João Sousa

I know - it does sometimes feel like folk won't really understand why having credible plans and a significant buffer is important until a new crisis rears its head.

The £21.7bn isstill be below the average OBR forecast error that far out, and that's with some impossible tightening built in

Our view - the £22bn headroom is a fiscal fiction. Allowing for fuel duty freezes and discounting non-credible spending cuts, it's more like £15bn - so not much higher than before, and not enough for normal forecast uncertainty

Letter: Here’s how the British economy reduces fiscal volatility From João Sousa, Deputy Director of the Fraser of Allander Institute, University of Strathclyde; Former Senior Fiscal Analyst, Office for Budget Responsibility, Glasgow, UK

My view on the two-forecast situation: it's not a time-consistent equilibrium to say there won't be policy accompanying a forecast. With Britain's institutions, the only way to reduce tinkering is to cut fiscal forecasts to one a year

www.ft.com/content/921d...

UK Budget Preview #3: Eyeing up a smorgasbord | FAI The UK Budget is still more than a week away, but it feels very long in the tooth already…

As the Chancellor reaches for the smorgasbord option on taxes, why is she in such a pickle? And is her planned headroom enough to avoid coming back next year for more?

Read our latest Budget preview below:

fraserofallander.org/uk-budget-pr...

This is clever briefing but not actually how OBR forecasts work.

The last pre-measures forecast, where this could have been true - was 31 Oct, 5 days before Chancellor's speech. Market determinants were also taken a month ago.

This suggest pulling back was the plan all along.

a photograph of professor sir John Curtice talking at our conference

What better way to round off our conference than with a look ahead to the 2026 Holyrood election with Professor Sir John Curtice? #FAI50

Mairi Spowage doing her welcome address at the FAI 50th anniversary conference in front of a large audience

a large audience in an auditorium

And we are live! What a great turn out 😁 Share your insights and pictures with us using #FAI50

A really interesting piece from @instituteforgovernment.org.uk - highlighting, as we have discussed at @fraserofallander.org as well, that the real crux of the issue is the looseness of the fiscal stance and inadequate buffer that makes every tiny shock enough to derail policy

shorturl.at/uDy1a

We Are All Lisa Cook Nobody is safe from weaponized government

If you think the attack on the Fed's Lisa Cook has nothing to do with you, you're wrong — any one of us may be next paulkrugman.substack.com/p/we-are-all...

UK fiscal issues aren't solved by ignoring poor productivity data or abolishing the OBR. Productivity growth has been averaging 0.5% a year. PSNB/GDP has averaged 4.8% since 2000-01, and 5.9% since 08-09.

This screams fiscal policy that's too loose, and policy and politicians must reckon with that

The Fraser of Allander Institute Economic and Social Policy Conference 2025 | FAI A few weeks ago, we announced our exciting plans to celebrate the Fraser of Allander’s 50th anniversary. As part of this, we were thrilled to announce that we are hosting our first economic and…

🚨 We're delighted to announce our (near) final programme for our Economic and Social Policy Conference on the 18th and 19th September, including a new live conference page on our site.

Read more: buff.ly/OK8xjug

This chart shows the decomposition of the change in the notional deficit from 2023-24 to 2024-25 - devolved spending is clearly the largest factor. Even if devolved revenues grew faster in percentage terms, devolved spending is just much larger a share of the economy

One of my favourite things of selling my last house was the estate agent telling me about his view on the outlook for Bank Rate

Firing the BLS Commissioner — the wonk in charge of the statisticians who track economic reality — is an authoritarian four alarm fire.

It will also backfire: You can't bend economic reality, but you can break the trust of markets. And biased data yields worse policy.

I have five legal surnames (even after naturalisation, thanks Portugal 🙃) and this is maddening on every flight. Have also learned that though spaces are allowed in some, surnames are all processed without spaces in the boarding passes, with weird results for initialing due to character limits

It's like everyone's forgotten the economic calculation debate and that information a priori is what's unknowable - can't calibrate a model around latent information whose results only materialise 20 years into the future.

Although it seems you can just write a report saying it 🤷

3/ A much more insightful – though perhaps less cheery – conclusion from looking at the SFC’s forecast is that by 2028-29, funding will be £0.7 billion lower than their central estimate published on 29 May

2/ It essentially assumes that no additional funding would have been made available for the Scottish Government in cash terms relative to that in 2025-26 – which is not a credible baseline

1/ We have seen Labour MPs and MSPs describing the SR event as increasing the block grant by £9.1 billion over the SR period. While Barnett consequentials add up to this, it's a figure that is neither transparent nor helpful

2025 Spending Review: a rollercoaster ride | FAI Earlier today, Rachel Reeves announced in the House of Commons the outcome of the Spending Review for both resource and capital budgets across…

The SR looks like a rollercoaster ride for public services. Boosts short-term, real-terms cuts in resource and capital in the medium-term; health and defence the main winners. ScotGov budget also grows by less than @scotfisccomm.bsky.social thought 2 weeks ago:

fraserofallander.org/2025-spendin...

8/ A cynic might suspect that the Chancellor knows this and is planning on finding a way of not having to deliver those planned cuts and avoid trade-offs on public services that are hard to stomach.

But that seems to be for another day, may even another year. Augustinian fiscal policy lives on

7/ And that leads us to what might be coming down the track. The overall envelope looks pretty undeliverable if you're going to protect some of the largest departments. It's a sort of 'mañana austerity' - this scenario is not implausible given total, but would imply a damaging rollercoaster effect

6/ And the other problem is that while it formally constrains spending, it doesn't really work as well as it seems. Look at the chart below - in every single SR period there has been significant policy in subsequent events, negating the seeming importance of this exercise

5/ The inflation and provision risk is passed to departments, which must manage it within set limits. This makes the Treasury's control function easier, but sets ministers against one another for an envelope that is broadly set already, and leads to strategic gaming that is hugely time consuming

4/ The problem? It failed to constrain public spending, particularly in the 1970s when inflation burst through the affordability of planned spending. The Treasury was forced to impose cash limits on departments to wrestle back control, and that is still largely the basis of the system

3/ That led to the creation of the Public Expenditure Survey, which intended to be an assessment of the level of provision of public services that - while moderated by medium term assessments of affordability - saw through short-term pressures and inflationary surprises

2/ As with many things in Britain, the architecture is old and creaking - and has been seen as a problem since the 1950s. A special committee recommended moving away from this tight leash of Treasury control over the process all the way back in 1961

1/ As we gear up for the UK Government's main spending review, it's worth pausing and looking at how we have ended up with this slightly bizarre system of ministers settling bilaterally with the Treasury in a sort of zero-sum world where the envelope has already been set

This is a crucial point. The two-forecast-a-year practice in the UK is a peculiarly British thing, and a remnant of the high inflation of the 1970s, when it became codified, and subsequently was put into the BRNA 2011. There's no convincing reason to do a forecast more than once a year