I came back from a year in Nairobi and walked into an American grocery store and did not recognize the prices.
That is not a metaphor. I stood in the cereal aisle holding a box I had bought dozens of times before and genuinely could not tell whether the number on the shelf tag was a mistake. It was not a mistake. That was just what cereal cost now.
The estrangement had a measurable backdrop. A Government Accountability Office analysis of Bureau of Labor Statistics data found that food prices rose 32.4 percent between December 2019 and December 2024. Memory had not become unreliable. The price environment had moved. ([files.gao.gov](https://files.gao.gov/reports/GAO-25-107451/index.html?utm_source=openai))
What interests me, however, is not another complaint about inflation. It is what the grocery store, as an institution, tells you while you shop—and whether you know how to read it.
The Shelf
A year away from American consumer infrastructure clarifies things that proximity obscures. In Nairobi, as I experienced it, household limits announced themselves quickly. You bought what you could afford. The distance between price and consequence was short.
The American grocery store speaks in a more elaborate register. There seem to be seventeen versions of the same yogurt. There is a budget tier, a middle tier, a premium tier and a store brand positioned near the bottom with what can only be described as a dignity problem.
The store calls this choice. From the aisle, it also looks like an income map.
The tiers allow several households to encounter the same category without encountering the same product. One shopper buys the familiar national brand. Another waits for the digital coupon. Another moves to the store brand. Another buys the premium version whose principal message is that its buyer need not perform these calculations.
This is abundance, certainly. It is also segmentation: a way of keeping households with different budgets inside the same store and, if possible, inside the same category.
The Store Brand
When I left, private-label products were already prominent. When I returned, they seemed to occupy more of the store. Industry sales data support the direction of that impression, though not the melodrama: between 2019 and 2024, store brands’ share rose from 19.1 to 20.7 percent of dollar sales and from 21.6 to 23.2 percent of unit sales. ([plma.com](https://www.plma.com/article/data-points-private-label-share-growth?utm_source=openai))
It would be too simple to call this only a story of consumer pressure. Store brands also grow because shoppers like them, because their quality has improved and because retailers develop them deliberately. But their economic function is hard to miss. They give the store somewhere to receive a shopper who is trading down.
That architecture is explicit rather than secret. Kroger, for example, describes its private-label portfolio as three tiers: premium, mainstream and value. It also reports that those products represented more than $32 billion of its fiscal 2024 sales. ([sec.gov](https://www.sec.gov/Archives/edgar/data/56873/000155837025004267/kr-20250201x10k.htm?utm_source=openai))
The store brand, then, is not merely the sad little box chosen after the national brand becomes unaffordable. It is a strategic asset. It can preserve the basket, retain the customer and turn an act of household retrenchment into an ordinary merchandising event.
The consumer still makes a sensible choice. The institution has simply prepared the choices in advance.
The Cereal Box
Cereal stopped me because it is so familiar. The cardboard box, the bright lettering, the quantity printed near the bottom: these are objects designed to be recognized without much thought. When the price becomes strange, the whole object becomes newly visible.
But the shelf tag cannot, by itself, tell me why the number changed. Food prices are shaped by production costs, labor, transport, energy, weather, disease, trade, competition, promotions and corporate pricing decisions. Government investigators have repeatedly cautioned that the contribution of any one factor can be difficult to isolate. ([gao.gov](https://www.gao.gov/products/gao-23-105846?utm_source=openai))
Shrinkflation is part of the cereal story, but it is not the whole story. GAO found that product downsizing made only a small contribution to inflation overall between 2019 and 2024, although its estimated effect was larger for breakfast cereal, contributing 1.6 percentage points to that category’s price increase. ([files.gao.gov](https://files.gao.gov/reports/GAO-25-107451/index.html?utm_source=openai))
That distinction matters. A smaller box is evidence of a particular pricing tactic. A surprising shelf tag is not proof of a single cause, much less proof of a conspiracy. What it does record is an outcome: after costs, contracts, promotions, competition and pricing strategy have finished their work, this is the number presented to the household.
The Second Store
The visible store sells groceries. Inside it now sits another store that sells access to the shopper.
Loyalty accounts record purchases. Apps deliver individualized promotions. Search results and digital circulars contain paid placement. Screens at checkout sell attention. Shelf assortments can be adjusted using transaction data. The grocery business is therefore not exhausted by the difference between the wholesale and retail price of a tomato or a box of cereal.
Major chains describe this second business plainly. Kroger says the traffic and data produced by its retail operation support high-margin businesses including analytics and third-party media. Albertsons says it uses customer and transaction data to target promotions and optimize shelf space, while its media business uses the company’s digital platforms and proprietary data. ([sec.gov](https://www.sec.gov/Archives/edgar/data/56873/000110465926037723/kr-20260131x10k.htm?utm_source=openai))
You are still the shopper. You are also part of the audience, the dataset and the sales proposition offered to suppliers.
This does not mean that production costs no longer matter or that every price is engineered from a secret psychological profile. It means that the modern grocery store has more ways to earn from a shopping trip than the shelf margin alone. The shelf tag belongs to that larger system.
The Negotiation Before the Aisle
The price of cereal is formed before I arrive. Producers negotiate with retailers. Retailers decide assortment, placement and promotion. Both sides study demand. The shopper participates by buying, refusing, substituting or leaving—but does not participate in the negotiations that created the available choices.
The shopper is not at the table. The shopper’s anticipated behavior is an input to the discussion.
The pandemic made the unequal leverage inside this system unusually visible. In a 2024 staff report, the Federal Trade Commission found that larger firms were often better able than smaller rivals to secure supplies during shortages. It also found that food-and-beverage retailers’ revenues relative to total costs remained above their pre-pandemic peak through the first three quarters of 2023. The FTC cautiously concluded that some firms appeared to have used rising costs as an opportunity to increase prices and profits. That finding does not establish that every price increase was excessive, but it does rule out the comforting idea that higher shelf prices merely copied higher costs in every case. ([ftc.gov](https://www.ftc.gov/news-events/news/press-releases/2024/03/ftc-releases-report-grocery-supply-chain-disruptions?utm_source=openai))
Inflation is therefore not only the movement of costs through a pipe. It is also a test of who can pass a cost forward, who can protect a margin and who must alter a household budget.
The Household
The same shelf increase does not produce the same consequence in every home.
In 2024, households in the lowest income quintile spent an average of 33 percent of their before-tax income on food. Households in the highest quintile spent 6.4 percent. These figures do not prove that lower-income households saw higher shelf tags. They show why an identical increase can impose a radically different burden. ([ers.usda.gov](https://www.ers.usda.gov/data-products/chart-gallery/58372?utm_source=openai))
A household with financial slack can preserve its preferences. A household without it must alter the basket. It can change brands, reduce quantities, abandon a category or remove something else from the budget.
If you already buy the least expensive acceptable version, the store contains fewer painless substitutions. Below the private-label floor is not another attractively designed tier. It is less of the item, a different item or no item.
This is not rhetorical poverty. USDA estimated that 13.7 percent of American households experienced food insecurity at some point in 2024, including 5.4 percent experiencing very low food security, in which eating patterns were disrupted or intake was reduced because resources were limited. ([ers.usda.gov](https://ers.usda.gov/publications/113622?utm_source=openai))
Moderate-income households often do what the store enables them to do: move from the national brand to the store brand, activate the coupon, buy the family size and congratulate themselves on being disciplined. They are being disciplined. They are also responding to an architecture built to retain their spending as their circumstances change.
There is no shame in this. The humor of the American grocery store is that it can turn austerity into a lifestyle decision. The package changes color, acquires the word value, and allows everyone involved to proceed without mentioning what happened.
The Institutional Signal
Institutional stress tends to move along gradients of leverage. Actors able to pass a burden onward usually try to do so; the burden becomes hardest to move when it reaches the household with the fewest alternatives.
This is a principle, not a claim that every firm wins or every household responds identically. Grocery retailers can be squeezed by suppliers. Small brands can be squeezed by retailers. Workers, farmers and distributors can absorb costs that never become visible on a shelf tag. The point is not that one actor controls the entire chain. The point is that the ability to refuse, substitute or renegotiate is unevenly distributed along it.
The grocery store makes that inequality unusually legible. The shelf tag is not a neutral fact about the world, but neither is it fiction. It is the residue of several negotiations—over costs, wages, supply, promotion, placement, data, margin and consumer tolerance—compressed into a number small enough to print beside a box.
Coming back from Nairobi did not make me angry at the American grocery store, exactly. It made me curious about what it was saying. What Nairobi had given me was not a superior economy but a different kind of legibility. Returning made the familiar machinery look strange enough to inspect.
Standing in the cereal aisle, I could not prove from one box who had gained, who had lost or which cost had moved first. I could see something more modest and more durable: the cost was real, its causes were plural, and its burden would depend heavily on how many alternatives a household still possessed.
That is what the store records every week. Not simply supply and demand, but the household meeting an institution that has already studied its likely behavior, arranged a ladder of substitutions and placed the final number on the shelf.