My wife has made the same casserole for thirty years. Two cans, one dish, no measuring.
Last month it came out wrong, and the fault was not the cook. It was the cans.
A study out of the University of Massachusetts this January put a number on it. Across the packaged food aisle, the average item lost 14.6% of its size from 2012 to 2019.
Fourteen point six percent. Hold that one. The researchers calculate it hid close to four percentage points of food inflation. The shelf tag never moved, and the tag is what gets measured.
3.0 pts
Of paper-product inflation came from size cuts alone, on the government’s own count
1.6 pts
The same figure for breakfast cereal
0.01%
The effect on national inflation per year, all goods and services together
Look at that third number. It is the honest one, and it cuts against the story. Across everything Americans buy, package shrinking moves national inflation by one hundredth of a percent a year. The statisticians are hiding nothing.
The reframe is the whole point today. Nobody lives in the average of everything. A retired household buys the same eleven or twelve items every week, and those items sit where the cutting is concentrated.
Paper goods, cereal, snacks, detergent. That is the basket of somebody who cooks at home and shops once a week.
The individual cuts are small on purpose. One chocolate bag went from 6.38 ounces to 5.32 at the same price. Take fifteen percent one year and ten three years later, and a quarter of the product is gone unnoticed.
Which is the same trick performed on a different counter. A number that holds still while the thing behind it gets smaller. Washington decides its version tomorrow afternoon. The market leans hard toward a rise, near 87 to 90 percent, with the other branch open until two.
One shrinking box, read down the list.
And the business that wins this one does not manufacture anything at all.
The honest limit is in the data and I will not dress it up. Downsizing is genuinely rare. One study of scanner records found roughly three thousand cases across twelve years. The bureau’s own count put the incidence under half a percent.
So anybody selling you shrinkflation as the reason your money feels short is overselling it. It is one mechanism among several, and it happens to be the visible one.
The arithmetic
The 14.6% average decline in packaged food size from 2012 to 2019, and the finding that it hid close to four percentage points of measured food inflation, come from research led at the University of Massachusetts Amherst, published in the International Journal of Industrial Organization in January 2026. The category contributions, 3.0 percentage points for household paper products and 1.6 for cereal, are from Government Accountability Office report GAO-25-107451, July 2025, which also found fewer than 5% of items downsized. The 0.01% a year all-items estimate and the 2,426 instances identified from 2015 to 2021 are the Bureau of Labor Statistics’ own; the roughly 3,259 scanner-data cases from 2006 to 2018 come from the same literature. The product examples are documented 2026 cases, and the compounding illustration is arithmetic. The Federal Open Market Committee announces at 2:00 PM Eastern on Wednesday 16 September. The roughly 87 to 90 percent probability of a rise is current market pricing, not a forecast by this letter. The read on who profits is Boomers Trade’s own.
Read the unit price, the small figure on the left edge of the shelf tag, in cents per ounce. It is printed by law and cannot be shrunk. Ours now says the casserole needs three cans. That is the whole story, standing in a kitchen, with a recipe card unchanged since 1994.
Andrew
Boomers Trade is written by someone getting older right alongside you, and watching who profits from it.
|