NOT regulated financial advice. Savvy Investor Guide is not regulated by the Financial Conduct Authority or the US Securities and Exchange Commission, and we are not financial advisers. Everything on this site is general financial information and education. Nothing here is personal financial advice. Before making any financial decision, consider speaking to a qualified independent financial adviser in your jurisdiction.
What this article covers: the UK Consumer Prices Index release for July 2026, published by the Office for National Statistics on 19 August; why the headline rate rose while the two measures the Bank of England watches hardest did not; what that split means for the Monetary Policy Committee meeting on 17 September; and what it does to the real return on cash savings and to mortgage pricing.
What it does not cover: a forecast of where inflation goes next. One month is one month, the August figure lands the day before the Bank decides, and this article will not pretend to know either number in advance.
In short
- UK CPI rose to 2.9% in the 12 months to July 2026, up from 2.6% in June. CPIH rose to 3.1% from 2.8%, and CPI rose 0.3% on the month.
- Core CPI did not move. It was 2.6% in July, unchanged from June. Core strips out energy, food, alcohol and tobacco, so a headline that jumps while core sits still tells you the movement came from the volatile items core excludes.
- Services inflation actually fell, from 3.6% to 3.4% on the CPI measure. Services is the gauge the Bank of England leans on hardest for domestically generated inflation. It went the opposite way to the headline.
- The ONS names the cause and it is the energy price cap. Mike Hardie, its deputy director for prices, attributes the rise to a sharp increase in gas prices following the change to the cap.
- Economists are unanimous on September. A Reuters poll of 64 economists conducted between 13 and 18 August found every single respondent expecting no change at the 17 September meeting. Bank Rate is 3.75% after a 6 to 3 vote to hold on 30 July.
- The August figure lands first. The ONS publishes August CPI at 07:00 on 16 September, the day before the Committee decides. That sequencing matters more than this month’s number.
- For savers, the arithmetic moved against you by 0.3 percentage points and nothing else changed. A 4.85% five-year rate is a 1.95% real return at 2.9% inflation, against 2.25% a month ago at 2.6%.
- The transatlantic picture has flipped. US CPI came in at 3.4%, down from 3.5%. The UK is accelerating while the US decelerates.
UK inflation in July 2026: what the numbers actually say
The Office for National Statistics published the July 2026 Consumer Prices Index at 07:00 on 19 August. The bulletin states that “the Consumer Prices Index (CPI) rose by 2.9% in the 12 months to July 2026”, up from 2.6% in June. The wider CPIH measure, which includes owner occupiers’ housing costs, “rose by 3.1% in the 12 months to July 2026”, up from 2.8%. On a monthly basis CPI “rose by 0.3% in July 2026, compared with a rise of 0.1% in July 2025”.
That is the sharpest monthly step in several months, and it is the number most coverage will lead on. Read on its own, it looks like UK inflation turning back up after a year of grinding it down.
Read alongside the rest of the release, it looks like something narrower and considerably less alarming.
The two numbers that did not move
The first is core CPI. The ONS records that “Core CPI (CPI excluding energy, food, alcohol and tobacco) rose by 2.6% in the 12 months to July 2026, unchanged from the 12 months to June”.
Core exists to answer one question: is this a broad price rise, or a few volatile things moving? It removes the four categories that swing hardest on global commodity prices and regulated pricing. If headline and core rise together, prices are climbing across the economy. If headline rises and core sits still, the movement is concentrated in what core takes out.
In July, core sat still. The entire 0.3 percentage point rise in the headline rate came from outside it.
The second number is services inflation, and this one did not merely hold. It fell. On the CPI measure services inflation went from 3.6% in June to 3.4% in July. That matters because services is the closest thing the Bank of England has to a read on domestically generated inflation. Goods prices are set substantially by global supply chains and the exchange rate. Services prices are set by British wages, British rents and British firms’ pricing decisions, which is what monetary policy can actually reach.
So the July release contains a headline rate moving up and the Bank’s preferred persistence gauge moving down, in the same month.
One technical note, because the two indices disagree here and the disagreement is real rather than an error. On the CPIH measure, services inflation was unchanged at 3.6% rather than falling. The two measures treat owner occupiers’ housing costs differently, and that difference is enough to move the services aggregate. Both figures come from the same ONS bulletin. Where this article says services eased, it means the CPI measure.
Where the rise came from: the energy price cap
The ONS does not leave this to inference. Mike Hardie, its deputy director for prices, said that “Inflation rose in July, driven by a sharp increase in gas prices following this month’s change to the energy price cap.” He added that “Other upward pressures included furniture prices falling by less than usual for this time of year, and also a smaller fall for clothing prices due to reduced discounting.”
The divisional figures in the bulletin line up with that account. Housing and household services, the division that contains domestic gas and electricity, went from an annual rate of 2.7% in June to 4.1% in July on the CPIH measure. Furniture and household goods moved from -0.2% to 1.0%, and clothing and footwear from -0.5% to 0.5%, both consistent with Hardie’s point that these categories fell by less than they usually do at this time of year rather than rising in any meaningful sense.
Pulling the other way was transport, which the bulletin names as the largest offsetting downward contribution. Its own annual rate fell from 5.7% to 3.6%.
A point on precision, since this is the kind of article where readers reasonably want the exact split. The ONS bulletin for this release does not publish a table of percentage point contributions by category. It gives the direction of the largest upward and offsetting contributions, and it gives annual rates by division. Several outlets will quote apparently precise contribution figures this week. Those are not in the ONS release, so this article names directions and divisional rates and stops there.
Why an energy cap rise is a different kind of inflation
A price cap change is a scheduled administrative event. Ofgem sets the cap on a formula, announces it in advance, and it takes effect on a known date. When it moves up, millions of household bills move up together in the same month, and the headline inflation rate registers that as a step.
Two consequences follow, and they point in opposite directions for a household.
The first is that it is entirely real. Your gas bill went up. The money left your account. Inflation measured on the things you actually buy is what erodes savings and wages, and no amount of pointing at core CPI changes the size of the direct debit.
The second is that it says very little about the future path of inflation, and interest rates are set on the future path. A cap step lifts the annual rate for twelve months and then drops out of the comparison on the anniversary, unless it is followed by another rise. That is what economists mean by a base effect. It is also, from the Bank’s point of view, close to the least informative kind of price rise there is, because raising Bank Rate does nothing whatsoever to the wholesale gas price or to Ofgem’s formula.
What this does to the 17 September decision
Bank Rate is 3.75%. The Monetary Policy Committee held it there on 30 July on a 6 to 3 vote, with the three preferring a rise. The next decision is Thursday 17 September.
Economists are unusually united. A Reuters poll conducted between 13 and 18 August, covering 64 economists, found that “All respondents forecast no change at the central bank’s next meeting in September.” The same poll reported that 56 of the 64, close to 90%, expect Bank Rate to be unchanged at 3.75% at the end of 2026, up from 83% of respondents in the July poll.
Market pricing agrees on direction and is vaguer than it looks on magnitude. Three separate trackers of overnight index swap pricing, all sampled within the same few days in mid August, put the probability of a hold at 72%, 79% and 87.1% respectively. They agree that a hold is the heavy favourite, that a rise is the tail risk being priced, and that a cut is priced at close to zero. They do not agree on the number, and none of them is a primary source, so treat “the market says roughly 75% to 85% hold” as the honest version and be suspicious of any article quoting a single figure to one decimal place.
We could not establish what the Bank itself thinks. As at 21 August we could find no on-record comment from any MPC member on this specific release, and the Bank’s own website was not reachable to us at the time of writing. That is a gap in what we know, not evidence that the Bank has said nothing.
The sequencing is the part worth diarising. The ONS publishes August CPI at 07:00 on 16 September, the day before the Committee meets. The Committee will therefore have a fresher number in front of it than the one in this article. If the August figure shows the cap step washing through with core still flat, July looks like a blip. If August shows core and services turning up too, the three dissenters from July have a considerably stronger case.
What it means for savers
The arithmetic is simple and it moved against you by exactly 0.3 percentage points. Real return is the rate you earn minus the rate prices rise. At 2.6% inflation, an account paying 4.85% returned 2.25% in real terms. At 2.9%, the same account returns 1.95%. Nothing about the account changed. The yardstick moved.
Using rates we verified at source this week, NS&I’s five-year Guaranteed Growth Bonds pay 4.85%, which is that 1.95% real return. Its Premium Bonds prize fund rate rises to 4.35% from the September draw, which is 1.45% above current inflation, with the standing caveat that a prize fund rate is an average across a prize distribution and is not what a typical holder with a modest balance actually receives. We set that out in our Premium Bonds piece and it gets more important as the advertised rate rises, not less.
Two things are worth holding together. Cash is still beating inflation, which was not true for most of 2022 and 2023. And tax has not gone anywhere: for a higher rate taxpayer who has used their Personal Savings Allowance, a 4.85% gross rate is 2.91% net, which against 2.9% inflation is a real return of roughly nothing at all. The headline rate on the poster and the number that reaches your pocket are different numbers, and inflation is only one of the two things separating them.
Alice Haine of Hargreaves Lansdown put the general case plainly after the release: “Inflation is never good news for savers. Even if it slows the fall in cash rates, rising prices still erode the real value of interest earned and reduce spending power over time.”
What it means for mortgage holders
Fixed mortgage rates do not track Bank Rate. They track the swap curve, which prices what markets expect Bank Rate to average over the term of the deal. That is why fixed rates can move in a week when the Bank has not met.
Haine noted after the print that “Two- and five-year swap rates have moved higher, which can feed into mortgage pricing, although competition among lenders has led to some rate cuts this week.” Both halves of that sentence are doing work. Swaps moved up on the inflation number, which pushes fixed pricing up. Lender competition is pushing it down. HSBC cut across its remortgage range on 19 August, the same morning the CPI figure was published, which is not a contradiction: those cuts were priced off the swap curve weeks earlier, and lenders reprice upwards faster than they reprice down.
The practical read for anyone with a fix ending in the next six months is that this release did not change the shape of the decision. It removed a little of the case for rates falling sooner. We track the lender moves in more detail in our running piece on the mortgage repricing waves, and the transmission from gilt yields to swap rates to your fixed rate is set out in our gilt yields explainer.
The transatlantic picture has flipped
In May we wrote about the April prints landing in opposite directions, with UK inflation falling to 2.8% while US inflation rose to 3.8%. Three months later the two economies have swapped roles.
The US Bureau of Labor Statistics reports that “The all items index rose 3.4 percent for the 12 months ending July after rising 3.5 percent for the 12 months ending June.” US core inflation was 2.5% in July, down from 2.6%. So American headline inflation is easing while British headline inflation accelerates, and the two core rates have converged to within a tenth of a point of each other.
The mechanism is the same one that drove the April divergence, running the other way. Both countries face broadly similar global energy costs. What differs is the regulatory plumbing on the household side. Britain passes wholesale gas costs to households through a capped price that resets on a schedule, so the pass-through arrives in visible steps. The United States does not have an equivalent national mechanism, so its energy pass-through is smoother and less synchronised. Neither arrangement is better at controlling inflation. They just distribute the same underlying cost differently across the calendar. The comparison is in our April divergence piece.
What this does not mean
- It does not mean inflation is back. One month in which core did not move and services fell is not a trend reversal. It is a scheduled administrative price change showing up where it was always going to show up.
- It does not mean the Bank will raise rates in September. Every economist in a 64-strong Reuters poll says otherwise, and market pricing puts a cut at close to zero and a rise as the tail.
- It does not mean your energy bill is going to keep rising at this pace. A cap step lifts the annual comparison for twelve months and then falls out of it, unless another rise follows. Ofgem’s next announcement is the thing to watch, not this figure.
- It does not mean cash has stopped beating inflation. At 2.9% it still does, before tax. After tax, for a higher rate taxpayer who has used their allowance, it roughly does not.
- And it does not settle anything. The August figure arrives on 16 September, the day before the Committee decides. That number, not this one, is the one that will move the meeting.
FAQ
Why does the headline rate matter if core did not move?
Because the headline rate is the one measured on the things you actually buy, and it is the one used to uprate a great many payments, from index-linked pensions to some student loan interest. Core is a diagnostic tool for reading where inflation is coming from and where it is likely to go. It is not a measure of your cost of living. Both numbers are useful, for different questions.
What is the difference between CPI and CPIH?
CPIH is CPI plus owner occupiers’ housing costs and Council Tax. It covers a broader slice of household spending, which is why the ONS designates it the more comprehensive measure, and it usually runs a little higher than CPI. In July 2026 CPIH was 3.1% against CPI’s 2.9%. The Bank of England’s 2% target is set against CPI, so CPI is the number that formally matters for interest rate decisions.
Will this push up my mortgage rate?
Not directly, and not by itself. Fixed rates are priced off swap rates, which moved up somewhat on the release. Lender competition has been pushing in the other direction at the same time, with cuts announced in the same week. The net effect on any individual product depends on which of those two forces is larger for that lender in that week. Variable and tracker rates move with Bank Rate, which was not changed by this release.
When is the next inflation figure published?
The ONS publishes Consumer price inflation, UK: August 2026 at 07:00 on 16 September 2026. That is the day before the Monetary Policy Committee announces its decision on 17 September, so the Committee will have seen it.
Is cash still worth holding at 2.9% inflation?
That is a question about your circumstances rather than about the inflation rate, and we do not give personal recommendations. What we can set out is the arithmetic. A gross rate above 2.9% is a positive real return before tax. Whether it stays positive after tax depends on your marginal rate and how much of your Personal Savings Allowance is left, and for a higher rate taxpayer who has used the allowance the margin at current rates is close to nil.
The Savvy Investor’s take
The useful skill on an inflation release is knowing which number answers which question. This one contained several, and they pointed in different directions.
If the question is “what is happening to my cost of living”, the answer is 2.9% and it went up, and the reason is sitting in your gas bill. If the question is “what will the Bank of England do”, the honest answer is that this release barely moved the argument, because the Committee sets policy against domestically generated persistence and both of the measures it uses for that either held still or fell.
What we would resist is the framing that will dominate the week, which treats 2.6% to 2.9% as an inflation problem returning. A regulated price cap moving on a published schedule is the most predictable component in the entire index. It was always going to do this. The number that would genuinely change the picture is core, and core did not move at all.
Diarise 16 September rather than reacting to 19 August.
Information, not advice. This article describes the UK Consumer Prices Index release for July 2026, published on 19 August 2026, and discusses the policy and market mechanics that follow from it. It is not personal financial advice. Savvy Investor Guide is not authorised or regulated by the Financial Conduct Authority. Where a regulated decision is involved, consult a qualified, FCA-authorised adviser. For free, government-backed guidance, contact MoneyHelper.
Key sources
- Office for National Statistics, Consumer price inflation, UK: July 2026, published 19 August 2026. Headline, CPIH, core, monthly change, divisional rates, services and goods.
- Office for National Statistics, Consumer price inflation, UK: August 2026 release page, giving the 16 September 2026 07:00 publication time.
- US Bureau of Labor Statistics, Consumer Price Index, July 2026, for the US headline and core comparison.
- Reuters poll of 64 economists conducted 13 to 18 August 2026 on the Bank of England rate path, reported 18 August 2026.
- Comment from Alice Haine of Hargreaves Lansdown on the savings and swap-rate implications, as reported on 19 August 2026.