The Role of the OBR

Twenty-eight panellists responded to this question. 43% of respondents thought that the OBR should maintain two comprehensive forecasts per year, while 11% favoured moving to a single annual forecast. One panellist thought that the OBR should evaluate Treasury forecasts rather than produce its own, and another panellist proposed that the OBR should move to an advisory-only role – issuing public commentary on fiscal risks without publishing formal forecasts. 39% of respondents thought that the OBR should move to probabilistic estimates of fiscal headroom rather than point forecasts.

Question 3: What do you see as the main reforms needed to preserve the OBR’s credibility and independence?

Nineteen panellists responded to this question, which was open-ended. A number of panellists proposed changes to how the OBR communicates uncertainty. Roger Farmer (University of Warwick) said that it “makes sense to recognise the potential uncertainty around assessments of fiscal sustainability based on alternative policy scenarios”. Michael Wickens (University of York) also supported the use of scenarios, noting that given the uncertainty of future policy, “the best the OBR can do is provide likely scenario outcomes”. He added that “probabilistic assessments of these would give too wide bands to be useful”. In contrast, Ethan Ilzetzki (LSE) thought that probabilistic estimates would allow for “more de-facto flexibility”, given the policymaking culture of “knee-jerk” responses to forecast changes.

Some panellists emphasised using scenarios for evaluating policy proposals rather than for communicating uncertainty. Thomas Sampson (LSE) argued that rather than having the OBR be the “arbiter of whether the fiscal rule is met”, a better alternative would be “for the OBR to advise on the possible fiscal impacts of different policy options”. David Cobham (Heriot-Watt University) argued that the OBR should be both able and obliged to consider alternative policy scenarios to assess “current political debates”, including the evaluation of policies proposed by “non-ruling as well as ruling parties”.

Another theme raised by the panel concerned resources and expertise. Ricardo Reis (LSE) thought that the OBR “needs to invest in its capacity to make forecasts and issue evaluations”. Vincent Sterk (UCL) considered the OBR to be understaffed, given its importance, and that “hiring more staff closer to the academic frontier” and “engaging more directly with the academic community” would help to improve forecast quality and bolster credibility. In a similar vein, Robert Kollmann (Université Libre de Bruxelles) suggested appointing senior academic economists to the OBR’s leadership.

Some panellists expressed differing views on the appropriate frequency of fiscal forecasts. Costas Milas (University of Liverpool) thought that the OBR should only provide a single yearly forecast of the fiscal rules, with a second yearly forecast that gives a more general outline of the economic outlook. Roger Farmer (University of Warwick) agreed, stating that “it makes sense to produce an assessment of [the fiscal mandate] on an annual basis”. In contrast, James Smith (Resolution Foundation) proposed that twice yearly forecasts should continue and that both forecasts should assess the fiscal rules. Instead, he noted that “a more transparent, regular process for re-evaluating supply side assumptions would be a process improvement” for the OBR.

Panellists also suggested drawing on insights from external forecasters. Jorge de Macedo (London Business School) recommended comparing fiscal policy assessments with other forecasters, similar to the setup between the IMF and OECD. David Aikman (NIESR) proposed “anchoring key assumptions such as trend growth or potential output to the median of external forecasters’ estimates” as a way of making such judgements more rules-based.