Everyday · Money & Consumers · Food
Grocery inflation slowed. Your grocery bill did not go backward.
A slower rate of increase is still an increase. And the basket average conceals a spread wide enough that two households in the same shop are having different years.
Two things are true at once, and the gap between them is where most of the frustration about grocery prices lives.
Inflation slowed. And your bill is still higher than last year. Neither statement contradicts the other, because a falling rate of increase is not a falling price. Prices went up more slowly. They did not come back down.
$100 basket
+ 8% → $108
+ 2% → $110.16
Inflation fell from 8 per cent to 2 per cent — a dramatic improvement by any measure. The basket still costs more than it did in either earlier year. Disinflation is a change in the rate; deflation would be a change in the level, and that is not what happened.
This is not a semantic point. It is the difference between a statistic that describes the economy accurately and a statistic that describes a household’s experience of it. Both are real; only one is on the receipt.
The 2026 forecast, and the fact that it moved
USDA’s Economic Research Service publishes a Food Price Outlook and updates it monthly as new Consumer Price Index and Producer Price Index data arrive. For 2026 it has been revised upward across the year.
Early 2026 releases → food-at-home about +2.3% to +2.5%
July 2026 release → food-at-home +2.8%
July 2026 release → all food +3.2%, away-from-home +3.8%
The spread is not a contradiction between sources. It is the same series revised as data landed, and each figure was correct on the date it was issued. A citation without its release date is therefore not a citation.
Two structural points follow. The Outlook tracks the annual average change — all months of the year against all months of the prior year — so it is not the month-to-month number a shopper feels. And it carries a prediction interval, sometimes a wide one, which is the forecast’s own statement of how much it might be wrong by.
Every figure on this page is a projection published by a federal agency, not an outcome. It will be revised again. When the year closes, the realised figure is what happened; until then the Outlook is the best available estimate and nothing stronger.
The average hides the thing you actually noticed
An overall grocery figure near 3 per cent describes a basket. Almost nothing in the basket moved at that rate, and the categories pulling hardest in each direction are the ones shoppers comment on.
Beef and veal is the standout increase, forecast in the July release to rise by roughly 10 to 11 per cent. The reason is supply rather than anything at the till: US cattle inventories are at multi-decade lows after a cyclical contraction of the herd, and farm-level cattle prices have risen accordingly. That is a structural condition, not a seasonal one, and it does not resolve quickly — a herd takes years to rebuild.
Eggs run the other way, forecast to fall substantially as flock sizes recover from earlier disruption. Egg prices are the clearest recent example of the whole problem: they rose sharply, then fell sharply, and a household that changed its shopping during the rise did not automatically notice the reversal.
- Rising faster than the basket — beef and veal most of all, with other meats, fish and seafood, sugar and sweets, and non-alcoholic beverages also running ahead.
- Falling — eggs, by a wide margin in every 2026 release.
- Near the basket average — cereal and bakery products, fresh vegetables and most of the rest.
Which of those you experience depends entirely on what you buy. A household that eats beef several times a week and a household that does not are shopping in the same store under the same published inflation rate and having two different years.
Eating at home and eating out have diverged
One trend in the 2026 data is consistent across every release: food away from home is rising faster than food at home, forecast at about 3.8 per cent against 2.8 per cent in the July release. That pattern has held for several years now.
The mechanism is that a restaurant sells labour, rent and energy as well as food. Wage rates, commercial rent and utility costs all enter the price of a meal out in a way they do not enter the price of a bag of rice. When those costs rise, restaurant prices follow more closely than grocery prices do.
For a household deciding where to spend, that gap is the actionable part of the whole dataset — more so than the headline rate, because it is a difference between two options rather than a description of one.
What this page does not claim
It does not tell you what your own grocery bill will do. A national forecast is an average across the country and across a standard basket; it is not a prediction about one household, one store or one region, and California prices are not the national average.
It also does not treat the forecast as settled. Every number here is a projection from a specific monthly release, carries a prediction interval, and has already been revised at least once this year. Where the figures on this page differ from a figure published earlier, both were accurate on their release dates — which is precisely why the release date is stated alongside each one.
Read the source directly at the ERS Food Price Outlook summary findings, which is updated monthly.
Related on this site: restaurant service charges and tips · Money & Consumers