2017. And of course, should the UK Government decide to spend these revenues on things such as reserved welfare powers, defence or reduce net borrowing, Scotland would not receive any consequentials.
A similar issue arises with the personal allowance. Because a higher proportion of Scottish income tax revenues come from people close to the personal allowance threshold, any increase in that threshold will have a disproportionate impact on Scottish revenues vis-à-vis the UK as a whole, leading to a relatively larger cut to the Scottish Budget than in England.

The final issue is around forecasting and the uncertainty that this involves.
The Scottish budget cycle will now be heavily dependent upon forecasts made by the Scottish Fiscal Commission and OBR – for taxes, welfare spending and all the individual block grant adjustments.
Forecasting isn’t easy – even in normal times!
With Brexit and many other uncertainties, robust forecasts are even more challenging.
In a Scottish context, they are some further challenges over and above all of this.
Firstly, the depth, coverage and timeliness of the economic data we have to make forecasts is not as good as in the UK as a whole. That means the forecasts of the Scottish Fiscal Commission will always – through no fault of their own – be subject to a greater margin of error.
Secondly, building the analytical capacity to model, forecast and evaluate the likely impacts of different policies or changes in economic changes, will take time and we can’t expect the Fiscal Commission to work miracles overnight.

Finally, the process for reconciling these forecasts with actual data is itself complicated and will require careful management.
The framework requires forecasts to be made in advance of an upcoming budget for how much revenue is expected to be raised in the forthcoming year. For income tax, and ultimately VAT, the amount of money forecast will then be transferred to the Scottish Budget by the UK Government. But the actual data will not be known with certainty until further down the line.
Once known, an adjustment will have to be made to reconcile the amount of money allocated in the first instance – which was based on that initial forecast – with what has actually been collected. This is because, in effect, HMRC will collect income tax revenues in Scotland but the UK Government has effectively written a cheque to the value of the Scottish Fiscal Commission’s forecast.
This process will take time to work through. And it creates the prospect, as this diagram shows, of forecast errors made in one year, still having an impact on the budget some 8 or 9 years down the line!

The new fiscal powers provide the Scottish Government with significant new powers and the opportunity to capture the benefits of economic success.
But depending upon how you view VAT assignation, between 50% to 60% of the Scottish Budget will still be determined by the spending and tax policy choices of the UK Government.
The fiscal framework attempts to bring coherence to the arrangements for taxation, borrowing, budget management and public spending.
But it’s complex and whilst it opens up new opportunities, it also exposes the Scottish Government to risks, some of which it has little or no scope to influence.
Whether the potential benefits outweigh the risks remain to be seen. But whatever happens, it is clearly going to be an interesting few years!