July CPI lands on Wednesday, August 12. June’s report was the softest inflation print in years — headline down to 3.5%, core at 2.6%, and the largest one-month decline in the index since April 2020. Almost all of that came from energy, which is precisely why July is the more informative release.
This article is for informational and educational purposes only. It is not financial advice, investment advice, or a recommendation to buy, sell, or hold any security, cryptocurrency, or financial product. Always verify data with official sources before making financial decisions.
Release time and where the data lands
Short answer: The Bureau of Labor Statistics publishes the July Consumer Price Index at 8:30 a.m. Eastern on Wednesday, August 12, 2026. The release contains headline and core CPI on both a monthly and annual basis, plus the component detail — shelter, energy, food, medical care, transportation services — that determines how the number gets interpreted. Markets trade the monthly core figure and the shelter line more than the headline annual rate.
Starting point: what June actually showed
| Measure | June 2026 | Prior / context |
|---|---|---|
| Headline CPI, year over year | 3.5% | Down from 4.2% in May; consensus was 3.9% |
| Headline CPI, month over month | −0.4% | Largest one-month decline since April 2020 |
| Core CPI, year over year | 2.6% | Down from 2.9%; consensus was 2.9% |
| Core CPI, month over month | Flat | |
| Shelter, month over month | +0.1% | Smallest monthly increase since January 2021 |
| Energy, month over month | −5.7% | Gasoline −9.7%, electricity −1.0% |
The report beat expectations on every headline measure. It is worth being precise about why: energy fell 5.7% in a single month, and gasoline alone dropped nearly 10%. Strip that out and the picture is softer than May but not dramatically so — core was flat on the month rather than negative.
The three lines that move markets
Core month over month. This is the number the rates market reacts to. The annual rate is arithmetic on twelve prior months and contains little new information; the monthly core change is the only part of the release that is genuinely new. A 0.1% versus 0.3% core print is the difference between a benign report and a problem, regardless of what the annual headline says.
Shelter. Roughly a third of the CPI basket and the largest single component. June’s 0.1% was the smallest monthly increase in more than five years, which is a meaningful development given that shelter’s lag behind market rents has been the main reason core inflation stayed elevated. The question for July is whether that was a genuine turn or a single soft month in a noisy series.
Core services excluding housing. Sometimes called supercore. It captures wage-sensitive service prices and is the component most closely tied to labour market conditions. It is not published as a headline figure — it has to be constructed from the detail tables — which is why it appears in commentary a few hours after the release rather than immediately.
Energy base effects after June’s collapse
June’s energy decline was large enough to distort the July comparison in a specific way. A one-month drop of that size does not repeat; if energy simply stops falling, the headline monthly figure mechanically rises even with no change in underlying inflation.
This is the standard base-effect trap and it produces a predictable pattern of misreading. A July headline of, say, 0.0% or +0.1% month over month would represent a substantial deceleration in the disinflationary impulse relative to June while looking like a benign number in isolation. The annual rate can fall while the monthly momentum deteriorates, and both statements can be true in the same release.
Goods prices after the tariff ruling
The Supreme Court struck down tariffs imposed under the International Emergency Economic Powers Act on February 20, 2026, in a 6-3 decision. Trade policy since has run through other statutory authorities, and the effective tariff rate has been in flux rather than simply falling to zero.
For CPI purposes the relevant question is core goods, which had been the channel through which tariff costs reached consumer prices. Import price pass-through operates with a lag of several months and is diluted by retailer margin absorption, inventory bought at old duty rates, and substitution across suppliers. A change in tariff policy in February does not produce a clean signal in July core goods — it produces a gradual, hard-to-isolate drift that is easily confused with ordinary goods disinflation.
Treat any commentary that attributes a specific decimal of the July core print to tariff policy with scepticism. The lag structure does not support that precision.
What the rates market has already priced
The Fed held at 3.50%–3.75% in July, with no fresh Summary of Economic Projections, leaving the June dot plot as the most recent official signal. There is no FOMC meeting in August; the next decision is September 15–16.
That gap matters for how this release trades. With five weeks between the CPI print and the next decision — and Jackson Hole in between on August 27–29 — a single inflation report has more room to move expectations than it would immediately before a meeting, but also more time to be superseded by the August jobs report and the following CPI. The initial move in short-dated rates on release morning frequently overstates the durable repricing.
How to read the release in the first ninety seconds
Check core month over month first, against consensus. Then check shelter month over month, because it determines whether a soft core reading is broad or concentrated. Then check whether the previous month was revised. Only then look at the annual rates, which are the numbers the headlines will lead with and the least informative part of the release.
One further check that takes ten seconds and prevents most misreadings: compare the headline and core monthly figures. If headline is much weaker than core, the softness is energy and food, and the Fed will largely discount it.
Risks, uncertainty, and limits
Consensus forecasts move in the days before release and the figures cited here reflect the position in early August. CPI is subject to seasonal adjustment revisions, and the BLS updates seasonal factors annually in a way that can change recent monthly figures without any new underlying data.
This is a guide to reading the release, not a forecast of it. Nothing here predicts the July print or the market reaction to it, and single monthly inflation readings have a poor record as standalone signals of anything.