Iain Stewart is the Member of Parliament for Milton Keynes South. Formerly a head-hunter, Iain was head of research for the Scottish Conservatives in the 1990s and co-authored “It’s Our Money, Who Spends It?” about the Barnett Formula and fiscal autonomy for Scotland.
And so the Barnett Formula once again steps into the political spotlight; this time as one of Conservative Home’s prescriptions for securing a Conservative majority in 2015. Reforming the formula is regularly cited as a mechanism through which, proponents argue, tensions and unfairness in the Union may be resolved.
Yet the formula endures into its fourth decade and successive Governments have parked reform in the “too difficult” box. The 2001 comment by Lord Foulkes, then a junior Scotland Office minister, is telling,
“If the SNP think that Barnett is too mean and the English Tories think that it is too generous, most sensible people would think that it is just about right”.
I do not write in defence of Barnett. Indeed, if we are entering a debate on what form “devo-max” might take, it is probably timely to consider whether Barnett remains fit for purpose. The formula is, however, a much misunderstood mechanism which is, in reality, only one component of a complex fiscal arrangement between Scotland (and Wales and North Ireland) and the UK Government.
Drawing on a book a co-authored in 2003 (It’s Our Money! Who Spends It? – copies still available), in this article I hope to shed some light on this oft-maligned formula, point to where the debate ought to focus and gently to suggest a path by which discussion might rationally proceed as we consider broader issues about the Union.
Critics of Barnett often start by highlighting the per capita levels of public expenditure in each part of the UK. The latest Treasury figures (2009-10) show that per capita spending in Scotland was about £1,400 higher than in England (£9940 v £8531). The Barnett Formula is cited as the perpetrator of this “largesse” shown to Scotland and calls for abolition/reform swiftly follow.
It is not, however, that simple. For a start, the Formula only applies to certain areas of public spending; those that are devolved to Holyrood. The Formula does not apply to large sections of spending; the main one being welfare. Almost 40% of identifiable public spending is not covered by Barnett.
Nor does Barnett determine the overall size of spending in Scotland. It is a formula for determining the annual changes to specific spending areas, based on Scotland’s population relative to that of England. In general terms, for any public service which is devolved to Scotland, Scotland will get about 8.5% of any increase in spending of that service in England. For example if the NHS in England gets an extra £1bn, Scotland would receive approximately £85m.
Each year, the changes for each spending programme are totalled up and an adjustment is made to the “Scottish block”; the annual grant from the Treasury to the Scottish Government. It is then up to the Scottish Government and Parliament to slice up the public spending cake as it wishes. It is not under any obligation to hypothecate each change. For example, Scotland could opt to spend the extra “NHS money” on education if it wanted to.
So Barnett does not determine either the overall size of public spending, or the level within each spending area. Why, then, is spending in Scotland apparently higher?
For part of the explanation we have to go back to the 1970s. The last needs-based assessment of public spending in Scotland was made by the Callaghan government in advance of the devolution legislation. Based on data for 1976/7, the Treasury led an inter-departmental study to assess the extent to which per capita spending in Scotland would need to be higher than in England to provide a comparable level of service. The conclusion was that it would need to be 16% higher (taking into account factors such as Scotland proportionately greater land mass, rural population, council housing stock and poorer health indicators) in order to provide an equal quality and breadth of provision of public services. However, it was found that actual spending in that year was 22% higher.
Partly to address this, and partly to avoid the need for annual “table thumping” by Secretaries of State for Scotland to determine the annual Scottish grant from the Treasury, the Barnett Formula was introduced. Ironically in light of the current controversy, Barnett was conceived as a convergence formula to reduce over a number of years the difference between Scottish and English spending levels. I do not wish to tax readers’ arithmetical skills but it is a fact that, all other factors remaining constant, if you apply year on year a pro-rata population based increase to a total that is significantly higher in relative terms, then in time the difference between the two will narrow.
Although there is no concrete data to prove this point absolutely (given my earlier note of caution about the territorial public spending per head figures), it would seem that Barnett has not delivered the convergence it was designed to do. Why?
There are two principal reasons for this. First, in the early years of the formula the population percentage figure was not updated. As Scotland had a proportionally declining population compared to that in England over these years, it meant that the Formula was delivering a bonus to Scotland. That anomaly was addressed in the 1990s by a change to an annual variable percentage, but for a decade or so a bias was built in.
Secondly, and more significantly, under Governments of all colours for reasons of political expediency separate funding deals outwith the scope of Barnett continued to be made between the Treasury and the Scotland Office which boosted Scotland’s funds. The former Scottish Secretary, Ian Lang, neatly set out the issue in his autobiography:
“The real scope for protecting Scotland’s interests lay in side deals and the special ad hoc negotiations that stood outside the corral of the ‘block and formula’. I calculated after two years as Secretary of State that the Barnett Formula had reduced the Scottish Office budget by £17m, whilst separate deals with the Treasury had increased it by £340m. The very existence of the Barnett Formula, far from inhibiting me, enabled me to concentrate on special deals to augment our resources”.
And it continues today. For example, the Autumn Statement contained a spending announcement for Scotland (to match fund Scottish Government spending regarding the Caledonian Sleeper services) outwith the confines of Barnett.
Therefore, if Barnett ended tomorrow the issue of comparative spending levels in Scotland and England would go on. There has not been a needs-based review of public spending in the UK since the 1970s, so we do not know for sure what different levels of spending there should be now to achieve a parity of service provision. This poses a number of questions.
Is it actually meaningful to compare Scotland and England as a whole given that within each nation, and within each region of each nation, there are large differences of need and current spending? For example health spending in Milton Keynes is effectively top-sliced to give extra resources to the North of England, yet arguably demographic changes in my constituency warrant that money being retained. Should we not be reassessing “needs” on a much more localised level throughout the UK?
Or should we question the need for a mechanism to deliver parity of service delivery now that there is devolution? “Devo-max” would mean that the Scottish Parliament and Government would be responsible for determining the total of its spending as well as the allocation to each spending programme.
That takes us to the proportion of tax revenue that comes from Scotland. Most of the debate centres on the spending side of the ledger; yet the quantity of tax receipts contributed by Scotland to the Exchequer is no less controversial.
My book examines this issue in some depth. In short, however, nobody is able to quantify the amount exactly. There have been, and are, a large number of studies which aim to calculate the amount but they all rest on a series of assumptions which are often hotly contested. Over the years, such studies have produced a wide spectrum of conclusions on Scotland’s fiscal position; from a healthy surplus to a substantial deficit. Hence the “we’re being robbed” cries from both sides of the border.
The kernel of the problem is that we have never assigned tax receipts on a territorial basis in this country. For example, how would income from North Sea oil and gas activities be apportioned between Scotland and England? There are many views. Similarly, how should we allocate corporation tax receipts from companies which are headquartered in England but generate a significant portion of their profits from activities in Scotland (and vice versa)?
Although difficult, I do not say that it is impossible to produce such accurate figures. But we do not have them now and that is not conducive to rational debate.
We may also have to consider the disaggregation of pension and social security receipts and payments. Scores of actuaries would have to be employed to apportion National Insurance receipts and liabilities between Scotland and England. In addition, there are territorial tax consequences arising from personal pension contributions. The UK Exchequer foregoes income tax revenue on personal pension contributions from people in Scotland, and has done for many years, in the expectation that it would later receive income tax on annuity payments. If a separate Scottish Exchequer were to receive this deferred income, how would that impact on the fiscal relationship? For these and other examples, again, it is not impossible to work out but we do not know at present.
My plea, therefore, is that before we decide on the future of the Barnett Formula, Devo Max and the like, we produce detailed and incontestable territorial public accounts in this country. It is only on that basis that debate on constitutional change can meaningfully proceed. Until then, we shall continue to have much heat, but little light, on the subject.