Three months of leaks: why the 2025 UK Budget exposed a broken process and how to fix it

The consequences of the drip, drip, drip of Treasury briefing to the press in advance were no more apparent than during the chaotic first two weeks of November. On the 4^th, the Chancellor interrupted the regularly scheduled programming of the nation’s radio and television broadcasts to all but confirm that the OBR’s underlying forecasts were bad enough to warrant breaking Labour’s manifesto commitment and increase income tax rates.

Less than ten days later, the Financial Times broke a story that the Government had at the last minute decided against the income tax increase. The Government’s line was that there had been “better than expected OBR forecasts,” even though the OBR’s published timetable clearly showed the pre-measures forecast had been locked down on 31^st October – a full five days before the Chancellor’s doom-laden breakfast speech. But the OBR, of course, could not rebut this assertion at the time.

Not a one-off

This is only an example of a pattern in which the OBR does not comment publicly before Budget day, but the Treasury can and does brief the media on what has happened to the forecasts and what it is planning to do about it. Looking at the FT’s pages alone, there were 14 articles between August and mid-November which rely wholly or partly on Treasury briefing. Language is often coded, but the pattern is unmistakable.

Table 1: Timeline of Budget process and Treasury briefing, August–November 2025

Date	FT article	Lead author	Themes	Briefing giveaway
15-Aug	Treasury prepares tax reforms in push to boost UK productivity	George Parker	£10bn productivity hit, tax reform	“said one ally of Reeves”
18-Aug	Rachel Reeves looks at UK property tax reforms in bid to boost growth	George Parker	Overhauling levies on housing, replacing stamp duty	“according to people familiar…”
19-Aug	Rachel Reeves explores reform of capital gains tax on expensive houses	Sam Fleming	Mansion tax	“according to people familiar…”
5-Sep	Sharp rise in UK pension lump sum withdrawals over tax concerns	Mary McDougall	Pension tax-free lump sum	“the Treasury is looking closely at…”
09-Oct	Rachel Reeves rules out ‘exit tax’ for wealthy people leaving UK	George Parker	Exit tax ruled out	“officials briefed on the chancellor’s thinking”
10-Oct	Rachel Reeves weighs capital gains tax increase to help plug UK’s £40bn hole	Sam Fleming	CGT increases being modelled	“according to government insiders”
21-Oct	Rachel Reeves targets tax partnerships in crackdown on wealthy Britons	George Parker	Taxing partnerships	“A Treasury spokesman […] highlighted”
27-Oct	Reeves faces £20bn hit to UK public finances from productivity doom loop	Sam Fleming	Details of the OBR forecast	“government officials have privately admitted”
31-Oct	Rachel Reeves plans Budget tax raid on expensive homes	George Parker	Higher bands on council tax	“One person close to Reeves’ thinking said”
05-Nov	Rachel Reeves set to spare UK banks from Budget tax raid	George Parker	No tax rises for banks	“one person briefed on her thinking”
07-Nov	Reeves plans £2bn Budget raid on UK retirement savings	Jim Pickard	Salary sacrifice restrictions	“government figures confirmed”
07-Nov	Income tax rise in manifesto’s ‘spirit’, say Labour figures	Jim Pickard	Income tax rise within manifesto spirit	“One Labour aide said”
13-Nov	Starmer and Reeves drop proposal to increase income tax rates	George Parker	Income tax rise dropped	“according to officials briefed on the move”
14-Nov	UK borrowing costs jump as investors lose faith in Reeves’ Budget	Ian Smith	So-called ‘smorgasbord’ approach	“People familiar with the matter”

Source: FAI analysis of FT articles

These private briefings are not a full reflection of the underlying OBR forecasts, but a collection of selective information leaked with the purpose of bedding in a narrative that serves the Treasury’s purpose, and which was in large part the driver of the subsequent debate over whether the Chancellor had been less than straight with the public.

Throughout this process, the OBR said nothing – it could not say anything given its place and role in the forecast process.

There are real costs to this setup

There’s not just a question of transparency – though of course that is important. But there are real economic costs to this endless pattern of briefing selective information.

We heard today that UK GDP contracted in October, which is unsurprising – in a volatile environment with an end date for the speculation (Budget Day), it pays to wait to avoid making a costly mistake based on incorrect information. That is even truer for business investment decisions, for which a change in government policy can be enough to render them no longer worthwhile. Market volatility too was increased, including wild swings on days of briefings such as the 14^th November.

And there is a fundamental democratic deficit too from the current setup. Both the OBR and the Treasury have full information about the interim forecast rounds as they happened; but while the former does not brief the media, the latter does, meaning that the media and the public get only a snippet of the underlying reality – the one the Treasury chooses to brief out. This means the public and other parties have access to much less and lower quality information, and have to make decisions based on that limited pool of data. Scrutiny of the government is limited because of the asymmetry of information.

How the forecast process actually works

The OBR runs an iterative process, with multiple rounds over the best part of three months. This allows it to take on new data, make judgements and get the data to converge to a stable base.

About 3 weeks before Budget Day, the pre-measures forecast is locked in. The Treasury then has three weeks to decide on policy, and the OBR uses that time to incorporate the likely effects of those policy decisions on the economy.

On Budget Day, both the OBR and the Treasury publish their documents which focus on the post-measures forecast – that is, accounting for underlying changes but also policy measures and their second-round effects.

But this “big bang” publication approach means the Treasury has had months of privileged information about the underlying forecast which they have selective leaked, while the OBR remains silent. By publication day, the narrative is set.

A simple change would have a big impact: let the OBR publish first

What if, instead of the OBR waiting until Budget Day, it published the pre-measures forecast three weeks before? Note that this is no different to the current timetable, so it wouldn’t require major changes to the working practices – but would mean that the baseline against which the Chancellor must make decisions would be available to everyone (public, markets, media, other parties) at the same time.

The Treasury would then have three weeks to respond with policy measures to formulate its budget. This is already the amount of time it has in the current process, and it would still be able to think about those changes in advance through the interim rounds. But there would be a possibility for further public debate as well regarding the size of the necessary fiscal policy response and how to approach that. Critically, that debate would be based on a known baseline rather than selectively leaked information.

A key design feature of this would be that the OBR would not publish the post-measures forecast immediately, but instead have time to consider their effects in the round at the next forecast. This is particularly an issue for what is known as second-round or indirect effects of policies on growth or inflation. These can be very hard to know with any certainty beforehand, and the OBR opining on their likely effect on the economy can put it in an impossible position which undermines its independence and credibility.

What this proposal would improve

The first benefit of the proposal would be to improve transparency and democratic accountability. It would provide the public at large with a shared baseline of information, restricting the Treasury’s window for creating a narrative without an OBR right of reply.

At the moment, the Government has all the information until the Chancellor’s speech, and the public, media organisations, think tanks, markets and opposition parties don’t get to see the underlying state of the public finances until all measures have been decided. This means, for example, that opposition parties need to reply immediately to the budget without having had time to digest the forecasts against which the Government has made its choices.

A three-week window between the OBR publishing and Budget Day would create an opportunity for parties, think tanks and the public to understand the state of the public finances and to come up with their own proposals for what should be done about any necessary changes in fiscal policy. This would be a step towards a more grown-up way of addressing the challenges to the public finances coming down the line.

The second key benefit would be to remove the OBR from its current impossible political position of judging the second-round effect of policy changes before they are even implemented. And even saying nothing means saying something. If it ignores the Treasury’s (often optimistic) claims about indirect effects, it gets accused by the Government of being overly pessimistic. And if it accepts it risks its credibility and independence and can be seen to do the Government a favour. The OBR can’t win.

In this proposal, the OBR is taken out of the firing line. The Treasury is free to make its own policy claims, as is right – so long as it owns them. If the Government is convinced of its own claims, then it should pursue those policies.

Market discipline would still apply in the first instance – Treasury claims have to be credible or investors will react badly. And the OBR then gets until the following forecast to account for these changes in the round, at a less politically charged point in time and – crucially – with more data.

And this would also reduce market volatility and business uncertainty. One objection to this reform might be that publishing forecasts without policies would create uncertainty. But that ignores the reality that we already have 3 months of leaks of partial information (see table 1). In this reform, the full baseline would be available on OBR publication day, followed by three weeks of policy debate – far better than the current 3 months of asymmetric information warfare.

A pragmatic solution

This is a proposal rooted in the current institutional framework. It uses the timings of the forecast as they already exist, and does not require lengthy and costly processes to create new legislation or institutions. And yet it aims to address the real dysfunction made abundantly clear by the 2025 Budget process.

There is an ongoing debate about the role of the OBR and how to preserve its independence. In this framework, the OBR’s technical forecasting role would be decoupled from the Treasury’s policymaking and removing the incentives to use the OBR as a political football in the fog of the pre-Budget period.

Click here for the full working paper version which documents the information asymmetry in the process, how the process would change in detail under this reform proposal and addresses potential objections. Any feedback would be welcome.