Peter Jones: Devolving tax could cost billions

The Barnett formula ensured public spending was higher in Scotland than the rest of the UK. Picture: PA

But what none of them seem to be addressing is that this could have severe consequences on Scottish public spending.

There is a dog that isn’t barking in this debate – the dear old Barnett formula. Readers will recall that Barnett determines not the block grant that the Treasury annually gives to the Scottish government (and to the Welsh and Northern Irish administrations) to spend, but the changes in it.

The fact that it just determines changes, and not the total sum, is crucial to understanding how, historically, Scots politicians have had more money to spend per Scot than English and Welsh have to spend on their people. But this spending advantage will come under intense scrutiny if proposals to transfer more tax powers start being discussed at Westminster.

Why? Let’s remember that Barnett was introduced as a temporary measure in 1979 when the then Labour government was trying to introduce a Scottish assembly that would have no tax powers. As we know, that effort failed, but Barnett remained.

What it did was take the existing block grants to Scotland, Wales, and Northern Ireland and add the refinement that for every extra £1 per head that Westminster decided to spend in England, £1 extra would also be spent in these three parts of the UK. If spending was cut in England, it would be cut in the  same per-capita way in the rest of the UK.

This meant that any historical bias or difference in any part of the UK in total amount of spending per capita was built into the system. And it so happened that the bias favoured Scotland. In 1979, per-capita spending was slightly more than 20 per cent higher in Scotland than average per-capita spending across the whole UK.

Barnett should have eventually reduced per-capita spending in each part of the UK to the same level. This is just an arithmetic effect – if you add, say, £5 to £100, that’s a 5 per cent increase, but if you add £5 to £120, it is only a four per cent increase.

But this “Barnett squeeze” didn’t happen to anything like the extent it should have. The reason is hotly debated in academia. Mainly, it is because successive Scottish secretaries and first ministers found ingenious ways to persuade the Treasury that extra lump sums over and above Barnett should be added to the Scottish spending total.

Oddly enough, when public spending is being reduced, the arithmetic effect is to increase Scotland’s relative advantage. The result is that, although there has been a bit of a squeeze, Scottish per-capita spending is still 13 per cent higher than the UK average, and 16 per cent higher than English per-capita spending.

Scottish politicians say this difference is justified because there is greater deprivation in Scotland requiring more money to be spent on poverty alleviation, the population is more sparsely spread, which makes it more expensive to provide public services, there is greater ill-health necessitating more health spending, and so on.

Boiled down, the argument is that Scotland should have higher spending because Scotland needs it. Barnett, however, does not determine public spending levels on the basis of need, but on the basis of population. It also carries the implicit assumption that the Scottish population has the same characteristics, on average, as the UK population.

This is almost certainly not true. It is also the case that there is a growing belief in political circles that public spending in the various parts of the UK really ought to be based on needs, ie varying population characteristics, rather than on population uniformity.

This view is strongly held in Wales, where there is a widespread view that the Welsh get a raw deal from Barnett. The Welsh have also tested what a needs-based formula for determining public spending might look like, and what results it might produce.

The work was carried out by the Independent Commission on Funding and Finance for Wales, chaired by Gerald Holtham, a fund manager and adviser on infrastructure spending to Wales’ first minister.

Its 2010 report devised seven indicators to assess needs across the UK. Three were demographic (number of older people, children, and people of ethnic origin), two measured deprivation (income poverty and ill-health), and two assessed costs of providing public services (sparsity of population  and a London-weighting allowance).

The needs assessment found that for every £100 spent in England, £115 should be spent in Wales, £121 in Northern Ireland, and £105 in Scotland. This Scottish figure is £11 per head less than is currently spent. If applied to the present Scottish block grant, it would mean a 9.5 per cent, or about £2.9 billion, cut to the Scottish budget.

This is big beer, dwarfing the current cuts being complained of by the Scottish government. The Holtham commission conceded that its assessment of Scottish needs was contestable as the functions devolved to Wales were more limited than those devolved to Scotland, but maintained that its figures were generally in the right ballpark.

You could argue that it is hardly surprising that a Welsh commission would come up with findings that are highly favourable to Wales. You could also dispute both the needs factors that it used and the weighting it gave to each factor in the needs assessment formula.

Nonetheless, the Holtham formulation is the only major needs assessment exercise on UK territorial spending to have been carried out since a similar Treasury study in 1978 which also found that Scotland was getting more spending than was warranted. And because of that, it would be the starting point in any political battle over devolved spending.

Re-opening debate about devolving more taxes would, because there would still be a block grant, also open debate about that. So, while there are good reasons for devolving more taxes, there are also 2.9 billion reasons why it might be a bad idea, not the least of which is that it would be dubbed the Union squeeze by the SNP.