The consumer price index report is the most market-moving scheduled release on the economic calendar. It is published monthly at 8:30am Eastern, and the August 2026 edition is due on 11 September, days before a Federal Reserve decision that markets rate as a coin flip.
Most coverage reduces it to a single number. The report contains far more, and the detail frequently tells a different story from the headline. This tutorial walks through how to read it.
What the index measures
The index tracks the cost of a representative basket of goods and services bought by households. Hundreds of items are included, each weighted by its share of typical spending.
Housing takes the largest weight by a wide margin. Food, transport and medical care follow. Prices are collected monthly and the basket's total cost is compared with earlier periods.
The four headline figures
Every report leads with four numbers, and confusing them is the most common error.
- Headline, monthly. How much the whole basket changed since last month.
- Headline, annual. Change over twelve months. This was 3.4% in July 2026.
- Core, monthly. Excluding food and energy.
- Core, annual. Twelve-month change excluding food and energy. This was 2.5% in July.
Why core matters more
Removing food and energy sounds like ignoring the things people actually buy. There is a reason for it.
Food and energy prices are volatile and driven by factors monetary policy cannot influence - weather, harvests, and geopolitics. Crude oil rising toward $91 a barrel on Middle East tension is not something interest rates can address.
Core is a better measure of underlying, persistent inflation. That is why the Fed weights it more heavily, and why a hot headline with a soft core is a very different report from a hot core.
Monthly versus annual
This distinction matters more than most people realise.
The annual rate includes eleven months of history. If prices jumped sharply last autumn and have been flat since, the annual figure stays elevated purely because of that old jump. When that month drops out of the calculation, the rate falls without anything changing in the present.
The monthly change describes what is happening now. Professionals often annualise the last three months to get a current run rate.
A falling annual rate alongside rising monthly momentum is a genuinely concerning combination, and headlines routinely miss it.
The components worth checking
Shelter
The largest single component. It is measured with a long lag, because it reflects rents across all existing tenancies rather than new agreements. Cooling in the actual rental market takes many months to appear here.
This lag is the strongest argument the doves on the Fed have: disinflation may already be in the pipeline without being visible yet.
Services excluding shelter
Often the most closely watched line of all. This captures things like haircuts, repairs, insurance and dining out - mostly the cost of people's time.
Because it is driven by wages, it is the stickiest part of inflation and the hardest for policy to shift. If this is running hot, the hawkish case is strong.
Core goods
Physical products excluding food and energy. This is where tariff effects appear most directly. This category spent much of the previous decade in deflation, quietly offsetting higher services inflation. When that stopped, the services problem became fully visible.
Energy
Excluded from core but included in headline. Given the recent oil surge, expect this to contribute meaningfully to the August headline figure.
How to read it in two minutes
A workable sequence for the morning of the release:
- Core monthly first. Compare with the expected figure. This single comparison drives most of the initial reaction.
- Core annual second. Confirms direction of travel.
- Services excluding shelter. Is the sticky part improving?
- Shelter. Still decelerating?
- Headline. Mainly to understand how much energy contributed.
Note that the headline comes last. That is deliberate, and it is the reverse of how most coverage presents it.
Surprise is what matters
The reaction is not determined by the level but by the difference from expectations.
Inflation at 3.4% sounds high, but if that is exactly what was forecast, it produces no reaction at all - it is already reflected in every price. A tenth of a percentage point difference on core, in either direction, is frequently enough to shift rate expectations substantially.
Checking the consensus forecast beforehand is therefore not optional.
The rounding quirk
Headline figures are published rounded to one decimal place, but the underlying calculation carries more precision.
Markets sometimes react to the rounded number, then reverse moments later when the unrounded figure shows the result sat at the extreme edge of the rounding band. This produces reversals that look arbitrary but are simply better information arriving.
Why this month's report matters unusually
The August report is the last significant inflation reading before the Fed decides, and the committee is genuinely split - the July vote was 9 to 3, and hike odds have hovered near 50%.
This means the release is effectively deciding a central bank meeting as well as informing inflation expectations. Two events priced through one number, which is why the reaction may exceed a typical inflation day.
What the report does not tell you
A few honest limitations.
It describes the past, not the future. It describes an average household that may look nothing like you. And it is an estimate built from sampling and statistical adjustment, revised over time and imperfect by construction.
The figures are broadly honest - independent measures using entirely different data track them closely - but they are approximations, and any individual's experience can legitimately differ.
The summary
Read core before headline, monthly before annual, and services excluding shelter before almost anything else. Know the expected figures beforehand, because surprise rather than level drives the reaction.
Doing that puts you ahead of most commentary, which typically leads with the annual headline number - the figure containing the most old news and the least information about what happens next.
For how to handle the trading side of the release, see trading inflation day without getting chopped to pieces.
The other inflation measure
Worth knowing, because it causes confusion: the consumer price index is not the measure the Federal Reserve formally targets.
The Fed's preferred gauge is a different index produced by a separate agency, using a somewhat different basket and updating its weightings more frequently to reflect changing spending patterns.
The two usually tell a similar story but often differ by a few tenths of a percentage point, with the Fed's preferred measure typically running slightly lower.
The consumer price index still moves markets more, because it is published earlier and is more widely followed. But when officials say inflation is at a particular level, they may be quoting a different number from the headline you read.
Seasonal adjustment
Nearly all the figures quoted are seasonally adjusted, and it is worth knowing what that means.
Some price movements happen predictably at certain times of year - holiday travel, seasonal clothing, energy for heating. Statisticians remove these regular patterns so that genuine underlying changes are visible.
The adjustment is necessary but imperfect. When behaviour patterns shift, the historical seasonal factors can become inaccurate, producing figures that overstate or understate what actually happened in a given month.
This is one reason single-month readings deserve caution and multi-month averages are more reliable.
Revisions and why they get missed
Consumer price index figures are revised less than employment data, but seasonal factors are updated periodically, which can change historical readings.
These revisions arrive without fanfare and are rarely reported. Occasionally a period that appeared to show cooling inflation gets restated as having been firmer, quietly changing the picture that policy decisions were based on.
Nobody trades this. It is worth knowing simply because it explains why the historical inflation series you look at today may not match what people saw at the time.
Putting a single report in proportion
A closing note on interpretation. One month's figure is a noisy observation of a slow-moving process.
Underlying inflation does not lurch about from month to month. What lurches is the measurement, affected by sampling, seasonal adjustment and one-off events in individual categories.
This is why officials repeatedly say they are looking for a sustained trend rather than reacting to a single print, and why a trader who concludes the entire inflation picture has changed on one report is usually over-reading.
The market will nonetheless react violently to that single report, because positioning has to adjust immediately. Both things are true at once: the report matters enormously for prices this week and rather little for the economic reality it describes.
Where to find the actual release
It is worth knowing that the full report is published free of charge by the statistical agency, in considerably more detail than any news summary provides.
The document includes tables broken down by category, showing both monthly and annual changes for each. Anyone wanting to know what specifically drove a surprise can find it there within a couple of minutes, rather than waiting for commentary to explain it.
The summary tables near the front are the practical starting point. They contain the headline and core figures along with the main component groups, which covers everything discussed above.
Reading the source directly a few times is genuinely worthwhile, even if you never do it again. It builds a sense of the report's structure that makes subsequent summaries far easier to interpret.
Building your own expectations
Over time it becomes possible to form a rough view before each release, and doing so improves your reading of the result.
The inputs are not obscure. Energy prices are observable daily, so the energy contribution is largely knowable in advance. Rent trends are published by private data providers well before they appear in official shelter figures. Wage growth from the employment report gives a steer on services.
None of this makes you competitive with professional forecasters. What it does is convert the release from a number arriving out of nowhere into a check on a view you already held, which is a much better way to learn.
The habit also protects against a common error: treating every surprise as significant. If you understood beforehand that the energy contribution would be large because oil rose toward $91, a hot headline figure is not a surprise at all. It is the expected consequence of something you already knew.
This article is general information, not financial advice. Do your own research or speak to a licensed professional before making money decisions.
General information, not advice. This article is published to everyone who reads it and takes no account of your circumstances, so it is not a personal recommendation. Trader Suite is not authorised or regulated by the FCA. Trading and investing involve a substantial risk of loss, and you should seek independent advice before acting. Our articles are researched and drafted with AI assistance and reviewed before publishing. Full risk disclosure.
TraderSuite Team
TraderSuite builds indicators and automated strategies for NinjaTrader 8. Our articles are written by the team, researched and drafted with AI assistance, and reviewed before publishing.






