Securing the long-term credibility of the OBR

If the verdict on the recent Spending Review was that George Osborne is a lucky chancellor then Robert Chote, the head of the OBR, is seriously unlucky. A surprise modelling revision and embarrassing mistakes in the previous forecasts have left the OBR facing criticism around its credibility. There are four reforms that the OBR could make, both to improve the accuracy of its forecasts and to protect itself against the inevitable flak next time it needs to change its numbers. 1. Presentation of figures The Chancellor made much of the receipts forecast improving by £27bn, which sounds like a huge annual swing. That perception has hurt the OBR. But that is actually a cumulative figure of increasing receipts across the five years of the forecast period. The bold headline statement is misleading about the scale of the revision. There used to be a Treasury rule that tax and spending numbers were only ever quoted on an annual basis. The Treasury caved in to pressure a long time ago, under Gordon Brown, to make big fiscal numbers sound even bigger by accumulating them. The OBR should not go along with this phoney presentation. It needs to make clear that it will uphold standards and only quote figures on an annual basis – and be prepared to call out both the Treasury and the Chancellor when they present numbers in ways that can mislead. 2. Transparent modelling The OBR has been damaged by the extremely convenient (for the Chancellor) discovery of modelling errors, which “found” an additional £ 6.1bn in VAT and National Insurance Contribution (NIC) receipts by 2019/20. These errors have been in the public finance forecasts for a long time. The OBR are right that, once discovered, it would clearly be wrong to subject the country to spending cuts or tax rises that were not justified by an up-to-date model. It is not the revisions themselves, but the perception that they came out of nowhere, that hurts the OBR. To make sure this doesn’t happen again, the OBR should implement a new and much more transparent process for exposing modelling changes. The OBR already does this in part – the VAT change, though not the NICs one, was flagged in a publication in October. This process should be standardised and decoupled from the OBR’s twice-yearly forecasts. The Opposition (and the media) should know the effect of modelling changes before the Chancellor stands up in the Commons. 3. Access to departments The tax credit debacle was not helped by the original cut being “miscosted” by £1bn – a 25% error. This was a Treasury costing, but one that the OBR ‘certified’ – despite not speaking to officials in HM Revenue and Customs (HMRC) and the Department for Work and Pensions (DWP) to verify the numbers. This was because the Treasury does not allow the OBR to speak to the departments directly responsible for tax credits policy. Without access to departments, how can the OBR credibly certify the figures? The root cause of this problem lies not with the OBR, but with the excessive secrecy the Treasury imposes on the budget process. The transparency provided by the OBR process has (eventually) exposed one error. Given the weakness of the budget process, you can be sure there have been many more over the years that the Treasury has not owned up to. The OBR needs to make clear that it will not certify costings without access to the relevant experts. This will helpfully expose when the Treasury is producing numbers without itself knowing they are right. 4. Forecasts not political assumptions As we have pointed out before, the OBR treats different kinds of spending in different ways: annual managed expenditure (AME) spending (on welfare and other benefits) is treated like tax receipts, by projecting forward the costs of current policies. Departmental expenditure limit (DEL) spending is treated fundamentally differently. Instead of forecasting future spending based on the cost of current policies, the OBR effectively uses the Chancellor’s assumption for what he’d like that spending to be – regardless of whether the Chancellor’s numbers are compatible with the Government’s current detailed spending policies. Unsurprisingly, this has not proved to be a reliable way of forecasting public finances. The Spending Review has now replaced the words of the Chancellor before the election with detailed spending policies, revealing a huge variation between the two. This episode has raised serious questions about whether the OBR is helping or hindering the debate on fiscal policy. The OBR needs to start forecasting DEL spending for periods not covered by a detailed Spending Review. This will not be easy, but many of the practical objections are overstated. And any model, even a simple one, is better than relying on the words of politicians. Conclusion The OBR has become an important part of the fiscal landscape. The personal authority and credibility of Robert Chote and his team have prevented serious accusations that the OBR deliberately made it easier for the Chancellor. It must now use this credibility to assert its independence from the Treasury, including taking a stronger line against misleading Treasury practices and acts of political convenience. If it fails to do so, the fiscal watchdog will start to look more like the Treasury’s poodle.