Mind The Gap: Critiquing the FT's Fiscal Gap Analysis

---

tl;dr
The FT recently published a figure of £1.8k per person pa. as the tax rises or spending cuts an independent Scotland would need to achieve sustainable deficit levels. By failing to quantify the impact of border friction and Scotland's need to create its own currency, this figure drastically understates the true scale of the longer-term economic challenge an independent Scotland would face - taking into account those factors, a longer-term figure of £4k per person would be more realistic

---

The Financial Times recently published an analysis ["Independent Scotland would face a large hole in its public finances"] which  summarised the implications of how Scotland's fiscal position has deteriorated since 2014:

"Based on the pro-independence Scottish National party’s previous assumptions, this would mean Scotland needed to raise taxes or cut public spending annually¹ by the equivalent of £1,765 per person in the period after exiting the UK so as to narrow the deficit to sustainable levels."

Anybody thinking £1.8k per person appropriately scales the economic challenge represented by separation has failed to realise the significance of that opening caveat: "Based on the pro-independence Scottish National party’s previous assumptions ...". 

The FT's headline figure significantly under-sells the scale of the fiscal challenge separation would cause; that caveat is akin to prefacing an analysis of the economic implications of Brexit by saying "Based on what the Leave campaign wrote on the side of a bus ...".

To help explain why the true challenge is so much greater than this analysis suggests, here's the FT's own summary chart

Let's first look at what has been included before considering the more important issue of what has not been included.

What has been included are "proposed spending cuts" of 1.2% of GDP². This figure is sourced from the SNP's Sustainable Growth Commission and - having spent more time than is healthy with that report [see Growth Commission Response: A Reality Check] - I am familiar with what lies behind the 1.2% saving and 5.9% "2018 financing gap"³ deficit assumption:

The 1.2% saving assumption is made up of 0.4% on defence spending, 0.4% on debt servicing and 0.8% on "UK Government spending allocated to Scotland", offset by 0.4% from oil revenues being diverted to a proposed "Fund for Future Generations".

The FT rightly notes "Although there were many assumptions made by Wilson that would likely be questioned by the UK government, his report was described as “not implausible” by the Institute for Fiscal Studies, an influential think-tank". 

As it happens, I am confident that I have carried out deeper analysis on "UK Government spending allocated to Scotland" than the IFS have. On that basis I have no hesitation in asserting that the 0.8% of GDP saving assumption is, in fact, deeply implausible. There are three reasons for this;

1. For an independent Scotland to make savings versus these reserved cost allocations, we would need to believe that standalone Scottish Treasury, Home Office (including Border Force), Cabinet Office, DCMS, DBEIS and National Crime Agency functions would cost less than 8.2% of the amount the UK currently spends on these departments (i.e. than the population share of these departments' costs allocated to Scotland in GERS). The empirical evidence shows that, far from proportionately costing less, government functions devolved to Holyrood require roughly twice as many civil servants per head of population than when those departments benefit from operating at the scale of the UK⁴.
2. The larger part of the SGC's "saving" comes from assuming that £2.4 billion of "spending that is allocated to Scotland but takes place elsewhere" could be "transferred to Scotland " and generate "revenue benefits of £0.6 billion"⁵ This assumption is based on a fundamental misunderstanding of the reserved spending allocations in GERS. I have checked my understanding directly with the Scottish Government economists responsible and am confident that £2.4 billion of transferrable spending simply does not exist⁶. If £2.4 billion of transferrable spending did exist, the SGC would be able to detail it with reference to reserved expenditure line items in the “Detailed Expenditure Database” which accompanies GERS. They won't, because they can't.
3. The SGC ignore the fact that GERS allocates to the rest of the UK nearly £200m of Scottish ferries costs, costs associated with Creative/Historic Scotland and nuclear decommissioning costs which take place in Scotland, all of which would be the responsibility of an independent Scotland
4. Although it's a somewhat theoretical figure at this stage: were Scotland to rejoin the EU, the GERS figures show that Scotland's EU contribution (without the advantage of the UK's abatement) would be £0.7bn

Accepting the SNP's assertion that an independent Scotland would see these net cost savings from UK Government spending allocated to Scotland (outside of defence and debt interest) is akin to accepting the £350m a week written on the side of a bus by the Leave campaign. The true figure in that case was nearer £180m a week (see Thoughts on EU Referendum) - but of course to focus on that figure alone would be to miss the real costs of Brexit.

Similarly if we focus only on the SNP's cost saving assumptions, we miss the true costs of Scexit. So while we could certainly debate the debt interest saving assumptions they make, let's instead consider what is not included in the FT's quantitative analysis.

The FT's assumption of a 9.9% deficit for 2025-26 is broadly in line with IFS projections (graph below) - but these are figures which assume Scotland remains within the UK and the UK single-market

This means that when the FT quantifies the fiscal gap (with spurious precision) at £1,756 per person, they are both taking at face value the SNP's wholly unrealistic "proposed spending cuts" and ignoring the other effects of separation (effects which the article goes on to mention in commentary only).

The article notes "A research paper by the LSE published in February said border problems stemming from Scotland leaving the UK would compound losses from Brexit, and estimated these were likely to result in Scottish incomes being between 6.3 per cent and 8.7 per cent lower in the long term compared with neither event happening."

The full research paper is worth reading [here > Disunited Kingdom? Brexit, trade and Scottish independence]. The figures quoted relate to income per capita and come with these caveats:

• "These numbers likely underestimate the losses caused by higher trade costs, as we do not account for any dynamic effects of trade on productivity"
• "Our estimates should be interpreted as the long-run effects of Brexit and independence after the economy has adjusted to changes in trade costs. Brexit studies typically argue that it will take 10-15 years for the full effects to materialise"

This means translating the LSE's assumptions into 2025-26 deficit gap terms is not straight-forward, but longer term this scale of decline in GDP would lead to a c.3.5% increase in the deficit gap (adding a further c.£1,100 to the per-head figure).

At the risk of labouring the point: the fiscal gap the FT have quantified ignores the economic impact of putting a border between Scotland and the rest of the UK. If you can't see how potentialy misleading that is, imagine an analysis of Brexit which both overstated the saving from stopping the UK's EU contribution and ignored the impact of trade friction from leaving the EU single-market and customs union!

The FT article goes on to mention "A further complication around borrowing relates to Scotland’s future currency arrangements". The genius of the SNP's currency plan is that it is so vague as to be impossible to robustly critique, so the FT is reduced to making the following general observation about borrowing: "With persistently high borrowing unlikely to succeed, an independent Scotland would need greatly improved economic performance to avoid tight spending control or higher taxes."

This massively understates the scale of the currency challenge. As explained in this These Islands paper [Choose Your Poison: The SNP's Currency Headache], the SNP's official strategy of indefinite Sterlingisation is unsustainable, meaning a new Scottish currency would have to be launched sooner rather than later. That means that rather than targeting a 3% deficit, a newly independent Scotland trying to support it's own currency and build credibility on the international capital markets would more realistically be striving to deliver a surplus. Indeed the Sustainable Growth Commission itself observed that “Small advanced economies have made fiscal prudence a strategic priority”, none of their chosen comparison "Small Advanced Economies" runs a deficit as large as 3% and in fact most run a surplus. Add to that the fact that the EU's Fiscal Compact specifies a target 0.5% deficit as measured across the cycle, and it's reasonable to assume an independent Scotland would be seeking to eliminate it's fiscal deficit - that would add another £900 to the fiscal gap.

If we add back the rather speculative 1.2% cost-savings, we increase the gap by a further £400 per head. 

Were we to add to that the currency associated risk (near certainty) of capital and talent flight, the scale of the fiscal gap would widen yet further. 

By now the point is hopefully well made: the FT's headline figure of £1.8k per person for the "fiscal gap" independence would create seriously under-states the true longer-term economic cost of Scexit - just adding together our figures here we can easily see a longer-term fiscal gap of £4.2k per head.

*****

Notes

1. Pedantic point: they mean "cut annual public spending" not "cut public spending annually"

2. [7.1% - 5.9%] = 1.2%; [9.9% - 8.7%] = 1.2%

3. The terminology is a little confusing: the "2018 financing gap" refers to the fiscal position that the Growth Commission (published in 2018) assumed an indy Scotland would inherit in 2021

4. See Civil Service Statistic 2020: I have completed detailed department by department analysis, but the simple sense-check here is that, despite the weight of central government departments in Westminster, it is still currently the case that on a per head of population basis there are currently 27% more civil servants employed in Scotland than in the rest of the UK. This is a reflection of the fact that there are more than twice as many civil servants per head of population in devolved functions and only c.7% fewer in reserved functions (primarily because of the large number of DWP and HMRC employees based in Scotland)

5. Sustainable Growth Commission B4.58

6. The Sustainable Growth Commission appear to have assumed that reserved spending (outside of defence, debt interest, accounting adjustments and international affairs) treated in GERS as "Non-Identifiable" can be relocated to Scotland. Drawing on line-level analysis of the GERS “Supplementary data - Detailed Expenditure Database” [summarised here > What's £7 billion between friends], we can see that this is clearly wrong: specific examples of spending (using 2018-19 figures) captured by this definition which clearly cannot be relocated include:

• Civil Service costs which already take place in Scotland (e.g. DWP & HMRC): £569m
• Overseas Pensions for ex-pat Scots: £336m
• EU transactions (cost of EU membership): £270m
• BBC costs which are already in Scotland: £251m
• Nuclear decommissioning costs which already take place in Scotland: £222m
• Other pensions responsibilities: £90m
• EEA Medical costs: £59m
• Financial Services Compensation Scheme Costs: £41m
• Maritime & Coastguard Agency costs: £26m

The above add up to £1.9 billion of costs which could not be relocated to Scotland and so could not deliver a fiscal multiplier effect - the true figure which could in theory be relocated is just £0.5bn (but these costs can't be both saved and transferred, which is what the SGC assume)