# A real tough choice: freeze the cap, and most of it pays for itself

_The price cap for January is set in November and, on where gas is now, it could go up by more than 20%. Freezing it would cost about £10bn. Lower inflation would hand roughly £7bn of that straight back through the index-linked debt book._

Neil Woodford · 21 September 2026 · 6 min read

![The cover of Private Eye from 12 November 1976: Healey Spells It Out](https://cdn.sanity.io/images/v3acfbvo/production/e7af4fd79777ca943b51826a78f1ad62a3e0e94e-1536x1024.png?w=1600&fit=max&auto=format)

---

Although Andy Burnham has claimed that he is, in effect, [match fit to make tough choices](https://www.noisecancelling.co/read/if-the-cap-fits), the truth is that so far he has made none. However, before too long he will be confronting them thick and fast. The first will come with the budget in October, when the Prime Minister and his Chancellor will have to grapple with significantly less fiscal headroom thanks to changed conditioning assumptions in the OBR's economic model. My sense is that the giveaways he will inevitably want to take credit for will have to be financed with tax increases, because the tough choice to cut welfare spending is one he will not confront.

_Neil on Burnham and his "difficult decisions":_ [If the cap fits: tax and spend, with better TikTok videos](https://www.noisecancelling.co/read/if-the-cap-fits) — Andy Haldane says the markets now see a traditional tax-and-spend socialist government with better TikTok videos. The Prime Minister’s reply was to cite scrapping digital ID as a difficult decision.

The other tough choice, and the one I want to focus on here, will come later in November when the energy price cap for the three months from January next year will be fixed. Based on where wholesale gas prices are now, and assuming they stay there, the cap could be going up by slightly more than 20%, and that is on top of the cumulative 17% increase over the last six months. _(Neil in the margin: Ofgem sets the cap four times a year and announces each level about five weeks before it takes effect. The January to March level is calculated from wholesale costs observed in a window that closes on 18 November, and is announced in late November. The cap now is £1,723 a year for a typical dual fuel household, from 1 October.)_

![Every new forecast for January is higher than the last](https://r4at4qm6kmohrtvq.public.blob.vercel-storage.com/charts/chart-tough-choice-jan-cap-8926eca0f30f-light.png)

_The window that sets the January cap does not close until 18 November. Every forecast made since gas took off in September has been higher than the one before it._

## The gas price, not the oil price

Interestingly, it is the gas price that is causing the biggest inflation headache in the UK now, and not the oil price. To put this into context, oil prices averaged $80 a barrel between 2010 and 2019 and have so far averaged $95 in 2026, despite the industry going through what the IEA has called [the greatest energy security threat in history](https://www.atlanticcouncil.org/event/the-ieas-fatih-birol-on-global-energy-market-resilience-in-a-moment-of-crisis/). That $15 premium, whilst uncomfortable, is not unbearable, whereas the doubling of the natural gas price in the UK since the start of July, to very nearly 200p per therm, is quickly becoming very uncomfortable indeed.

![2022 makes this year’s gas spike look tame](https://r4at4qm6kmohrtvq.public.blob.vercel-storage.com/charts/chartDoc.chancellor.c1-uk-gas-f014cee18c52-light.png)

_The 2022 spike peaked at 568p a therm on 29 August. Gas is 184p now. Whatever this year’s rise is, it is not 2022, and in 2022 somebody capped the bill._

_[Embedded media](https://www.cornwall-insight.com/press-releases/energy-bills-forecast-to-hit-three-year-high/)_

> Our energy bills remain tied to events thousands of miles away
>
> — [Dr Craig Lowrey, Principal Consultant, Cornwall Insight](https://www.cornwall-insight.com/press-releases/energy-bills-forecast-to-hit-three-year-high/)

For a leader who highlighted, on a recent tour of the UK, a singular and primary focus on alleviating the cost of living, this is not something that he can duck. However, one would be forgiven for thinking that he is between a rock and a hard place, given that his predecessor has already maxed out the credit card and significantly increased tax to pay for it, giving Mr Burnham only a very limited set of options. Having said that, fortunately there may be a way for him to significantly soften the blow of this energy price impact without increasing taxes or cutting spending to pay for it, albeit that it is not obvious and might cut across some obtuse socialist ideology. It might also confront resistance from his mate next door and from the Treasury, based on the "not invented here" principle, but it could help to solve a big problem for him if gas prices remain where they are now. Here is how.

## Yields are a rounding error. Inflation is not.

Contrary to consensus thinking on [gilt yields and funding costs](https://www.noisecancelling.co/read/the-fed-hikes-the-bank-holds-and-gilts-rally), although it is true that higher ten-year yields, if maintained, will in the long term increase the debt interest burden, the impact of higher conventional yields in the short term is pretty minimal: basically a rounding error in terms of total managed spending of £1.4trn in 26/27, and total debt interest costs of £111bn. Having said that, interest costs are materially sensitive to inflation outcomes because of their impact on index-linked coupons and principal. The outstanding book of index-linked government debt is just under £690bn, or about a quarter of the total book of debt. In summary, if inflation increases by a certain percentage, an index-linked bond's principal value increases by a corresponding amount. Although the fixed coupon percentage does not change, because it is calculated as a percentage of the upwardly adjusted principal, it also increases. This means that, in round terms, if there is a 1% increase in UK inflation over twelve months, the increased cost of servicing the book of index-linked debt is just under £7bn. Consequently, if via some mechanism the government could intervene to lower recorded inflation (CPI), then there would be a corresponding saving in the index-linked debt interest bill, amongst other savings, and hey presto, a saving without cutting welfare. _(Neil in the margin: Index-linked gilts uprate both the coupon and the principal in line with the RPI, with a lag. The uplift on the principal accrues every year but is only paid out at redemption, which is why a year of higher inflation shows up in the debt interest line long before anyone is handed the cash.)_

**£688.5bn** — The stock of index-linked debt at end-December 2025, a quarter of the government's wholesale debt portfolio

![A quarter of the debt book moves with inflation](https://r4at4qm6kmohrtvq.public.blob.vercel-storage.com/charts/chart-tough-choice-index-linked-53927092bb0b-light.png)

_This is the part of the debt that reacts to a price cap decision. Yields matter over decades; the index-linked book reprices with the next RPI print._

_[Embedded media](https://www.gov.uk/government/publications/debt-management-report-2026-27/debt-management-report-2026-27)_

## What happens if he does nothing

But before outlining what the Prime Minister could do, I should explain what is likely to happen without any intervention. The energy price cap is fixed four times a year and it is one of the biggest single administered prices in [the CPI basket](https://www.noisecancelling.co/read/inflation-myths). If gas prices remain where they are now until November, then it is likely that this part of the basket will increase by just over 20% for the period January through to March, which will, with slightly higher food prices, push headline inflation close to about 4% in January. This in turn, being about 1% above where inflation is now, will increase index-linked debt funding costs by about £7bn if maintained for a full twelve months, which I think is unlikely, but it helps with the maths. _(Neil in the margin: CPI inflation was 3.1% in August, up from 2.9% in July, on the ONS release of 16 September. Motor fuels did most of the work. The January cap would arrive on top of that.)_

So, without any intervention, here is another government spending increase Mr Burnham and Mr Healey will not have bargained for. And of course, higher headline inflation also directly increases other expenditure items the government has to fund.

## Freeze it

However, the action that could provide significant relief here, both for consumers and for the government, is if Mr Burnham decided to freeze the energy price cap at its current level in November. That would have a significantly positive impact on headline inflation, but it would come at a cost as well.

First to the cost. A 20% increase in the cap will lead to a £350 increase if maintained for a full year. Assuming all 29 million households in the UK are gas consumers, which is clearly not true but helps with the maths, that leads to an increase in energy costs for all households of £10.15bn, and the same full-year cost to the government if it steps in to prevent the increase. The £10bn would be paid to the gas industry to compensate them for the difference between the retail and the wholesale price. Clearly it cannot afford that, but it is not the whole picture, because if the government intervenes the CPI does not get pushed up in January. If we assume that CPI as a result is 1% lower for a full year because of the intervention, and it may not be as much as this, but the maths is easier, that will save £7bn in index-linked interest costs. So the net cost is a much more manageable £3bn, and probably less given other savings the government would benefit from as a result of lower inflation. A lower inflation rate would of course also deliver other benefits to business and households in the form of lower interest rates, with the MPC not under pressure to hike, and lower mortgage rates.

![What a freeze costs, once the debt book pays its share](https://r4at4qm6kmohrtvq.public.blob.vercel-storage.com/charts/chart-tough-choice-freeze-arithmetic-11c47947106d-light.png)

_The gross number is the one that gets quoted. The net number is the one that matters, and it is a third of it._

So, if Mr Burnham had the minerals to do this, and Liz Truss did back in 2022, he not only really does make a tough decision but also matches the cost-of-living rhetoric. And maybe if normality returns to energy markets by the spring, the whole thing goes away and his political capital is increased. The potential blocker, though, is not insubstantial for those in the Labour party wedded to socialist principles rather than to pragmatism, because a scheme like this provides relief to all households. It is quick and very cheap to implement: the government only has to write a few cheques to a few energy suppliers. It is not means tested, and so the government ends up giving relief to richer households. Some in the party could not swallow this, placing principle above pragmatism despite the benefit to all and to the wider economy. _(Neil in the margin: The Energy Price Guarantee, announced in September 2022, held the typical annual bill at £2,500 by paying suppliers the difference between that and the cap. It was originally to run for two years; Jeremy Hunt cut it back to six months a month later.)_

## Will it happen?

When this idea was put in front of Rachel Reeves, political resistance meant it was not adopted, and the same self-serving nonsense might again prevent what to me looks like a simple and eminently sensible idea from being implemented. We will see if Burnham and [Healey](https://www.noisecancelling.co/read/the-new-chancellors-misdiagnosis) are more pragmatic than their predecessors, but if they do nothing, the economic hole that is currently being dug for them will get deeper and deeper by the day.