Robert Chote, its first full time Chairman, explains how the UK’s Budget watchdog came about, and examines some of the critiques it has faced
When applications were invited to chair the Office for Budget Responsibility in the summer of 2010, I consulted two distinguished former Treasury officials. One urged me to go for it, on the grounds that creating the OBR was a worthwhile attempt to improve fiscal policymaking in the UK and that it would be an interesting personal challenge. The other told me not to touch it with a bargepole, as it was probably a cynical ruse by the then Chancellor to dodge responsibility for a misguided fiscal strategy, and it would do my reputation no good at all.
In the end I decided to apply, served ten years and hugely enjoyed it. But it is for others to judge who gave the better advice. While the OBR is now firmly established in the UK’s economic policy architecture and generally well-regarded, it has not been without critics and opponents. Some, inevitably, take issue with specific forecast judgements. Others have deeper objections, arguing that it thwarts legitimate ministerial policy ambitions and reinforces a growth-sapping Treasury orthodoxy.
The OBR is a ‘non-departmental public body’ with a decision-making committee of three appointed by the Chancellor (subject to Treasury Select Committee veto) and 50 civil servants. Created by the Coalition Government as an independent-but-official watchdog, it was given four tasks: to publish five-year forecasts for the economy and public finances; to police the government’s targets for borrowing and debt; to scrutinise the cost/yield of policy measures; and to assess long-term sustainability. Later it was asked to do more on devolution, welfare spending and fiscal risks.
There are also some things Parliament told the OBR not to do: examine alternative policies (including those put forward by opposition parties), make policy recommendations and comment on the merits of individual measures.
Why create an OBR?
The OBR is one of 50-plus independent fiscal institutions (IFIs) around the world, providing non-partisan analysis of governments’ finances. IFIs are a response to ‘deficit bias’ and ‘procyclicality’ in fiscal policy: governments borrow more on average than they should, and too often they amplify the ups and downs of the economy by spending the proceeds of temporary booms – either cynically or because they fool themselves that they have improved its underlying performance. IFIs help by increasing transparency, discouraging short-termism and pointing out potential over-optimism.
The rise in public sector debt from the 1970s, and the apparent success of independent central banks in depoliticising interest rate setting, stirred interest in creating IFIs. Perhaps, economists thought, some of that stardust could be sprinkled over fiscal policy. Most countries tried to improve fiscal management initially by adopting explicit fiscal targets, but the OBR was one of a wave of IFIs created following the 2007-09 financial crisis when such targets proved insufficient.
The OBR can also be seen in the context of long-running debates in the UK about the transparency and independence of official economic and fiscal forecasts.
Transparency
Chancellors have long recognised that rigorous assessment of the economic and fiscal outlook should inform Budget decisions, but many have been ambivalent about publishing forecasts. The Labour Government started publishing them in 1947, hailed by the Daily Mail as “an annual orgy of pompous prediction”. Rab Butler stopped in 1951, fearing they smacked of socialist planning and could unsettle markets. Selwyn Lloyd re-started and Reginald Maudling re-stopped.
Roy Jenkins released GDP forecasts again in 1968 and they have been published (and widened) ever since. An amendment to the 1975 Industry Act required the Treasury to publish forecasts for the economy, although oddly not the public finances. These typically looked ahead two years, but from 1980 they were accompanied by five-year public finance projections to underpin the Conservatives’ Medium-Term Financial Strategy. Official forecasts have kept this medium-term focus ever since.
Why the ambivalence about publication? One obvious answer is that forecasts are almost always wrong, and politicians find this embarrassing. In 1969 Sir Alec Cairncross, head of the Government Economic Service, explained “some of the inconveniences from the ministerial point of view of putting official forecasts in front of an unsophisticated public”. He argued: “Error is not the main hostage that a Chancellor thinks he is giving to his political opponents. It is rather that an official forecast becomes in the act of publication a plan… They imply that the Government is content that events should be in accord with the forecast. This exposes [it] to immediate criticism not only for what it has done, but for what it has not.”
Independence
What of independence? Pre-OBR the formal position was that Treasury forecasts were those of the Chancellor himself, although they varied a lot in enthusiasm for the detail. Kenneth Clarke wrote: “I never took a close interest in the contents of the [Red Book], which was prepared by my officials…. They said that in the end the figures ‘were your figures, Chancellor’ and implied that it was not unusual for ministers to substitute their own… I regarded this suggestion as outrageous”. His successor was more hands-on.
The first tentative step towards forecast independence came in 1992 with Norman Lamont’s Panel of Independent Forecasters, a methodologically diverse ensemble who produced a range of forecasts so wide that the Treasury’s could not help but look reasonable within it. Gordon Brown axed the panel, asking the National Audit Office to audit selected forecast assumptions. The NAO did not want the job. It lacked expertise and rightly feared that ministers would overstate the imprimatur it was providing.
The first serious proposal for a UK IFI came from academic economist Simon Wren-Lewis in 1996. In office a year later, Mr Brown boosted fiscal transparency in several ways but did not create an IFI – although he later made the case for one by publishing consistently overoptimistic fiscal forecasts and manipulating the ‘golden rule’ for borrowing. At the 2005 election, the Conservatives and Liberal Democrats both proposed an IFI within the NAO, as now exists in France, Finland and Lithuania. The Conservatives first proposed a stand-alone OBR in 2008, initially suggesting that it should give policy advice but later changing their minds. Paradoxically, this strengthened the OBR’s position by avoiding the suspicion that its forecasts would be tailored to support policies that it rather than the government wanted. Labour opposed the creation of the OBR right up to the 2010 election, but then supported it immediately thereafter provided there were sufficient guarantees to ensure its independence.
The OBR in practice
Inserting the OBR into Budget preparations went remarkably smoothly, and it helped that OBR staff were transplanted from the Treasury’s economic and fiscal forecasting teams rather than duplicated. There was no attempt to re-invent the forecasting wheel and the Treasury resisted the temptation to cut costs.
Forecast preparation and Budget-making is now a multi-week iterative process in which the OBR, the Treasury and other departments exchange information privately as the details of the forecast and policy package are firmed up in parallel. As soon as the Chancellor’s speech concludes, the OBR publishes its forecast, explains the impact of policy decisions and pronounces on compliance with the fiscal rules. This requires a clear timetable, set out well in advance and adhered to. The forensic scrutiny the OBR applies to each revenue and spending line, and the costing of specific policy measures, is a huge (but largely invisible and under-appreciated) benefit of this process.
This all worked well during the Coalition years, partly because the need for both parties to sign policies off imposed extra discipline. Gone were the days when measures could be inserted or dropped while the Red Book was at the printers in Vauxhall. But after 2015 the forecast and policy process became messier, with less notice of fiscal events, compressed timetables and more deadlines missed. This caused mistakes and inconsistencies between the economic forecast, fiscal forecast and policy package.
Several factors contributed: the return to single party rule, less Treasury experience at Number 10, testier PM/Chancellor relationships, the weakening of the Treasury’s authority by the Brexit vote, the disruption of the Parliamentary timetable by the Brexit negotiations and – most dramatically – the pandemic. The formal rules and informal practices that underpin the OBR’s work were designed for tax and spending decisions being made twice a year, not 14 times as between spring 2020 and 2021.
The shortcuts necessary during Covid made it harder to persuade ministers to re-embrace proper process thereafter. Most famously, when Liz Truss became Prime Minister in 2022, her Chancellor Kwasi Kwarteng introduced a mini-Budget with huge unfunded tax cuts without requesting (indeed rejecting) an accompanying OBR forecast and scrutiny of the measures. Financial markets revolted and the rest is history. Not surprisingly, his successor Jeremy Hunt was more solicitous to the OBR and Hunt’s successor Rachel Reeves has legislated to bolster its role.
Turning to the OBR’s forecasting and its implications for policy, from the outset we wanted to ensure that as well as publishing central forecasts, we would always show our working transparently and emphasise the uncertainty around every forecast. As it happens, the OBR’s forecast errors for GDP growth and the budget deficit were smaller during the pre-Covid/Ukraine decade than previously under the Treasury, although in-year fiscal forecasts were, surprisingly, harder to improve.
The OBR’s GDP forecasts have tended to lie near the average of unofficial ones and those of the Bank of England. Not by design, but this is no bad thing. If the OBR were ever to present a fiscal forecast based on a far-from-consensus macro view, the analysis of the public finances where it really adds value would simply be ignored.
Some critics argue that the OBR underestimated the impact of ‘austerity’ on economic growth after 2010, although current data show a stronger recovery than initially recorded. Once the government has taken its fiscal policy decisions, the OBR generally assumes that the Bank will set monetary policy to deliver an overall level of spending consistent with the inflation target. This means that fiscal belt-tightening only reduces growth temporarily until the Bank can offset it – assuming that it is willing and able to do so. We published forecasts in the early 2010s showing an extended period of spare capacity in the economy, from which readers might have concluded that we felt the Bank needed more help from fiscal policy to bolster demand. That signal would have been clearer had we also forecast that inflation would persistently undershoot the target, but this would have had the Bank and Treasury straight on the phone.
Unelected and overmighty?
What of the argument that, as a bunch of unelected technocrats, the OBR inappropriately constrains ministerial policy ambitions and indulges its own institutional biases, for example, by favouring immigration? The purist response is that the OBR neither takes policy decisions nor makes recommendations, even when its forecasts show the government missing its targets. Chancellors are free to disagree with its judgements, remain off-track and explain their decisions accordingly. But, as we have seen, it is a brave or foolhardy Chancellor who simply tries to sideline it.
One complaint sometimes heard from Right and Left is that the OBR discourages radical policies to improve long-term growth performance – be they large tax cuts or public investment programmes – because its forecasts understate the benefits. Theory and experience suggest that the OBR is probably right to be sceptical, but the issue is also one of timescale. Even on an optimistic view, policies of this sort are likely to involve large upfront costs while most of the beneficial impact on economic potential and tax revenues will lie beyond the five-year forecast horizon the OBR was given. The OBR’s longer-term projections are more stylised and get much less attention.
Partly reflecting worries that good policy deeds go unrewarded, the OBR is more transparent than in my day about the impact that government policy has on its forecasts for economic potential via labour supply, investment and productivity. But I fear this will have encouraged one-sided and time-consuming lobbying by the Treasury, pleading for even the tiniest upward revisions to the forecasts for potential GDP (which are anyway hugely uncertain) to secure every possible pound of budgetary room for manoeuvre.
One key determinant of the future size of the economy is the size of the population, which in part reflects net migration flows. The OBR has been accused of overstating the fiscal benefits of immigration because its forecasts include the boost to tax revenues without reflecting the need to spend more on public services. This is to misunderstand how the forecast works. Other than in the very short term, the amount to be spent on public services is a policy choice that the Chancellor makes at each fiscal event and that the OBR plugs into its forecast with only minor adjustments. If the OBR hands the Treasury a pre-measures forecast with a bigger population, it is for the Chancellor not the OBR to decide whether to increase its cash spending plans to reflect this.
More generally, it is not for the OBR to predict or recommend what the government should spend on public services or whether it should seek to increase or reduce their quantity or quality – these are core political choices. The OBR’s role is to make the implications of these decisions for the public finances transparent.
Worth the candle?
Looking back over what is now almost 15 years, has the OBR experiment been worthwhile? No-one would argue that it has transformed the health of the public finances for the better. Chancellors over this period have embraced and abandoned more fiscal rules than Henry VIII did wives. Public debt is much higher as a share of GDP, but that reflects major shocks to the economy – some unavoidable and some self-inflicted – more than the sorts of temptations that IFIs were designed to help ministers resist, even though these have also played their part. I suspect that Chancellors have on average behaved less recklessly than they would have done without the OBR and it has also encouraged them to address some long-term issues – population ageing, for example – rather than simply ignoring them or hoping for the best.
The OBR’s economic and fiscal forecast errors were smaller on average during the pre-Covid/Ukraine decade than previously under the Treasury, then unsurprisingly larger. But whether the Treasury would have done better in parallel is impossible to know. And benchmarking against external forecasts is misleading because they typically include expectations of how policy might change which OBR forecasts explicitly do not.
But I would argue (with obvious bias) that the OBR’s biggest contribution – which alone justifies the experiment – is the transparency and scrutiny that it has brought to the public finances, built on a foundation of independence, openness, professional judgement and thoughtful communication. People are bound to disagree with some of its forecast judgements, and I am sure that there are many things it could or should have done better. But when governments are taxing, spending, borrowing and lending many billions of pounds of our money, having some independent experts keep a beady eye on what they are doing is no bad thing.
Robert Chote was Chairman of the Office of Budget Responsibility from 2010 to 2020. He is currently Chair of the UK Statistics Authority and the Northern Ireland Fiscal Council, but writes here in a personal capacity.
