There are two consumers in the data right now, and they refuse to agree with each other. The University of Michigan's sentiment index fell to about 44.8 in May from 49.8 in April, parked just below the previous historic trough set in June 2022, per the university's surveys. Cost of living remains the dominant complaint, with 57% of respondents spontaneously citing high prices as the thing eroding their finances, and roughly two-thirds saying they've cut back on spending, as Retail Insight Network summarized. By the survey data, the American shopper is miserable.
By the receipts, the American shopper is fine. Retail sales have kept grinding higher — up 0.5% month over month and 5.2% year over year in the most recent Census reading — and the through-line of 2026 has been that people keep buying even as they tell pollsters they're tightening up. PYMNTS has called it bluntly: consumers keep spending even as confidence wavers. Feelings and behavior have decoupled, and retail is being asked to merchandise to both at once.
The gap between how people feel and what they do
The divergence is real but not mysterious. Friday's jobs report showed wages still rising 3.4% year over year, a hair ahead of cooling inflation, so households have nominal room to keep transacting. What's changed is the character of the spending. It's defensive — concentrated in essentials, replenishment, and the small "cheap thrills" that deliver a hit of discretionary pleasure without a discretionary-sized bill. That's the same current Endcap has tracked through trade-down earnings season, from Costco's double-digit comps to Five Below and Dollar General, and through our look at Memorial Day, when participation rose even as the average ticket fell sharply. People are still showing up. They're just spending selectively and trading down within the cart.
The two-speed consumer, quantified
Underneath the average sits a barbell. Higher-income households are still spending freely — often trading over to value formats like Costco not out of need but because the math is simply better — while lower-income and non-degree households, the ones Deloitte's consumer tracking shows are most exposed to gas and essentials inflation, have pulled back hardest. A retailer's exposure to this split now matters more than its read on the "average" consumer, because there is no average consumer to sell to.
Why the next three weeks decide it
The argument gets settled soon. Amazon pulled Prime Day forward to June 23–26, three weeks earlier than its traditional July slot, and Target and Walmart have stacked counter-events into the same week — a calendar shift Endcap flagged when Amazon confirmed the dates. That makes late June the earliest and biggest read on discretionary appetite all summer. If weak sentiment finally bites, it will show up in conversion and average order value across that window, and the "feelings don't equal behavior" trade falls apart. If spending holds through a deal event this large, the pattern survives into back-to-school and the people writing the gloomy sentiment numbers will keep getting overruled by the people swiping the cards. Retailers don't have to guess much longer — but they do have to be inventoried correctly for whichever consumer actually shows up.