The consensus read on the US retail sales August surge is straightforward: Americans are spending freely and the consumer is fine. The Census Bureau data does support a strong headline number. But the composition of that spending, and where structural share is quietly shifting, complicates the tidy narrative.
Overall retail sales rose 1.2% in August from July to $773 billion, seasonally adjusted, and were up 6.0% year-over-year. Strip out gas stations, and the picture is only marginally softer: sales still rose 1.1% month-on-month and 4.9% year-over-year. On the surface, resilient. Below the surface, a significant amount of that headline strength is a statistical echo.
The Prime Day Distortion the Headline Number Buries
Part of the August rebound reflects a mechanical recapture of July’s weakness. Amazon Prime Day shifted into June this year from July last year, which pulled the spending spike forward, depressed July, and made August’s comparative look better than it is. The three-month moving averages, which smooth that distortion, tell the more honest version: ecommerce’s three-month average rose 0.6% month-on-month and 10% year-over-year, still strong, but considerably less dramatic than the raw August print of 2.6%.
None of this makes the underlying trend bad. It simply means the US retail sales August surge is partly catch-up, and investors pricing in a sudden re-acceleration of consumer momentum may be getting ahead of themselves.
Ecommerce’s Structural Dominance Is the Real Story
Ecommerce is now, on a 12-month basis, the largest single retail category, with an 18% share of total retail sales, ahead of motor vehicle dealers which have slipped to 17%. That is a structural shift, not a cyclical blip, and it accelerates with each reporting period.
The enrichment data suggests the move is broader than Census Bureau monthly figures alone capture. According to Publicis Commerce, e-commerce grew 7.5% in the first half of 2024 and now represents 22% of core retail sales for the year, a materially higher share than the monthly Census data implies. Meanwhile, the U.S. Census Bureau‘s own Q2 2024 e-commerce report puts the channel’s share at 16.0% of total sales in the second quarter, a figure that varies with how “total sales” is defined but confirms the directional dominance. Brick-and-mortar retail is not collapsing, but it is conceding ground in categories it once considered stable, including groceries.
Walmart and Costco are worth naming here precisely because the report places them in two separate categories simultaneously: their physical stores land in “general merchandise,” while their online operations count toward ecommerce. The practical effect is that both businesses are growing in the data twice over, while traditional grocery retailers and specialist retailers are counted once and declining in share. That accounting quirk flatters the general merchandise and ecommerce lines at the expense of categories that cannot split themselves the same way.
Restaurants and bars contributed a 1.2% monthly increase to $105 billion, seasonally adjusted, up 5.8% year-over-year. As a barometer of genuine discretionary confidence rather than obligatory spending, that number carries weight. Food services and drinking places are now the third-largest retail category, with a 12% share, and the gap between restaurant spending and grocery spending has widened considerably since 2019, when restaurant spending first exceeded food and beverage store sales.
Gas stations, predictably, moved with prices rather than volumes. A 3.1% monthly jump to $62 billion, and 21% year-over-year, reflects the gasoline price spike in August, not a surge in miles driven or consumer exuberance. It inflates the aggregate and should be discounted accordingly by anyone trying to read the underlying demand signal.
The miscellaneous stores category, up 1.8% month-on-month and 14.0% year-over-year to $17 billion, deserves more attention than it typically receives. Cannabis stores sit within this category and are the primary explanation for what the report describes as a multi-year massive surge. The trajectory has external support: Business of Cannabis projects US cannabis sales will exceed $32 billion in 2024, record growth that retail aggregates quietly absorb without the category ever receiving a dedicated line. For anyone tracking the US retail sales August surge by sector, this omission is worth noting.
The consumer is spending. The question the headline number does not answer is whether the composition of that spending, heavy in experiences, ecommerce and cannabis, light in big-ticket durables and traditional grocery, reflects confidence or a specific kind of selective resilience that is easier to sustain in some rate environments than others. Motor vehicle dealers, still a $130 billion monthly category, grew just 1.7% year-over-year, a pace that sits well below the aggregate and below ecommerce by a wide margin.
