Duty of fiscal candour: why public spending figures should be presented much more clearly

One of the many regrettable consequences of the prorogation of parliament was that Sajid Javid was able to announce big increases in public spending for the financial year starting next April while dodging the detailed scrutiny that should then follow. In ordinary circumstances the Treasury select committee of MPs would quiz the chancellor of the exchequer thoroughly soon after his statement to the House of Commons. With so little time before parliament was suspended Javid was able to duck out of an appearance.

The chancellor’s avoidance tactics were in keeping with his statement to the House on 4^th September where he was twice reprimanded by the Speaker John Bercow for talking politics rather than public finance. When Javid did get to the point, his main headline announcement was a 4.1 per cent real increase in day-to-day spending on public services, amounting to an extra £13.8bn between 2019-20 and 2020-21. However, the chancellor failed to mention that such expenditure was already set to rise by 1.3 per cent after inflation, because of the boost to the NHS under way through a five-year plan announced by Theresa May in June 2018. The specific impact of Javid’s measures was thus an additional 2.8 per cent, a figure we did not hear.

Public spending statistics are inherently complex, which makes it all the more important that they are presented as transparently as possible, spelling out the way that they have been put together. Instead, too often, they are confusing and obscure. Though Javid presented his figures in self-serving fashion, there is nothing new in chancellors doing that. A more reliable source is the Treasury’s accompanying report that sets out the full facts and figures behind the speech to MPs. However, on this occasion it was particularly tricky to follow.

Search for that headline number of £13.8bn and you find it tucked away in an annex where a table sets out departmental budgets for day-to-day spending this year and next. Only there will you find that Javid was referring to real increases in 2020-21 prices. The effect of using next year’s higher prices is to exaggerate the increase (a bit), but what matters is that anyone listening to the chancellor would reasonably have expected him to present it in today’s prices. This was for example the way in which May set out that five-year plan for the NHS in England last year, when she promised an increase of £20.5bn by 2023-24, in 2018-19 prices.

That is not the only issue raised by the Treasury’s table. Indeed, at first sight it is baffling since it shows overall day-to-day spending rising by £15.4bn in real terms (from £336.9bn in 2019-20 to £352.3bn in 2020-21) rather than the declared increase of £13.8bn. You have to trawl through 15 footnotes to establish that the £13.8bn is calculated by restating the spending for this year to allow for classification changes that have temporarily lowered departmental expenditure in 2019-20.

It would have been far better if the Treasury had set that out explicitly in its table rather than burying it in the small print. And this was not the only resort to discreet footnotes in the report. Unlike Javid or any other Treasury minister, Robert Chote, head of the Office for Budget Responsibility (OBR), a watchdog, did manage to find the time to face questions from the Treasury Committee. Speaking to MPs on 9^th September he pointed out that the government’s claim that the new spending plans complied with its fiscal rules rested on a footnote referring to the OBR’s March forecast. The budgetary outlook has darkened since that now elderly assessment.

The distortion that the Treasury was ironing out to reach its £13.8bn figure is not unusual. It arises when an item of expenditure is relabelled between the two main categories into which total spending is split. The first is the departmental budgets whose plans for 2020-21 Javid announced this month. Making up nearly half of total spending, these “DELs” (departmental expenditure limits) typically, though not on this occasion, have their plans set three or four years ahead. The rest of spending, AME (annually managed expenditure), is planned on a yearly basis. Switches between these two categories can distort the picture over time within each of them while changing nothing overall.

However useful this overall approach may be as a method of budgetary control for the Treasury, it creates havoc with the historical figures. The Institute for Fiscal Studies published a study last year of the Treasury’s effectiveness in controlling public spending between 1993 and 2015, a period during which the DEL/AME regime has predominantly been in force. “A really important finding” of the analysis, noted the think tank, was "the remarkable difficulty we have experienced in collecting and interpreting data on plans and outturns which are consistent over time." A particular difficulty in more recent years has been the switch of spending in Scotland out of DEL into AME as a result of the new Scottish fiscal settlement.

Adding to the complexity of the spending figures is the use of commercial accounting practices by Whitehall since the early 2000s. Since then there has been a general shift across developed economies in the same direction, according to Delphine Moretti, a public budgeting specialist at the Organisation for Economic Co-operation and Development. But Britain is one of only three among the 36 countries that belong to the OECD (the other two are Australia and New Zealand) which have gone the whole hog, by applying the private-accounting standards called IFRS in the public sector.

The case for the accounting reform is to reinforce financial discipline within government, for example by making provisions to cover the likely cost of future liabilities that have been incurred. The downside is that it has made Britain’s public expenditure figures even trickier to follow. In particular it has driven a wedge between those reported and budgeted by departments and the crucial headline spending figures in the national accounts, which by contrast recognise provisions only when they are settled. Removing that wedge requires accounting adjustments, which are shoved into AME. These can be huge: in 2018-19 they stripped out for example over £80bn of public-sector pension costs recorded under the commercial accounting standards but not in the national accounts.

Given these brow-furrowing complexities, it is vital that ministers and departments present spending figures in a much clearer fashion. Exhortation alone will achieve little. The OBR has proved its worth over almost a decade in providing independent economic and fiscal forecasts. Its remit should be expanded to provide a transparent and comprehensible presentation of public spending figures that matter so much but are understood too little.