Thursday 26 August 2021

GERS 2021 - So What?

Every year the Scottish Government's Chief Statistician provides an updated analysis of the state of Scotland's public sector finances by publishing the GERS report [Government Expenditure & Revenue Scotland 2020-21].

This really shouldn't be a very controversial event, but since the independence referendum in 2014 each publication of GERS has been met with a veritable tsunami of media commentary and online debate, with both sides of the constitutional debate seeking to spin the figures in their favour.

In recent years, the GERS figures have shown Scotland's fiscal deficit (as a percentage of GDP or on per person basis) to be much larger than the UK's. Unionists claim this demonstrates the value of fiscal pooling & sharing within the UK while nationalists suggest it somehow proves that Scotland would be better off as an independent nation or - particularly in stagnant backwaters on social media - they seek to undermine trust in the figures. 

GERS Deniers

Those nationalists who promote GERS-denial are indulging in a strategy which is explicitly1 intended to distract from the simple economic facts. To get dragged into defending the integrity of the GERS figures is to play into GERS-deniers' hands. They don't need to win the arguments, they just need their supporters to see the figures being argued over and let confirmation bias do the rest: "People are arguing about the figures so they obviously can't be trusted - I'll just go with what my gut tells me."

The inherent advantage GERS-deniers have was neatly summarised by the influential thinker George Horne back in 1786:

"Pertness and ignorance may ask a question in three lines, which it will cost learning and ingenuity thirty pages to answer. When this is done, the same question shall be triumphantly asked again the next year, as if nothing had ever been written upon the subject." 

Elsewhere on this blog I have written more than my share of "thirty pages to answer" those trying to undermine the figures [most recently here]. But all anybody really needs to know to dimiss the vast majority of GERS-deniers are these three indisputable facts: 

  • The decision to publish the GERS report is the Scottish Government's alone [as confirmed by this FOI response]
  • GERS is compiled by the Scottish Government's own team of statisticians and economists (in St Andrew's House, Edinburgh) using methodologies and assumptions they have chosen following years of extensive consultation
  • GERS is an accredited National Statistics publication and carries the quality mark to prove it

Now that's out of the way, let's focus on what the GERS report actually tells us.

The GERS Deficit

In essence the GERS report is a very simple publication which sets out to address just three questions:

  1. What revenues were raised by Scotland?
  2. How much did the country pay for the public services that were consumed?
  3. To what extent did the revenues raised cover the costs of these public services?
Throughout this blog all figures quoted are on the basis that Scotland is allocated a "geographical share" of North Sea revenue. In layman's terms this means Scotland gets to keep the revenues generated by "Scotland's Oil"2. So we need just three numbers to answer those question for 2020-21:
  1. £62.8 billion of revenues were raised by Scotland
  2. £99.2 billion was the cost of public services consumed by the country
  3. £36.3 billion was the difference between revenues raised and the costs of these public services
So the GERS figures show Scotland (as part of the UK) running a deficit3 of £36.3 billion or 22.4% of GDP (compared to the overall UK deficit of 14.2% of GDP).

In the context of the independence debate this 8.2% deficit gap is an important number, because (all else being equal) it shows how much bigger an independent Scotland's deficit would be than that Scots currently share as part of the UK. Of course all things wouldn't be equal, but we'll come to that.

The following point can't be stressed too much: while Scotland is an integral part of the UK, the GERS deficit is merely a paper exercise:
  • Scotland's public spending is not constrained by the GERS deficit, because the capacity for public spending in Scotland is ultimately determined by the UK's overall economic position
  • Scotland is not accruing a debt liability as a result of the notional GERS deficit, because Scotland pools and shares its deficit and debt with the rest of the UK

The Deficit Gap (aka the Fiscal Transfer)

While some fringe nationalists might want to argue that Scotland shouldn't be responsible for any of the UK's debt, the GERS figures expose how morally indefensible that position is. Scots get their fair share of benefit from UK public spending, it would be outrageous to suggest they shouldn't bear any responsibility for the debt used to fund it.

What is widely accepted (and is implied within the GERS calculations) is that Scotland should bear a population share of the UK's shared debt4.

Those Nationalists who try to dismiss any talk of a Fiscal Transfer "because it's borrowing on our behalf" are simply misunderstanding what the Fiscal Transfer is: it's the difference between Scotland's population share of the UK's deficit (i.e. Scotland's share of the increase in the UK's debt) and the GERS deficit (i.e. Scotland's actual contribution to that increase in the UK's debt).

To illustrate for 2020-21 when the UK's deficit was £297.7bn:
  • Scotland's 8.1% population share of the UK's deficit was £24.3 billion - this is the implied increase in Scotland's share of the UK's debt liability
  • Scotland's actual contribution to the UK's deficit (i.e. the GERS deficit) was £36.3 billion - this is the implied contribution Scotland's economy actually made to that debt liability 
  • The difference - the benefit to Scotland of only having to assume a population share of the UK's deficit (in the form of a population share of the UK's debt) was £12.1 billion5
So in 2020-21 GERS shows an implied fiscal transfer between the rest of the UK and Scotland of £12.1 billion - that's £2,210 for every man, woman and child in Scotland.

To reiterate (because it comes up time and time again): this is absolutely not funded by debt which Scotland is expected to assume a liability for, it is the difference between Scotland's population share of the UK's debt and Scotland's contribution to that debt.

In a final attempt to hammer this point home: anybody arguing that the fiscal transfer doesn't exist would have to argue that Scotland's share of the UK's debt went up by £36.3 billion last year (i.e. that Scotland would be responsible for 12.2% of the increase in the UK's debt, because Scotland was responsible for 12.2% of the UK's deficit6). I don't think any of them are arguing for that. 


Isn't this just a snap-shot?

GERS actually provides data going back for 23 years. While most commentary focuses on the absolute size of Scotland's deficit over that time, this blog is more interested in understanding how the Fiscal Transfer (aka Deficit Gap) has changed:


As the graph above clearly shows, the Fiscal Transfer in Scotland's favour has been well over £10 billion a year for each of the last six years. In fact, the data shows that Scotland has only been a net fiscal contributor to the UK in two of the last 23 years7

We will come on to show that the historical fluctuations in the fiscal transfer are almost entirely explained by movements in North Sea revenues.

While the GERS data only goes back to 98-99, the Scottish Government's Scottish National Accounts Project (SNAP) previously attempted to take the analysis back to 1980. These older figures need to be treated with some caution (they appear generous to Scotland8), but they do illustrate that in the years when North Sea oil was booming, Scotland could be considered a very significant net fiscal contributor to the UK.



This graph is evidence of pooling and sharing within the UK working over time. When North Sea oil was booming, Scotland was a major contributor to the UK; now North Sea oil revenues are a mere trickle, Scotland is a major beneficiary. 

As I have observed elsewhere, the direction of the fiscal transfer doesn't seem to faze Scottish nationalists - in whatever direction the transfer flows, it will be argued as a reason to break up the UK.

That said, the longer the current economic situation persists, the harder it becomes for nationalists to sustain the "Westminster stole our oil" grievance. Over this 41-year period - the longest period for which we have the data and a period which assumes a starting point highly favourable to the nationalists' argument - Scotland is in fact a net beneficiary of fiscal pooling and sharing within the UK (the red bars outweigh the black).

This is neatly illustrated by the cumulative picture:




This shows that, as of now, Scotland has "got back" from the UK (largely in the form of higher public spending) more than it "put in" during the 1980's as as result of the oil boom years.


If Scotland's Deficit is so large, doesn't that prove being in the UK is bad for Scotland?

The simple answer to this question is: No.

To explain why, we need to understand the extent to which that larger deficit is a function of lower revenue generation (relatively poor economic performance) and to what extent it is due to higher public spending (doing relatively well out of pooling and sharing).

The GERS report puts Scotland's revenue and spending numbers in context by comparing them to figures for the whole of the UK (including Scotland) on a per person basis. This shows us that in 2020-21:
  1. Scotland generated £382 less revenue per person than the UK average [Table S.4]
  2. Scotland's spending was £1,828 more per person than the UK average [Table S.6]
Although the report doesn't put this figure in a summary table, the inevitable result of the figures above is that Scotland's deficit per person is £2,210 greater than the UK average. Multiply that £2,210 per head by Scotland's population of 5.47m and we've found another way to get the implied Net Fiscal Transfer of £12.1 billion. Analysis is satisfying sometimes, isn't it?

This tells us that more than 80% of the Fiscal Transfer is explained by the fact the UK's system of pooling and sharing leads to higher spending per person in Scotland. It's hard to see how even the most committed separatist could argue that this is a bad thing.

Once we appreciate that the Deficit Gap (aka Net Fiscal Transfer) is the summation of differences in revenue generation and spending per person, we can squeeze some more insight from the historical GERS figures. 

In the graph below:
  • The red line shows how much higher Scotland's spending per person has been than the UK average over time
  • The black line shows how much higher/lower Scotland's revenue per person has been
  • The green line shows how much lower Scotland's onshore revenue (i.e. excluding North Sea revenues) has been

The gap between the red and black lines is the Fiscal Transfer; the gap between the red and green lines is the Onshore Deficit Gap (i.e. the size of Fiscal Transfer that would exist if there were no North Sea revenues).

There's a lot to digest in this graph:
  • The red line shows that the gap between spending for Scotland and the UK average has actually grown over this period9
  • The green line shows that Scotland's Onshore Revenue generation only slightly lags the UK average, although that gap grew materially between 2014 and 20171O
  • Throughout this period there has been a large Onshore Deficit Gap (the gap between red and green lines) which is revealed when North Sea revenues decline
  • The scale of the Onshore Deficit Gap is consistently explained largely by relatively higher Scottish public spending
  • Fluctuations in the Fiscal Transfer are dominated by fluctuations in North Sea revenues (the gap between the black and green lines) - when oil revenues decline, the Onshore Deficit Gap is revealed. This was clear in 2014 and illustrates why the economic case for independence was always a reckless gamble on future oil revenues

Isn't this because GERS unfairly allocates UK-Wide spending to Scotland?

In a word: No.

In GERS, UK-wide costs considered to benefit the whole of the UK (e.g. Defence, Overseas Development Aid, UK Debt Interest, Whitehall Departmental Costs, DWP and HMRC overheads etc.) are allocated to Scotland on a per person basis (i.e. Scotland is allocated an 8.1% population share of these costs). This means that by definition there is no difference between the cost per person allocated to Scotland and to the rest of the UK in these categories.

How much an independent Scotland might spend to replicated the services delivered by these currently shared UK costs is a not a question the GERS report is designed to answer. But as I argue elsewhere, it seems highly unlikely that an independent Scotland could replicate HMRC, DWP, Treasury, FCO, DFID, Cabinet Office, BEIS etc. functions for less than the 8.1% of these UK departments' costs which are allocated to Scotland in GERS. 


So why is Scotland's spend per person so much higher?

The table below shows where the £1,828 per person higher expenditure comes from in 2020-2111


Note that there is no difference in spend per person on Defence and International Services because, as explained above, these are allocated in GERS on a per person basis.

Scotland has a lower population density than the UK average, extensive remote and island communities and particular demographic challenges which makes it inevitable that it will cost more per person to deliver equivalent services in some categories. Whether the actual level of higher spend the GERS figures reveal can be entirely justified on a like-for-like cost-to-serve basis is an unanswered question.

The detail behind these figures deserves an entire blog in its own right, but  to pick some highlights;
  1. The biggest absolute difference in spend per person is on Social Protection (i.e. pensions and benefits), an area which is almost entirely reserved to Westminster12. This demonstrates a fundamental but too often over-looked point: under current constitutional arrangements, Scotland is guaranteed to get its needs-based share of reserved social protection spend. The amount spent in scotland is determined directly by how many people in Scotland qualify for the relevant pensions and benefits, no Barnett Formula calculations or inter-governmental negotiation are required. Scotland gets a significantly higher share of Social Protection spending than the UK average as a result and - assuming one believes in the principle of UK-wide social solidarity - that is demonstrably fair (and an advert for the merits of reservation).
  2. Those seeking to undermine the GERS methodology often make an uninformed fuss [as discussed at length here] about the Transport spending allocation, mainly because it used to include an allocation for HS2 costs which were not incurred in Scotland. This was never a material issue, but in a welcome move the Scottish Government's statisticians have chosen to now completely exclude any HS2 allocations from the GERS figures. This means that there are literally no infrastructure costs which take place outside Scotland (transport or otherwise) allocated to Scotland in GERS13
  3. The fact that Public and Common Services is relatively so much higher than the UK is another trigger for those who like to look at a number they think doesn't feel right and cry "foul!". Scotland has a large public sector and these numbers are entirely reasonable and fully explained within GERS, as detailed here
  4. The figure which may surprise some is that Health spending is only 1% higher per person in Scotland than the UK average.  We will discuss this further in the next section.

What does this tell us about the Scottish Government's priorities?

Perhaps the most interesting way to look at the realtive spend per person data is to consider how it has changed over time. The graph below shows the differences between Scotland and the average of rest of the UK spend per person over time, by category (the health spending line has been highlighted):


The relative decline in health spending over recent years is particularly marked. It is of course the case that funding increases in the health service in the rest of the UK are automatically passed on to Scotland via the Barnett Formula, so the decision to deprioritise spending on health has been the Scottish Government's alone. This shouldn't come as a surprise given that in April 2021 (i.e. using 2019-20 figures) the IFS observed "Official estimates suggest Scottish health spending per person now 3% higher than in England, compared with 22% at the start of devolution".  

Again we could write an entire blog on what each of these lines tells us, but it's perhaps worth drawing attention to three of them:
  • The relative increase in education spending 
  • The relative increase in Public Order & Safety (i.e. Police) spending - presumably largely due to the centralisation of Police Scotland
  • The recent relative up-tick in enterprise and economic development spending
Whether these relative spending increases have delivered improvements in service delivery is, at best, a moot point.

There is one other piece of information buried in the GERS tables which seems worth highlighting. Table 3.2 shows that in the last year Scottish Local Government spending declined by 0.9% whereas in the UK overall it increased by 16.5%.  It is hard to see how one could interpret this as anything other than the SNP Government centralising and taking power away from local government. 


Aren't these figures meaningless when it comes to the case for Independence?

The GERS figures are historical actual figures, so they can only reflect the performance of the Scottish economy under current constitutional arrangements. That is: sharing the UK's machinery of state, operating within a borderless UK single market, sharing a central bank and currency, pooling and sharing resources and accepting that some tax and spend decisions are reserved to Westminster.

Those caveats matter, but they certainly don't make the figures meaningless: they still tell us what revenue the Scottish economy currently generates and how much it costs to deliver the public services that Scots are used to receiving. They provide a base-line against which those wishing to make an economic case for separation can argue what would change.

None of that prevents us being able to draw some simple conclusions from the figures as they stand, as the SNP proved in 2014 when their chosen line on the GERS figures was:
"Scotland accounted for 9.3% of UK public spending between 2008-09 and 2012-13, while generating 9.5% of tax receipts - it put in more than it got out. It suggests that tax receipts are currently 14% higher in Scotland than the rest of the UK"
The latest GERS figures allow us to update this statement for the most recent five year period:
"Scotland accounted for 9.2% of UK public spending between 2016-17 and 2020-21, while generating just 7.9% of tax receipts - it put in less than it got out. It suggests that tax receipts are currently 3.5% lower in Scotland than the rest of the UK"
So back when the SNP could lay claim to significant North Sea revenues, the GERS figures were used to argue that Scotland "put in more than it got out" because Scotland had higher per person tax receipts which more than offset Scotland's higher per person spending14.

Since then North Sea revenues have declined (to the surprise of nobody who was paying attention in 2014) and so the same analysis now shows that despite generating slightly [3.5%] lower tax receipts than the the UK average, Scotland continues to benefit from significantly [11.2%] higher public spending.


Is this fair?

Whether all of this is fair or not depends on your perspective. It can certainly be argued that this is merely pooling and sharing of resources across the UK working effectively over time. When oil boomed in the 1980's,  Scotland "put in more than it got out"; now oil revenues have declined to a trickle, Scotland "gets out more than it puts in".

From a less transactional and more philosophical perspective: if you believe that fellow citizens of this union should not have the quality of their healthcare, education or social welfare constrained by the revenue generating performance of their constituent nation (or region) then yes, this is probably fair.

If you take a narrower view and think that pooling and sharing is weakness, that a tighter line should be drawn around the groupings that should have to "stand on their own two feet" then I guess you will see the fiscal transfer revelaed by the GERS figures as a reason to break up the UK.

I know where I stand.

*****

Notes

1. It was explicitly stated in one of those "Ten Things That Will Help Us Win Independence" type articles in The National some years back, but I confess I can't be bothered to find it again
 
2. In case anybody doubts that: in recent years Scotland has actually been apportioned more than 100% of the UK's North Sea Revenues, because PRT rebates are weighted towards fields in English waters

3. aka a negative Net Fiscal Balance 

4. This was the starting position assumed in the Independence White Paper and the SNP's more recent Growth Commission report - and the GERS report itself allocates a population share of the shared UK debt interest to Scotland. For detail geeks: public sector pension fund interest receipts and expenditure are known for Scotland and - because Scotland's public sector employment is larger - the share of both is larger than Scotland's population share (which excites some people when they see the allocation on the cost side, but per GERS p.7 Q2, the receipts and expenditure largely cancel each other out anyway)


5. There's a rounding effect in the number presentation - with one more decimal place we can see £24.26 billion minus £36.34 billion = £12.08 billion

6. 

GERS-denier in Chief Richard Murphy has stated in previous years "I have been continually bemused by the fact that GERS [..] says that Scotland runs a deficit so  much larger in proportionate terms than that for the UK as a whole [..] Proportionately the Scottish deficit is suggested to be, after North Sea revenue is taken into account, 3.45 times that of the UK as a whole". I can only presume he will be even more bemused now that the deficit has grown and Scotland's deficit is proportionately only 1.6 times larger than the UK's as a whole. 

Or let's look at what Iain McWhirter said in the Herald last year: "But even then, as the tax expert Professor Richard Murphy has pointed out, it is absurd to claim, as GERS has in recent years, that Scotland with 8% of the UK population accounts for nearly 60% of the entire UK deficit. This just doesn't make sense." It was never absurd and just because Iain couldn't get his head around the maths involved didn't mean it didn't make sense (as I have explained many times, for example here and here). Now the absolute deficit has grown, Scotland is only responsible for 12% of the UK's deficit - I can only assume Iain will see this as further proof of the absurdity of fiscal arithmetic.

7. The figures being used in 2014 showed a slightly more favourable position for Scotland, but the GERS figures have since been revised, most notably by amending with more accurate (lower) North Sea revenue figures

8. Whereas historical GERS figures are continually restated to reflect changes in accounting policies and more up-to-date understanding of the data, the SNAP figures were last updated in 2012-13 and so will not be strictly comparable. In the years where SNAP and GERS overlap, the SNAP deficit is on average £1.5bn lower than the GERS deficit (i.e. the SNAP data is favourable to Scotland compared to our more recent understanding). By inspection, the net difference is mainly explained by higher spending allocations in GERS compared with historical SNAP data.

9. Some people assume the "Barnett Squeeze" should inevitably lead to convergence in spending - this isn't the case because of differential rates of population growth between Scotland and the rest of the UK. If you think about it: if spending didn't change at all but Scotland's population grows more slowly than the rest of the UK's, Scotland's spend per head will be less diluted by population growth than that of the rest of the UK. For more on this topic, see here > What is the Barnet Squeeze

10. The reasons behind this deserve a blog of their own, but some obvious factors to consider are
  • The decline in the North Sea industry will have had knock-on effects on the onshore economy
  • Increases in personal tax rates in recent years have either led to a change in, or a better understanding of, the actual number of higher rate tax payers in Scotland
  • The ongoing constitutional uncertainty, which can surely not have helped business confidence
11. For those who are curious, here is the same data vs rUK as well as vs UK (including Scotland)


12. see Box 3.2 and Table 4.4 in the GERS report

13. You can check this by surfing the "supplementary-expenditure-database" which accompanies the GERS report - the biggest source of confusion is the allocation of UK Government Network Rail costs (the organisation responsible for Scotland’s rail infrastructure) which is quite reasonably allocated based on actual Train Operating Company usage in Scotland

14. Since those figures were quoted, the GERS figures have been revised down (mainly by correcting down the oil revenues attributed to Scotland) - so using the updated figures for the same time period we would actually say: "Scotland accounted for 9.3% of UK public spending between 2008-09 and 2012-13, while generating 9.3% of tax receipts". So in fact even during those relative boom years for North Sea revenue, Scotland merely paid its way on the SNP's own terms

Wednesday 25 August 2021

Richard Murphy on GERS: The Perpetually Baffled Man

I'm writing a serious blog on GERS and really don't want it to get bogged down by engaging with the frankly bonkers claims made by the perpetually baffled GERS-denier in chief, Richard Murphy. 

To engage with Murphy on the topic of GERS is as pointless and frustrating as trying to play chess with a pigeon (while wrestling a pig). But I recognise that some of you might think that describes an amusing spectator sport, so let me use this blog to deal with some of his recent - um - "contributions to the debate".

My starting point is his appearance on the "GERS - Scotonomics Special". This is an hour of video and I only made it 5 minutes in before I realised that I simply couldn't sit through any more of it because life, like me, is too short. 

The fact that this entire blog is required to unpick what is basically just one minute of his extraordinarly comprehensive wrongness I hope explains why I will try to avoid engaging in any further pigeon-chess-playing-while-pig-wresting. 

Suffice to say that this blog alone should be enough to convince any rational observer how spectacularly misplaced Murphy's arrogance is when it comes to criticising the work of the Scottish Government's statisticians. If more convincing is needed, see here and here or - if you want to see him taken apart in front of a Holyrood committee by the inestimable Jackie Baillie - here.

The point where I gave up the will to go on appears at about 3:30 in that video - I quote him here verbatim:

"... when you look at the expenditure side, what you actually see [is] the vast proportion of the expenses of the UK are apportioned to Scotland on the basis of population - 8.1% - and yet I put out figures on the blog this morning and show that the vast majority of the expenses actually represent more than 8.1% of the total UK spend and in some cases vastly higher .. in terms of Public Administration costs something like 13 to 14% of the total costs of the UK are apparently incurred in Scotland that's why I think frankly what's happened is they've taken the UK figure and then they've taken the figures for running Scotland and added the two together not allowing for the fact that some things obviously aren't done twice in Scotland ..,. they've done that sort of crass stuff and I'm afraid to say I just think that this is really poor accounting [and on and on he goes ...]"

Honestly it's hard to know where to start with this level of wrongness, but let me try and be calm and forensic.

The blog of his he refers to is the typically boldly titled "Why GERS is wrong - yet again". That blog contains even more nonsense which it's hard not to get distracted by, but I'll relegate the most obvious howlers to a footnote1 so I can wrestle here specifically with his bizarre assertions about Public Administration (by which his blog makes clear he means the GERS line item "Public & Common Services") cost allocations.

The figure that has him so exercised is £2,410m of Public & Common Services expenditure allocated to Scotland [see Table 3.1 on page 28 of the GERS report] which is indeed 13.7%c of total UK costs.

First of all, the fact that he admits to being baffled by how much higher than the UK average Scotland's Public Administration costs are shows how poorly he understands Scotland.  If he'd read page 7 of the GERS report (remembering that Scotland accounts for just 8.1% of the UK population) he would have seen a broad hint:

"... around 9.1% of UK spending is undertaken for Scotland, slightly higher than a population share. While direct estimates of spend in Scotland are not available, this is consistent with broader indicators of public sector activity in Scotland, which show that the public sector plays a larger role in Scotland than the UK as a whole. For example, around 10% of UK public sector employees are based in Scotland, with regional pay differences resulting in around 9.3% of the UK paybill spent in Scotland." - GERS page 7.

If he'd made it as far as page 35 he's have seen Table 3.8 in GERS shows that £1,683m of the £2,410m of Public & Common Services costs in GERS are directly incurred by the Scottish Government and Local Authorities. So what he's getting exercised about is in fact an allocation of £726m for Public & Common services delivered by the UK Government.

Cross-referenceing Tables 3.8 and 3.7 would have shown him that this £726mn is an allocation of just 4.6% of those UK Government Costs2. So his wild assertion that 8.1% of all UK Public & Common Services costs have simply been added on to Scotland's costs is, obviously, wrong. Suggesting that such a basic error could have been made merely advertises the fact that he has no conception of the depth and detail of the work that goes into compiling the GERS figures.

But is that 4.6% reasonable? 

Well intuitively it doesn't sound weird given how much of the UK's integrated machinery of state Scotland still relies upon and we're dealing here with only £0.7bn out of a £36.3bn deficit.

But we don't have to rely on intuition, we can dig deeper. The "supplementary-expenditure-database" which accompanies GERS provides detailed line-level detail to back-up all the GERS spending allocations for the prior year (when £624m of UK Government Public & Common Services costs were allocated to Scotland).

This database shows us that roughy half of those UK government costs3 allocated to Scotland are a simple population share of HMRC costs. Now HMRC is a big department and it's a shared UK resource. This cost allocation is not duplicating anything  it's not something "done twice in Scotland" so - surprise surprise - none of Murphy's alleged double-counting is going on here. 

As an aside: frankly it's highly unlikely that Scotland could replicate the full service offered by HMRC for as little as 8.1% of what it costs the UK today, so this could be argued as a very good value deal - but that's a different question for another day. 

These HMRC costs helpfully illustrate the wrong-headedness of another of Murphy's frequently voiced claims: that GERS is wrong because we're incurring spending outside Scotland and not seeing the tax generating benefits of that spending in Scotland. A quick google would lead him to the UK's civil service statistics which tell us that 7,730 HMRC employees are based (and generating tax) in Scotland.  That's 12.8% of them so, even allowing for salary differentials, more HMRC employee costs are incurred in Scotland than allocated to Scotland in GERS. [A similar situation exists with DWP staff: 8,820 or 9.7% of DWP employees are paying tax in Scotland].

Another aside: that Civil Service headcount data allows us to cross-check the figure quoted on page 7 of GERS - and sure enough 47,590 or 9.97% of all UK-based Civil Servants are based in Scotland. It's almost as if the team who sit in St Andrew's House working on the GERS report have a better understanding of what's going on than Murphy does, isn't it?

So we're now down to explaining just £0.3bn of Public & Common Services costs incurred by the UK Government and charged to Scotland in GERS. The detailed database shows us that £0.2bn of them relate to a population share of HM Treasury, the Cabinet office and the Department for International Development (DFID). Whether Scotland could replicate these services for just 8.1% of the UK's costs charged in GERS is again open to debate (a debate to which I predict the conclusion would be: no). That said, although 18% of UK-based Foreign, Commonwealth & Development Office4 employees are based in Scotland, it is of course true that the vast majority of HM Treasury and Cabinet Office employees are in London. 

The £0.1bn that's left (£142m if you really care) is Scotland's share of House of Commons, IPSA, House of Lords and a few other bits and bobs.

So there are some costs which occur in England that are allocated to Scotland - but equally there are some costs that occur in Scotland that are allocated to England (just think of those HMRC and DWP employees or see below to learn about how Scottish ferry costs are treated). 

Murphy blusters that he is sure the amount spent in England which is allocated to Scotland (and could be transferred to Scotland to generate more tax) is "some billions" - needless to say that is a nonsensical number based on nothing more than a cumulation of ignorant assertions.

I've done the analysis in detail here, but the simple answer is it wouldn't be more than a couple of hundred million at most (and you can't both save costs and transfer them!). 

If you can work up a grievance out of the possible fiscal multiplier effect from moving maybe a couple of hundred million of spending within the context of what is a £12 billion fiscal transfer5 from the rest of the UK to Scotland ... well good luck to you.

***

Not only could I do this exercise for every expenditure category in GERS, I have done. I know right? What kind of a lunatic would spend time familiarising themselves with the detail behind the figures before opining on how they were compiled?

Having been through that detail with a fine-tooth comb, I do have some disagreements with the GERS methodology. For example I think the EU contribution calculation attributes too high a share of the UK's rebate to Scotland and I think the treatment of Renewable Obligation Certificates understates the effective subsidy from the rest of the UK to the Scottish Renewables sector. But that doesn't mean I rubbish GERS: I make polite enquiries of the team responsible in the Scottish Government to test and build my understanding. Even where I continue to disagree, I recognise the reasonableness of the approach that has been taken.

Not only is it patently obvious that Murphy has never taken the time to check any of his wild theories with the team who actually compile the figures, he clearly hasn't even read the report.

To illustrate my point, here's just one other extract from that YouTube clip which somebody has drawn my attention to. Again I quote him verbatim [from about 5:18]

"HS2 - which is not going to come near Scotland for decades and decades if ever - is being apportioned to Scotland and there's no benefit from it. We know that the mass of public transport investment in the UK is in the South East of England and that's being apportioned to Scotland. This is nonsense..."

He's right, but only insofar as everything he says there is nonsense.

It really couldn't be clearer, it's there in black and white on GERS page 11: "in this edition of GERS none of the expenditure associated with High Speed 2 is allocated to Scotland".

HS2 used to be the exception that proved the rule, because none of any other "transport investment ... in the South East of England" has ever been allocated to Scotland in GERS6

It's rather useful that HS2 is now completely excluded from GERS (it's been removed from prior years as well) because

  1. This saves me having to explain how relatively trivial that HS2 figure was - just 2.3% of HS2 costs used to be allocated in GERS (based on the business case assessment of the knock-on value to the Scottish economy) and last year that would have been just £75m

  2. There is now only one example where GERS allocates transport infrastructure spending in one part of the country to another - and that's where £69m (30%) of the Scottish Government's Ferries costs are allocated to the rest of the UK7. I wonder why Murphy hasn't kicked up a fuss about that?

You do have to wonder why, every time he's "baffled" or "simply does not understand" GERS, Murphy leaps to the conclusion that the figures must be wrong and the Scottish Government's own economists must be lazy and/or incompetent. Surely it couldn't be that he's a grifter who's spotted a chance to build a profile by saying what cybernats and reality-denying nationalists want to hear?

As a parting thought: imagine being part of the team that has spent years refining the GERS methodology and compiling the reports, only to have this man - on the basis of transparently superficial "analysis" and without bothering to gain even the most rudimentary understanding of the data he's dealing with - to accuse you of doing "crass stuff" and being guilty of "really poor accounting".

He really is beyond the pale.


Chokka-bloke, checking out.

***

Footnotes (the first one's a doozy)

1. Some other howlers in Muprphy's blog

  • "Scottish data is inflated compared to the UK by including local pension fund costs, which are not in the UK data, which you might say is odd" - I would say it is odd, because his assertion that those costs are not in the UK data is simply and obviously untrue! The figures are shown in the supplementary-expenditure-database which accompanies GERS (something Murphy has clearly never troubled himself to look at), where we can see that Scotland is allocated £2.0bn of the UK's £17.2bn of public sector pensions debt interest.
  • He's goes on to be baffled by debt interest allocations - "why has Scotland got a disproportionately large share of interest payments, picking up almost 15% too much?". It's not 15% "too much", as he would know if he bothered reading the GERS report. Below are direct quotes from the main GERS report which explain clearly: why Scotland picks up a higher share (9.3%) of debt interest payments than our population share (8.1%); that this is entirely explained by a correction to the historical treatment of public sector pension fund interest expenditure; that this is consistent with the way interest income is apportioned (and so the net effect on the stated deficit of these public sector pension fund interest expense and income allocations is roughly zero); that these are local government pension funds where data are available for Scotland.  
    • "GERS includes two categories of interest spending. The first is reserved UK debt interest, and Scotland is allocated a population share of this, amounting to £2.3 billion in 2020-21. The second is interest spending associated with public sector pension funds. These funds also generate interest income, and in 2020-21 Scotland is apportioned £1.5 billion of interest expenditure associated with public sector pensions, and £1.4 billion of interest income" GERS page 7
    • "In previous editions of GERS, Scotland has been allocated a population share of all UK public sector interest expenditure. In this year’s publication, this has been changed to separate out interest expenditure which is associated with public sector pension funds. These are local government funds which generate both interest revenue and expenditure, and where data are available for Scotland. This change brings the treatment of expenditure associated with these funds into line with the treatment of their revenue, which was already allocated to Scotland using Scottish data" GERS page 11
    • Consistent with the CRA, interest expenditure by public sector pension funds is shown as spending by HM Treasury." GERS page 30
    • "The main methodology change in GERS 2020-21 relates to the treatment of public sector interest expenditure, where interest expenditure associated with Scottish Local Government pension funds is separated out. All this expenditure is treated as Scottish, rather than apportioning Scotland a population share." GERS page 25
[Given Scotland's larger public sector, the fact that public sector pensions costs are higher than the UK average is singularly unsurprising]

2. For those who care about following the audit-trail
  • Numerator: £726 [Table 3.8]
  • Denominator: Total UK P&CS costs of £17,524 [Table 3.7] less Scottish P&CS costs of £1,683 [Table 3.8]= £15,841

3. £294m out of £624m to be precise - so 47%

4. In September 2020 the Foreign & Commonwealth Office (FCO) and Department for International Development (DFID) merged to form the Foreign, Commonwealth and Development Office (FCDO)

5. This will be covered by my other blog, but it's an easy enough number to calculate:
  • Scotland's 8.1% population share of the UK's deficit was £24.26billion - this is the implied increase in Scotland's share of the UK's debt liability

  • Scotland's actual contribution to the UK's deficit (i.e. the GERS deficit) was £36.34 billion - this is the implied contribution Scotland's economy actually made to that debt liability
     
  • The difference - the benefit to Scotland of only having to assume a population share of the UK's deficit (in the form of a population share of the UK's debt) was £12.1 billion
So in 2020-21 the GERS figures show an implied fiscal transfer between the rest of the UK and Scotland of £12.1 billion - that's £2,210 for every man, woman and child in Scotland.

6. at least not in the 8 years that I've been studying these figures

7. per the "supplementary-expenditure-database" which accompanies GERS: 30% of Scottish Government costs associated with Caledonian Maritime Assets Limited and Ferry Service Department are apportioned to rUK (reflecting an assessment of the share of the value of Scottish ferry services enjoyed by travellers from the rest of the UK)