Amazon Made Whole Foods Cheap. Its Hourly Employees Are the Cost.

Amazon is betting on Whole Foods harder than it ever has, and Whole Foods is betting against its own hourly workers' union harder than it ever has, and those are the same fact viewed from two aisles.

Here's the collision. In January, Amazon said it would close all of its Amazon Fresh and Amazon Go stores and convert some of them into Whole Foods locations, effectively conceding that its own grocery brands couldn't win and that Whole Foods would carry the company's physical grocery ambitions. Meanwhile the first union in Whole Foods' history — at the Philadelphia Center City store — has been certified for more than a year and a half, and management still won't sit down to bargain a first contract, which is a choice that has a price. A labor story and a grocery-pricing story. Actually one story: the hourly employee is where the two ends of Whole Foods' economics meet.

The frequency engine wearing an apron

The weird thing about Amazon owning Whole Foods is that it owns the one business whose economics are the opposite of everything else Amazon does. Amazon runs on scale, software, and fees. Whole Foods runs on butchers, fishmongers, and line cooks paid hourly to operate around 540 stores that produced $5.79 billion in the second quarter — a 4% increase, and roughly 3% of Amazon's total quarterly sales of $200.6 billion. And grocery is a brutal business: net profit margins run between 1% and 3%, a band so thin that a single bad quarter of spoilage or wage pressure can erase the year.

Nobody in the strategy tells you to value Whole Foods on its own margin, because it isn't the business. Its job is to be a "frequency engine" — the reason a Prime household keeps buying perishables weekly, which feeds subscription renewal, advertising, and the delivery network that carries the real profits. Amazon's own wording gives this away: when it reports its paid-units growth, the metric explicitly excludes Whole Foods, because Whole Foods isn't a unit-sales story, it's a habit story. Grocery is the wedge. AWS, ads, and logistics are what Amazon actually makes money on.

Cheap kale, paid for somewhere

That explains why Amazon can push Whole Foods to shed its "Whole Paycheck" reputation without flinching. The company has doubled its promotions, given Prime members an extra 10% off sale items, and leaned on the 365 private label — now over 4,000 products, with private-label sales up 60% since 2019 — all while narrowing the price gap with Walmart and Kroger. Whole Foods CEO Jason Buechel, who now also runs all of Amazon's worldwide grocery stores, told Yahoo Finance the chain has been named the lowest-priced U.S. retailer every year since 2017. That's a strange boast for the store whose nickname has been "Whole Paycheck" since the 1990s, but it makes sense once you see who's underwriting the cheaper prices: the high-margin machine next door.

This is, in effect, a cross-subsidy. Amazon can afford for Whole Foods to compete on price because the resulting frequency pays off inside Prime, ads, and the delivery network — categories where Amazon's margins live. The grocery business doesn't need to be a great standalone business; it needs to be a great habit.

Where the price cut actually lands

Now the other end of the story. In a thin-margin, hourly-labor business, the flexible schedule is the shock absorber. When you compete on price, you squeeze hours and scheduling and wages, because labor is the largest cost you actually control. At Whole Foods that pressure has a history behind it: the chain's higher wage bill — for in-store cafes, prepared foods, and cheesemongers — is itself a reason its prices were ever high in the first place.

That is the structural setting for the first union in Amazon-owned Whole Foods history. In January 2025, workers at the Philadelphia Center City store voted 130–100 to join UFCW Local 1776. Whole Foods challenged the result with a string of objections, and in June 2026 the National Labor Relations Board denied review, saying the company raised "no substantial issues warranting review." Whole Foods said it "strongly disagrees"; the union said there is "no reason at this point" Amazon shouldn't come to the table. It still hasn't.

I want to be careful about what one store of roughly 540 is worth, because financially it is basically nothing. A hundred and thirty unionized votes can't move a cost line in a $2.7 trillion company whose real story this quarter is AWS growing 37% and advertising up 26%. The wrong way to read this is as a Whole Foods labor apocalypse. The right way is as the first organized claim on the exact cushion — the schedule — that makes Whole Foods' cheap-price strategy possible. A union is an instrument that converts "the manager can flex your hours however it likes" into "that flexibility has to be bargained and paid for." One store doesn't change the economics. A pattern of them would, because nothing puts a floor under thin grocery margins faster than locking in hour guarantees and wage steps across a chain that has decided to win on price.

So for an Amazon shareholder, the hourly Whole Foods employee is a footnote with a message. Amazon is deliberately running a physical grocery business it barely needs to profit from, in a price war, in a category whose margins are 1%–3%, against hourly workers who have just proven they can organize a store and win at the labor board. Amazon has decided that grocery is the wedge it will subsidize to drive frequency — and the hourly schedule is where the remaining cost of that decision lands. It can program the pricing, the Dash Carts, and the checkout. It cannot yet program the schedule, and the union is the first instrument trying to put a price on it. If that instrument stays confined to one store in Philadelphia, it's a rounding error. If it spreads, it's the one lining in Whole Foods' economics that isn't so easily subsidized away.