Professor Murphy and Deckchairs

I see Professor Richard Murphy has been offering more of his ineffably obtuse observations on the Scottish Government's GERS figures. As sure as night follows day, we can expect to see a logic-mangling column in The National next week and the usual suspects seizing on his musings to proudly proclaim "see: we know nothing!".

I'm almost impressed by the lengths Murphy and his cheer-leaders will go to trying to avoid facing the simple truth that the GERS figures reveal. That simple truth is that Scotland's notional independent finances look weaker than the UK's in total because we spend far more per person on comparable services than the average of the rest of the UK.

That's it.

There's other stuff going on of course, but the simple explanation for Scotland's higher deficit per capita - the Deficit Gap between Scotland and the rest of the UK - is that we spend more per person in Scotland on public services.

***

Here's all you really need to know [regular readers please forgive the repetition, but this needs to be understood to explain why Murphy's aim is so far off target]:

This graph shows Scotland's relative per capita revenue generation and public spending versus the rest of the UK¹. The green line shows that Scotland’s onshore economy (i.e. excluding oil) consistently generates about £350/person less than the rest of the UK average. The black line shows what happens when we add Scotland’s volatile oil revenues to the picture. When the black line is above the axis this means Scotland generates more tax per head than the rest of the UK (something used as a proud boast by the SNP during the independence referendum but - as the graph clearly shows - only ever the case because of oil revenues).

The red line shows Scotland’s relatively higher public spending, a figure which has risen in recent years to over £1,500/person more than the rest of UK. The Deficit Gap - the very existence of which Prof Murphy tells us bemuses him - is the difference between the red and black lines. This is how much bigger Scotland's deficit per capita is than the rest of the UK's. A small part of the reason is because we generate less revenue, but the vast majority of the reason is that we simply spend more.

It's not hard to understand is it? Any even half-competent analyst would look at this data and say "we have to understand why Scotland spends more per capita on public services, that's clearly the main reason why the deficit gap exists".

Which brings us to the less than half-competent GERS analyst that is Professor Richard Murphy. Earlier this week he offered the following nugget of an insight:

"what GERS still shows is the improbable likelihood that the (sic) Scotland is disproportionately responsible for the UK deficit"

The "improbable likelihood" -   Prof Murphy's position appears to be that he simply can't get his head round why this could be true, despite the fact that the reason is staring him in the face: we spend much more per capita than the rest of the UK.  This isn't about estimates or allocations: this is known expenditure data and it isn't surprising

Let me reiterate each of these points.

1. It's not about allocations. As we'll come on to see, Murphy appears to have only just noticed the concept of non-identifiable expenditure (that is "expenditure that cannot be identified as benefiting a particular country or region of the UK but is instead incurred on behalf of the UK as a whole"). The point here - and I can't emphasise this enough - is that the vast majority (near as dammit all²) non-identifiable expenditure is allocated to Scotland on a population basis. Scotland's spend per capita on these non-identifiable costs (that are allocated on a per capita basis) is, by definition, exactly the same as the per capita spend for the rest of the UK. So when we're looking at per capita spend differences, this has absolutely nothing to do with population-allocated costs.

2. It's known expenditure data. These aren't estimates or survey based allocations (as is sometimes necessarily the case on the revenue side) - when we're looking at the differences in per capita spend we are dealing with known, actual, undisputed data.

Maybe it needs laying out more clearly. Below is a simple table I've derived from the data provided as support to the latest GERS figures. It compares 2016-17 GERS reported spend per capita in Scotland with spend/capita in the rest of the UK.

[As an aside: you can use this table to find ways of closing the £1,500 per capita spending gap, including making assumptions about what an independent Scotland might replace those population-allocated costs with. So if we spend *nothing* on defence, we'd save £565/capita vs our GERS expenditure; if we cut our Social protection budget (including pensions) by 9%, we'd be back to the UK average and have saved £408/capita.  Have a play - £1,566 per capita is an awful lot of money]


3. It isn't Surprising. Look at the table above: the big spend/capita differences are no surprise to anybody familiar with Scotland's lower population density and the characteristics of Scotland's population. As the Fraser of Allandar Institute recently said

We know that Scotland spends more per head than the UK both because of how much is spent on things like health, education, economic development etc. but also our slightly higher number of people entitled to benefits associated with issues such as long-term ill health etc. There are also some minor technical issues, like the fact that Scottish Water is a public asset in Scotland but not elsewhere.

***

So faced with this frankly rather clear and easy to understand picture, what does Professor Murphy do? Does he start to look at the higher per capita spend areas and understand why we spend more money, whether that is indeed justified by need, whether we could find savings there if we needed to?

Of course not.

Like an accident investigator looking at the Titanic disaster and saying "I'm frankly bemused by why that ship sank - I want to hear more about the way the deck-chairs were arranged, I think that might explain it" he disappears off down a rabbit hole questioning accounting treatments he clearly hasn't understood or thought through - don't look at the iceberg folks!

Here's what Murphy's latest epistle from the planet bonkers actually says:

"I have been continually bemused by the fact that GERS – Government Expenditure and Revenue Scotland – and its equivalent data for Wales and Northern Ireland – says that Scotland runs a deficit so  much larger in proportionate terms than that for the UK as a whole"

[Comment: look at the graph above. If Murphy is still bemused as to why that deficit gap exists, he surely just needs to understand where and why we spend more per capita on public services? That's clearly the main explanatory variable here.] 

"What follows is speculation at present: think of it as an idea put out for peer review right now and not a final argument"

[Translation: I know I can't defend any of this, I just need to feed those wanting to dismiss the GERS figures and this is the best I can muster.]

"Until 2013 Scotland collected more per head than the rest of the UK, Now it collects less: this is an obvious reason why the scale of its deficit appears to be growing". 

[Comment: Yes Richard, the black line approaches the green line - there's no mystery here, those of us paying attention saw this coming]

"Much, but not all of my criticism of GERS has focussed on the fact that almost all the significant revenue figures are estimates based on either data extrapolation of the whole of the UK or on relatively small samples for Scotland meaning that I think that there is doubt about whether all the major tax revenues are fairly stated"

---

This is a side-show to the main-event, but it's worth pausing here. Given Murphy's overall position of being bemused by the deficit gap, I think we can safely say he's implying his doubt about whether the figures are fairly stated suggests he thinks they may be understated. That's certainly how his pro-independence cheer-leaders interpret this proclamation.

In fact - as I and many others have argued - any assumption bias that exists is a/ not material to the debate and b/ likely to favour Scotland (because of obvious political pressure to do so).

So when the Scottish Government Statisticians chose different assumptions to HMRC for Scotland's share of corporation tax, they were assumptions that were more favourable to Scotland. Similarly when it came to how we calculate Scotland's geographic share of oil revenues, Scottish Government statisticians chose a methodology which favoured Scotland. In both cases, following consultation and reflection, the Scottish Government's statisticians have accepted that HMRC assumptions are now more accurate and have revised past figures appropriately³.

Sure enough if we look at the empirical data, the effect of subsequent revisions to past figures in GERS has nearly always been to make the initially reported figures appear to have been optimistic, as the graph below rather neatly demonstrates:

I've chosen to go back to 2007-08 data and 2011-12 GERS reporting as those figures (yellow on the chart above) were the ones relied on by the Independence White Paper and the "last five year" figures quoted therein. The red figures are the latest numbers released this week. So if one were to correct the White Paper's text (p.599) with what we now know to be more accurate figures:

Since 2007/08, Scotland has run an average net fiscal deficit of £8.3 billion £10.0 billion (5.9 per cent 6.8 percent of GDP). [..] In 2011/12, the latest year for which data is available, Scotland is estimated have run a net fiscal deficit equivalent to 5.0 per cent 7.0 percent of GDP. In the same year the UK is estimated to have had a deficit of 7.9 per cent  7.1 percent of GDP. 

Or maybe we could restate it with the latest five year's figures instead:

Since 2007/08 2012/13, Scotland has run an average net fiscal deficit of £8.3 billion £14.0 billion (5.9 per cent 9.0 percent of GDP). [..] In 2011/12 2016-17, the latest year for which data is available, Scotland is estimated have run a net fiscal deficit equivalent to 5.0 per cent 8.3 percent of GDP. In the same year the UK is estimated to have had a deficit of 7.9 per cent  2.4 percent of GDP. 

The big downward revisions in this year's report are mainly the impact of accepting that the past methodology used for reporting Scotland's geographic share of oil income was overly generous to Scotland (see note 3 for details).

As you can clearly see, the net effect of this correction is to adjust down the revenues allocated to Scotland by nearly £7bn over the last 10 years, to adjust the revenues down in each of the last five years. 

So having understood all of that, now read how "Professor of Practice in International Political Economy at City, University of London" Richard Murphy describes these adjustments 

"Some changes, e.g. on oil revenues, have taken place, with modest up-ratings in Scottish revenues as a result"

And we're meant to take this guy's comments seriously?

---

OK, back to the main event and Murphy's blog:

It is however said that the real problem is in spending [...] here things look really awry

You'd think he'd now actually look to see where and why this is the case, wouldn't you? No such luck. What follows is a painfully convoluted attempt to argue that the accounting approach used in GERS is flawed and that we're unfairly allocated some costs and/or not allocated the tax income associated with those costs because of a failure to appropriately "match" in an accounting sense.

This is basically our accident investigator saying "but those deck-chairs: are we sure they weren't rearranged in a such a way that caused uneven weight distribution and thereby contributed to the otherwise inexplicable sinking of the ship?"

I have been through what he's written on the accounting technicalities, I really have. Suffice to say there's nothing of material significance in the points he makes and nothing which hasn't been discussed before (albeit in the more esoteric backwaters of the debate, because the issues don't pass any reasonable materiality threshold in the context of the constitutional debate).

I could elaborate more, I really could - but I don't see why I should have to spend time explaining why the way the deck-chairs were positioned really doesn't matter - after all: it's Saturday, the sun is almost shining, and I want to go and ride my bike.

*** Addendum ***

If you aren't convinced that Murphy is wrong in his latest stumbling foray into the technicalities of GERS, the ever-diligent Fraser Whyte has written a thorough debunking of his "argument" here

*****

---

Notes:

---

1. All the figures I quote here are Scotland versus the rest of the UK (rUK) where rUK = [UK - Scotland]. Some confusion may arise because most commentary you'll read is based on Scotland vs UK (as a whole, including Scotland). The latter is easier to do (because that's how the GERS report shows the data) and perfectly valid if we're considering our choice to either "share with UK" or "go it alone". I think comparing to rUK is more informative for the ongoing debate about fiscal transfers - if you like it's consistent with the SNP's "us" vs "them" approach as opposed to the indyref question which was  "us alone" vs "us with them". To be honest this subtlety doesn't make a jot of a difference to the overall conclusions.

3. Allocation of non-identifiable expenditure is almost all done on a population (per capita) basis. [The biggest numbers are of course those related to defence and debt interest]

3. Some details on revisions to GERS:

Corporation Tax: GERS used to use a methodology that differed from HMRC and favoured Scotland. Following consultation, in the 2014-15 GERS the Scottish Government's statisticians agreed that the HMRC's assumptions were more appropriate and so figures were revised down (compared to those used during the Independence Referendum)

Oil Revenues: As explained within the GERS report itself and the GERS consultation document