# Clean bowled again

_The consensus said the economy would flatline in July. It grew 0.4%, and the first half of 2026 was the fastest in the G7. The budget on 28 October will be built on five-year projections from the institution that missed all of it._

Neil Woodford · 14 September 2026 · 7 min read

![A cricketer is bowled out](https://cdn.sanity.io/images/v3acfbvo/production/5a13b63c5ec945920344f3bd16e93e2ebfcb73dd-2484x1634.jpg?w=1600&fit=max&auto=format)

---

My favourite cricketing metaphor, which has featured quite regularly in recent editions of Noise Cancelling, was the first thing that came to mind when I saw [the UK's GDP data](https://www.ons.gov.uk/economy/grossdomesticproductgdp/bulletins/gdpmonthlyestimateuk/july2026) on Friday last week. Instead of the zero growth in July that consensus had forecast, once again the UK economy has surprised doom-laden expectations by delivering growth of 0.4%.

_[Embedded media](https://x.com/ONS/status/2098290472879825192)_

July's better than expected outcome followed a similarly better than expected outcome in the first quarter of 0.6% and steady growth in the second quarter of 0.4%. According to the ONS, GDP in July is also estimated to be 1.6% higher than in the same month in 2025. Given that growth in the third and fourth quarters last year was so disappointing (0.1% in both quarters) the comparators going into the final months of this year are pretty favourable. _(Neil in the margin: The Office for National Statistics, the UK's official statistics agency and the body that compiles GDP. Neil leans on it heavily here as the counterweight to the forecasters he's needling.)_

![Flat for six months, then away again](https://r4at4qm6kmohrtvq.public.blob.vercel-storage.com/charts/chartDoc.cleanbowled.c1-monthly-gdp-7ba4aea14f08-light.png)

_From July 2025 to January 2026 the economy went precisely nowhere. That is the stretch the doom-laden commentary was written about. Since January it has grown 1.5%, and nobody has rewritten the commentary._

![The two quarters the forecasters are still catching up with](https://r4at4qm6kmohrtvq.public.blob.vercel-storage.com/charts/chartDoc.cleanbowled.c2-quarterly-88b0b78200be-light.png)

_0.6% and then 0.4%. Set against 0.1% in each of the last two quarters of 2025, which is what makes the comparators for the rest of this year so forgiving._

Even if the economy slows from July's robust backdrop, this year is going to come in a long way above what was forecast for it. The arithmetic is worth doing properly, because it is more striking than it first looks. **For the consensus figure of 1.1% to be right, the economy has to stop growing altogether for six months.** If the second half merely repeats the second quarter's pace, the year lands at about 1.4%; at anything like the first quarter's pace, 1.5% is in sight. And July has already started the third quarter 0.6% above the second quarter's average level, so a good part of that is banked before August and September are even counted.

## The detail is as interesting as the headline

Whilst production (industrial production, which includes manufacturing) was down over the latest three months, as was construction, services output, which is by far the largest component of the economy at about [80% of output](https://www.ons.gov.uk/economy/grossdomesticproductgdp/articles/bluebook2026industryimpactanalysis/2026-08-20), was up 0.6% over the same period. In the month of July itself all three sectors grew: services by 0.4%, production by 0.2% and construction by 0.1%.

The growth was concentrated in the parts of the economy you would want it to be concentrated in. Professional, scientific and technical activities rose 2.1% over the three months, driven by a 7% jump in scientific research and development. Information and communication rose 2.5%, mainly on a 4.4% increase in computer programming, consultancy and related activities. The ONS is careful about why, and I think rightly so: [“many of the businesses reporting the largest turnover in July 2026 are involved in activities related to artificial intelligence and cloud computing”](https://www.ons.gov.uk/economy/grossdomesticproductgdp/bulletins/gdpmonthlyestimateuk/july2026), though “it is difficult for us to quantify the exact impact”. That is an honest answer, and a more interesting one than a firm number would have been.

![The bit of the economy that is actually growing](https://r4at4qm6kmohrtvq.public.blob.vercel-storage.com/charts/chartDoc.cleanbowled.c6-services-ded9d70a8df4-light.png)

_Computer programming and consultancy grew 4.4% in three months. The ONS says the firms reporting the biggest turnover in it are doing AI and cloud work, but that it cannot put a number on that. Neither can I. It is worth watching._

_[Embedded media](https://www.youtube.com/watch?v=aQXkc8Kc92w)_

## The usual cast of institutional nobility

The reason clean bowled came to mind is that, predictably, this outcome is considerably better than was forecast by the economic establishment. Both [the Bank of England](https://www.bankofengland.co.uk/monetary-policy-report/2026/july-2026) and [the OBR](https://obr.uk/efo/economic-and-fiscal-outlook-march-2026/) have 1.1% for 2026 (the Bank's forecast was updated in July, the OBR's in March), and [the OECD](https://www.oecd.org/en/publications/2026/06/oecd-economic-outlook-volume-2026-issue-1_8be0dba6/full-report/united-kingdom_ee1bed4d.html) is more downbeat still at 0.9%. [The independent forecasters ](https://www.gov.uk/government/statistics/forecasts-for-the-uk-economy-august-2026)[HM Treasury polls](https://www.gov.uk/government/statistics/forecasts-for-the-uk-economy-august-2026) average 1.1%, in a range of 0.7% to 1.3%. The first half of the year, on its own, delivered 1%. _(Neil in the margin: The Treasury gathers and publishes a monthly survey of City and academic economists' forecasts. The point of citing the average is that no serious independent forecaster came close either.)_

I won't go on, but suffice to say that the usual cast of institutional nobility has once again got it totally wrong, as they did with inflation, [as I wrote a fortnight ago](https://www.noisecancelling.co/read/inflation-myths). I also [set this argument out in August](https://www.noisecancelling.co/read/june-gdp-and-the-forecasts), when June's number did the same thing to the same forecasts. The pattern is not that the official forecasters are occasionally wrong. It is that they have been wrong in the same direction, on the same variables, all year.

_[Embedded media](https://www.investing.com/news/economic-indicators/uk-economy-grew-04-in-july-4897234)_

> Britain's economy has slowed after a strong start to the year, but growth of 0.4 per cent in the second quarter still leaves the UK leading the pack ahead of its G7 peers.
>
> — [Stephen Hunsaker, Economist, Resolution Foundation](https://www.resolutionfoundation.org/press-releases/economy-slows-but-doesnt-stall-as-the-uk-economy-leads-the-g7-on-growth/)

## Nostradamus, and the forecasts nobody marks

Nevertheless, as usual, the OBR's projections will be revered as some kind of modern-day prophecy from Nostradamus, and the financial media will in turn slavishly opine on how dreadful the outlook for everything is. The trouble, of course, is that the OBR could not forecast Christmas.

Its track record is as awful as the Bank of England's. This year both institutions have again failed dismally to get anywhere near the right forecast for growth. Their inflation forecasts fared a little better, although the Bank's attempt to quantify the impact of the conflict in the Gulf was ludicrously wide of the mark, [as I set out at the time](https://www.noisecancelling.co/read/consensus-clean-bowled-again). And most ironic of all, it was only last November that [the OBR downgraded its long-term UK productivity growth forecast from 1.3% to 1%](https://obr.uk/efo/economic-and-fiscal-outlook-november-2025/), just a few months before [the ONS revised up its productivity calculations](https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/labourproductivity/articles/ukproductivityintroduction/januarytomarch2026andoctobertodecember2025) after admitting that [its Labour Force Survey](https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes/articles/labourforcesurveyqualityupdate/july2026) could no longer be relied upon as the denominator in that critical ratio. I have written about that at length, twice: [the short version](https://www.noisecancelling.co/read/fessing-up-uk-productivity) and [the long one](https://www.noisecancelling.co/read/the-great-uk-productivity-myth). Even the FT's Martin Wolf has today half-heartedly acknowledged that the UK's productivity performance has been better than had previously been measured in recent years. _(Neil in the margin: Productivity is output divided by hours (or workers) worked. If the Labour Force Survey overstated that bottom line, measured productivity was understated — hence the upward revision when the ONS admitted the survey was unreliable.)_

## Why any of this matters now

The particular relevance of these numbers now is that we are in the run-in to [the budget on 28 October](https://www.gov.uk/government/publications/chancellor-letter-to-the-treasury-select-committee-tsc-budget-2026-date), when we will be informed by Mr Healey what the government will be doing with tax and public spending within the growth and borrowing constraints that the OBR will place around it. Those constraints will be a direct product of the OBR's longer-term projections – or conditioning assumptions, as they are called – over the next five years.

My question is this. If the OBR cannot be relied upon to get anywhere close to near-term growth outcomes, why should it be relied upon to get its five-year projections right, and why should these finger-in-the-air forecasts be the basis for tax and spending decisions taken now?

_[Embedded media](https://dailybusinessgroup.co.uk/2026/09/boost-for-healey-as-uk-economy-in-surprise-expansion/)_

I think the controversial answer is that this process is fundamentally flawed, and that the OBR's economic musings should not be relied upon to drive policy decisions. Interestingly, barely anyone else does it this way. [Belgium](https://www.plan.be/en/publications/fiscal-councils-independent-forecasts-and) and [the Netherlands](https://www.cpb.nl/en) have a similar structure, where an independent body's forecasts are formally adopted by the government. Other developed economies, including Canada and the United States, have independent bodies that make forecasts and evaluate policy decisions – [the Congressional Budget Office](https://www.cbo.gov/about/overview) is the obvious example – but leave macro-economic forecasting to their finance ministries. The output of those independent bodies is treated as an advisory input. It is not binding.

## Very little they can control

In fact, what Mr Healey and his Prime Minister are going to find out quite quickly is that despite all the talk of radicalism, there is very little that they can control. [I went through the Chancellor's first big speech a few days ago](https://www.noisecancelling.co/read/the-new-chancellors-misdiagnosis), and the gap between the ambition and the levers is the whole story. Consider what is actually on the list of things that will be decided for them rather than by them:

- Oil and gas prices, and household energy prices.

- Treasury yields and gilt yields.

- Fiscal headroom, and the OBR's conditioning assumptions that set it.

- Interest rates.

- And even government spending, which is the one thing they should be able to control and can't.

Because these things will be decided for them and not by them, this government's room for manoeuvre is severely limited, and made all the more so by [the reckless spending and tax decisions taken in the first two years of office](https://www.noisecancelling.co/read/how-not-to-run-an-economy). This might seem a little left field, but I wonder if both of them might be wishing that [Liz Truss](https://www.instituteforgovernment.org.uk/article/comment/blocking-obr-forecasts-undermines-credibility-liz-trusss-economic-plans)[ had been a little more successful when she first sidelined the OBR in September 2022](https://www.instituteforgovernment.org.uk/article/comment/blocking-obr-forecasts-undermines-credibility-liz-trusss-economic-plans) and then [called for it to be abolished](https://finance.yahoo.com/news/liz-truss-calls-obr-abolished-060000562.html). _(Neil in the margin: Truss's September 2022 mini-budget bypassed the OBR entirely and triggered a gilt-market rout that ended her premiership in weeks. Neil's tongue is firmly in cheek about wishing that campaign against the watchdog had gone better.)_

## A clean pair of heels

Once again, the economic consensus has got the UK economy wrong. Instead of the pedestrian growth rate most institutions have been expecting, the UK economy appears to have shown a clean pair of heels to the rest of the G7 so far in 2026. It is hardly super-charged, and two quarters is not a trend. But given the headwinds of higher taxes, excessively high interest rates and higher energy prices, it is a pretty good outcome, and I think it bodes well for a better growth outcome in 2027 too.

![A clean pair of heels](https://r4at4qm6kmohrtvq.public.blob.vercel-storage.com/charts/chartDoc.cleanbowled.c4-g7-b9679db9182a-light.png)

_Hardly super-charged, and two quarters is not a trend. But this is the economy that was supposed to be the sick man of the G7, and it is currently top of the table._

## Postscript

This is a critical week for both the US and UK economies, given that [the Fed](https://www.federalreserve.gov/monetarypolicy/fomccalendars.htm) and [the Bank of England](https://www.bankofengland.co.uk/monetary-policy/upcoming-mpc-dates) will both be making interest rate decisions – the Fed on Wednesday, the MPC on Thursday. Market consensus is that the Fed will raise and the Bank will hold. I am not so sure about the first of those.

Given what is happening to [core inflation](https://www.bls.gov/news.release/cpi.nr0.htm)[ in the US](https://www.bls.gov/news.release/cpi.nr0.htm) (it fell to 2.4% in the latest data, the lowest since March 2021) and to [wage growth](https://www.bls.gov/news.release/empsit.nr0.htm), which is just over 3% and falling, I think the Fed will decide to keep rates on hold. _(Neil in the margin: Core strips out volatile food and energy prices to show the underlying trend the Fed actually targets. At 2.4% it's near the 2% goal, which is why Neil doubts a rate rise.)_

![Neither of these is asking for higher rates](https://r4at4qm6kmohrtvq.public.blob.vercel-storage.com/charts/chartDoc.cleanbowled.c8-us-core-wages-764379144e31-light.png)

_Core inflation at 2.4%, the lowest since March 2021, and wage growth at 3.1% and falling. Look at these two lines and tell me the case for tightening._

As for the MPC, [three members of the committee voted for an increase in July](https://www.bankofengland.co.uk/monetary-policy-summary-and-minutes/2026/july-2026) and will presumably do so again, but my bet is that the rest of the nine will vote for the much more sensible outcome, which is to hold rates steady. [I still think the next move is down](https://www.noisecancelling.co/read/why-i-can-see-uk-interest-rates-below-3-next-year). _(Neil in the margin: The Bank of England's Monetary Policy Committee — the nine-member body that sets Bank Rate. A split with three dissenters, as in July, signals how finely balanced the decision is.)_

As for the oil price, I am convinced that something pretty odd is going on in the oil market and that the spike above $100 is hard to justify, [as I wrote on Friday](https://www.noisecancelling.co/read/oil-over-100-and-the-ecb-reaches-for-the-cricket-bat). In this respect I am expecting the price to ease back to where it came from a couple of weeks ago, when it was trading at about $90 a barrel. It should be interesting.

![Twenty dollars in a fortnight](https://r4at4qm6kmohrtvq.public.blob.vercel-storage.com/charts/chartDoc.cleanbowled.c7-brent-356d7feb40cd-light.png)

_Twenty dollars in a fortnight on a limited exchange of fire, while tanker volumes through Hormuz have been recovering. I do not think this price is describing the physical market._