BGI’s latest report – “Financing Scotland: Is there a workable financial settlement for Scottish devolution?” – points out that the airy assurances given in the heat of the referendum debate that financial devolution could go ahead while retaining the Barnett formula may not prove so easy to deliver in practice.
The formula was introduced in 1978 as a simple population-based system for allocating Scotland a share of changes in UK public expenditure. It doesn’t say what is the right level of UK resources going to Scotland (and Wales and Northern Ireland). It just says what the change should be. If £100 is added by the Treasury to spending in England on an activity that has been devolved to Scotland, then Scotland gets X% of that added to its block grant – where X is the population of Scotland relative to England, currently around 10%. So the increase in spending per head in Scotland in cash terms is roughly the same as England – but the base level of the block grant is just what spending happened to be when the formula was introduced, plus years of subsequent adjustments.
The settlement recommended by the Smith Commission and welcomed by the PM will mean that more services (employment and training provision for example) will be devolved and so will fall within the scope of Barnett. Some welfare spending will be devolved for the first time. Scotland will be given the revenue from income tax in Scotland and a share of VAT revenues, and powers to change income tax rates and thresholds.
Where does all this leave Barnett? Suddenly an obscure but very simple formula has been caught up in a welter of complicated adjustments and indexations. Can it take the strain?
The first source of tension will be the devolution of some welfare spending. This is currently outside Barnett because the numbers on benefit may move differently in Scotland to England for all sorts of economic, social and demographic reasons. Smith proposes limiting devolution to only a small slice of welfare spending which is not much influenced by economic conditions: benefits for carers and disabled people. The proposal seems to be that an amount equal to current spending on these benefits in Scotland is added to the block grant outside Barnett and then ‘indexed’ in future. It is not clear what it would be indexed to or whether there will be any provision for future adjustment
Second, the devolution of tax powers and revenues will introduce considerable complexity. The Smith plan is that an estimate should be made of Scottish income tax receipts and receipts from the first 10% of VAT. This should be paid by the Treasury to Scotland and the block grant correspondingly reduced. And ‘future growth in the reduction to the block grant should be indexed appropriately’.
This seems to mean that total Scottish revenues from the UK, which now consist of roughly £28 billion of block grant, would in future be made up of perhaps £11 billion of income tax and £5 billion of VAT, leaving only £12 billion or so of block grant driven by Barnett. It is not at all clear why the reduction in the block grant would be indexed, as proposed by Smith, or to what.
Even more complex is the handling of UK changes to income tax. Suppose the UK increases the basic rate while Scotland leaves its rate unchanged and that the proceeds of the UK increase go to fund a mix of defence (non-devolved) spending across the UK and health (devolved) spending in England. Scotland would appear to gain the benefit of stronger UK defence without paying anything towards it. And if Barnett applied to the increase in health spending the Scottish block grant would benefit from that too! Smith says ‘changes to taxes in the rest of the UK, for which responsibility in Scotland has been devolved, should only affect public spending in the rest of the UK.’ But no explanation is given and no mechanism is outlined to achieve this. And in practice of course receipts from specific taxes are almost never linked to changes in specific spending programmes.
Finally, change will inevitably focus attention on the Scottish ‘advantage’. On a simple measure of public spending per head that is widely quoted Scotland gets 19% more than England. But stripping welfare spending out of this calculation increases the advantage to 27%.
In conclusion, the Barnett formula has proved pretty robust over past decades but perhaps the key to its survival has been that it is so simple that it has been able to tick over in the background without too much challenge. But it is hard to see Barnett surviving when it comes under the spotlight. The new constitutional settlement, and the financial settlement which it rests on, must surely deserve wider debate.