A smaller September reading for prices excluding food and energy does not mean food or fuel got cheaper. In the Personal Income and Outlays release issued at 8:30 a.m. EDT on October 29, 2026, the U.S. Bureau of Economic Analysis said the PCE price index rose 0.5 percent from August and 3.6 percent from September 2025. Excluding food and energy, it rose 0.3 percent on the month and 3.1 percent on the year. Both pairs are increases. The lower pair is a narrower index, not a finding that groceries, gasoline, or other energy products fell.
These figures describe September 2026. They were checked on November 5, 2026, a week after publication, so they are not a new November price report. BEA says the next release, scheduled for November 25, 2026, will supersede the October 29 estimates.
Two indexes, not a food or fuel receipt
PCE means personal consumption expenditures, the national-accounts record of household spending on goods and services. The PCE price index follows prices across that broad basket. BEA also reports the index after removing food and energy. Analysts often call the second measure core PCE. The label is a convention. It does not mean the excluded items are optional or stable for a household budget.
In September the indexes did not match. On the month, 0.5 percent is larger than 0.3 percent. On the year, 3.6 percent is larger than 3.1 percent. Subtracting the published rates, and labeling the result as arithmetic rather than a BEA component, leaves gaps of 0.2 percentage point and 0.5 percentage point. Those gaps are differences between two indexes. They are not estimates of the change in food prices, energy prices, or fuel prices.
Weights are why the subtraction fails. Food and energy are only part of the full basket. If their prices move differently from other prices, they pull the full index away from the narrower one. The size of that pull depends on their price changes and on their spending shares. This release prints neither the shares nor the separate price changes. Treating the 0.5-point annual gap as food-and-fuel inflation uses a calculation the table does not support.
The category name matters too. BEA says “excluding food and energy,” not “excluding fuel.” Energy in this index is broader than gasoline or diesel on one receipt. The supplied release does not list gasoline, heating fuel, electricity, or a grocery line. It cannot show which of those prices rose, which were flat, or which fell.
Smaller than the headline, not a decline
“Smaller” here means smaller than the full PCE price index, not smaller than zero. A 0.3 percent monthly increase is still an increase. A 3.1 percent rise from a year earlier is still a rise. Neither figure says prices inside the core basket fell back to an earlier level.
The August column makes the monthly comparison sharper. BEA’s table shows both the PCE price index and the excluding-food-and-energy index at 0.3 percent in August. In September the full index moved to 0.5 percent, while the narrower index stayed at 0.3 percent. Core’s monthly rate did not slow relative to the August rate printed here. The headline sped up, and the indexes separated. Hearing only that core was lower can hide that unchanged 0.3 percent core rate.
The annual rates answer another question. They compare September 2026 with September 2025, not September with August. A latest-month acceleration can sit inside a year-over-year rate that also reflects the earlier eleven months. This text gives September’s annual rates only. It does not give August’s annual price rates, so it supports no claim that the annual rates themselves accelerated.
Do not turn the monthly rate into the annual rate by multiplying. Repeating 0.5 percent for twelve months is not the published 3.6 percent change from a year earlier; the annual figure reflects the actual path. The same distinction separates 0.3 percent from 3.1 percent.
Exclusion is a measurement choice
Leaving food and energy out asks how fast the rest of the basket changed. A core index can make that other pace easier to inspect. The choice is about measurement, not relief at a store or a pump.
The September pattern does not show a drop in the excluded items. If those prices had fallen enough, given their weights, the full index could have risen more slowly than the core index. It rose faster on both horizons instead. That is consistent with food and energy adding to the full index, but the release never splits the gap into food versus energy.
Assume two different indexes start at 100, an illustration, not a BEA price level. A 3.6 percent rise reaches 103.60. A 3.1 percent rise reaches 103.10. The 0.50 difference is between those constructed levels. It is not 50 cents off a grocery trip and not a per-gallon change.
The national indexes are not a household bill. Heavy gasoline spending can diverge from either measure, and light energy spending can look closer to the core basket only by coincidence. The release does not say a particular bill rose 3.6, 3.1, 0.5, or 0.3 percent.
What to watch on November 25
Income and spending lines in the same release are not price quotes. Current-dollar PCE rose 0.5 percent, while real PCE was less than 0.1 percent in the text and 0.0 in the rounded table. Those quantity and dollar figures do not show a food or fuel price.
BEA’s next Personal Income and Outlays release is scheduled for November 25, 2026, at 8:30 a.m. EST, and it will supersede the October 29 estimates while adding October. Watch whether September’s full index stays above the excluding-food-and-energy index, and whether October’s two rates move together or split. Separate food and energy prices are not in this release. Until they are printed, the smaller core reading means only that the narrower basket rose less than the full basket, not that food or fuel got cheaper.