On this page
The short version: The Consumer Price Index (CPI) comes out once a month from the U.S. Bureau of Labor Statistics (BLS), at 8:30 a.m. Eastern Time, usually between the 10th and the 15th of the month. Each release carries two headline numbers that do not measure the same thing: the seasonally adjusted monthly change and the unadjusted 12-month change. Headline CPI covers everything; core CPI leaves out food and energy. Shelter is the largest single component — roughly a third of the index — and it moves slowly, so it often decides whether the month looks hot or cool. This guide explains how to read the release, what the Federal Reserve actually watches, and which numbers are noise.
What CPI measures
The CPI tracks the average change over time in prices paid by urban consumers for a basket of goods and services: food, energy, shelter, apparel, transportation, medical care, and more. The figure most people quote is CPI-U, the index for all urban consumers, reported as a month-over-month change and a year-over-year change. The BLS publishes the release on bls.gov the moment it comes out, and the full tables post at the same time. A practical habit: read the BLS release itself before reading anyone’s commentary on it. The method for checking a claim against its original source is the same one in our guide on how to verify the news.
Headline vs. core: two different questions
Headline CPI is the all-items index — everything in the basket. Core CPI is all items less food and energy. The two answer different questions. Headline tells you what households actually paid; core tells you what the underlying trend looks like once the most volatile categories are stripped out. Food and energy prices swing on weather, harvests, and geopolitics, so a headline surprise driven by gasoline can reverse the next month. A core surprise driven by services — rent, medical care, insurance — is harder to dismiss, because services prices move slowly and reflect wage pressure. When the two disagree, markets pay more attention to core.
In the October 2026 release (published November 10, 2026), headline CPI rose 0.4% on the month and core CPI rose 0.3%, the same pace as in September and August. Shelter rose 0.3% and accounted for more than 30% of the monthly all-items increase. Used cars and trucks fell 1.2%, while recreation rose 0.8% and medical care rose 0.3%. Those details matter: a 0.4% headline driven by shelter reads differently from a 0.4% driven by gasoline.
The monthly change and the 12-month change are not the same thing
Each release leads with two numbers. The monthly change is seasonally adjusted — the BLS quotes it first. The 12-month change is computed on the unadjusted index. The 12-month rate does not measure the month: it settles twelve months at once. A high 12-month rate can persist for months after monthly prints cool down, and it can fall sharply when a hot month from a year ago drops out of the window — the base effect. That is why a single month rarely changes the story, and why three consecutive months in the same direction matter more than one.
A real example of how the record can break: the BLS did not collect October 2025 price data because of a lapse in federal funding, so it could not publish October 2025 indexes for most series — and therefore could not publish 12-month percent changes for October 2026. The October 2026 release carried monthly changes only. When a 12-month figure is missing, say so; do not compute one from a different index.
Why shelter decides the month
Shelter — rent and owners’ equivalent rent — is the largest weight in the index, roughly a third of it, and it is slow-moving. A 0.1 percentage-point surprise in monthly core CPI is roughly a 1.2-point difference in annualized terms, which is why seemingly small misses can move rate markets. The components to check after the headline: shelter, core services excluding shelter (sometimes called “supercore,” an informal but widely watched gauge of wage-driven inflation), core goods (vehicles, apparel, household goods — sensitive to supply chains and tariffs), and energy and food (volatile; they push headline away from core). If shelter is the whole story, the print is sticky; if energy is the whole story, it may not last.
What the Fed watches — and what it does not target
The Federal Reserve’s inflation target is 2% on the PCE price index, not the CPI. CPI comes out first each month, so it is the first comprehensive read on inflation — but it is not the Fed’s target measure. CPI and PCE differ in weights and scope (PCE covers a broader basket and uses different formulas), so the two can diverge by several tenths of a point. Read the Fed’s decision the same way you read the CPI: the statement, the dot plot, and the press conference each add a layer. Our guide on how to read a Fed rate decision walks through that document, and the Fed’s own rule proposals — with comment deadlines and effective dates — are covered in Federal Reserve rule proposals.
A five-minute routine for release morning
1. Check the time: 8:30 a.m. ET, on the BLS schedule (the next release is Thursday, December 10, 2026, covering November 2026 data). 2. Read the monthly change first (seasonally adjusted), then the 12-month change (unadjusted) — and note if the 12-month figure is missing. 3. Read core, not just headline. 4. Find what drove the month: shelter, energy, or something else. 5. Compare against consensus expectations, not against last month — markets move on the surprise. 6. Wait for the details: the BLS tables show which categories moved, and the jobs report the same week adds the labor-market half of the picture. Our guide on how to read the jobs report covers nonfarm payrolls and unemployment the same way. Company claims about prices and margins need the same treatment — check the filing, not the press release, as in A Company Says “Record Quarter”.
Common mistakes
Mistake 1: quoting the 12-month rate as if it described the month. It describes twelve months. A 3% annual rate with a 0.1% monthly print is a cooling economy, not a hot one.
Mistake 2: treating a gasoline-driven headline as a trend. Energy reverses; services persist.
Mistake 3: annualizing one month. Multiplying a monthly change by 12 exaggerates noise; three months of data annualize more honestly.
Mistake 4: ignoring what is missing. If the BLS could not publish a year-ago index, the 12-month change does not exist — do not invent one.
Figures in this guide are from the BLS CPI news release for October 2026 (USDL-26-1685, published November 10, 2026) and the BLS release schedule; check bls.gov for the latest release. This article is for information only and is not investment advice.
