Market Analysis • November 17, 2026
October Retail Sales Rose 0.2%—But the Estimate Couldn’t Rule Out Zero
The November 17, 2026 retail-sales release delivered a headline that sounds firmer than the evidence: October sales increased 0.2%, but the Census Bureau’s 90% confidence interval includes zero. In other words, the estimate points to a small nominal gain, yet the data cannot distinguish that change from no monthly growth. The release states this plainly—and also notes that sales are not adjusted for price changes.
Here’s what the numbers reveal:
- Total retail and food-services sales growth slowed from 0.8% in September to 0.2% in October; the October estimate is not statistically distinguishable from zero.
- September’s headline growth rate was unrevised at 0.8%. That does not tell us whether the September sales level or individual categories were revised.
- Auto and parts dealers’ sales fell to $137.0 billion in October, while gasoline-station and nonstore sales increased.
- Sales were 6.8% higher than a year earlier, but that is a comparison of nominal dollars—not proof that consumers bought 6.8% more.
- The release supports caution about October’s momentum, not a conclusion that households are under financial stress.
A Softer Headline, Without a Downward September Revision
The monthly point estimates show a clear deceleration in total retail and food-services sales. But the size of that slowdown needs careful handling: October’s estimate carries enough sampling uncertainty that the reported 0.2% should not be treated as definitive evidence of growth—or decline.
| Monthly sales growth | August 2026 | September 2026 | October 2026 |
|---|---|---|---|
| Total retail and food services | 1.5% | 0.8% | 0.2%* |
| Excluding motor vehicles and parts | 1.6% | 1.1% | 0.6% |
| Excluding motor vehicles, parts and gasoline stations | 1.4% | 0.5% | 0.5% |
October’s 90% confidence interval includes zero. The release says comparisons across industries have not been tested for statistical significance.
September’s headline monthly change remained 0.8% in the November 17 release. That matters: October’s weaker point estimate did not arrive alongside a downward revision to September’s headline growth rate. Still, the available earlier-release material does not provide enough detail to determine whether September’s sales level or category estimates changed.
The measures excluding autos, and excluding both autos and gasoline stations, tell a more resilient—but narrower—story. Growth excluding autos was 0.6% in October, while growth excluding autos and gasoline stations held at 0.5%. These figures suggest that the headline slowdown was not mirrored in every aggregate. They do not establish that the differences between measures, or between months, are statistically significant.
Autos Slipped as Gasoline and Nonstore Sales Climbed
October’s category estimates show different currents beneath the modest headline. Auto and parts dealers recorded the largest visible decline among the listed categories, while gasoline stations and nonstore retailers moved higher.
| Category | September 2026 | October 2026 | October reading |
|---|---|---|---|
| Motor vehicle and parts dealers | $139.2B | $137.0B | Declined |
| Gasoline stations | $67.6B | $68.6B | Increased |
| Nonstore retailers | $126.7B | $128.2B | Increased |
| Furniture and home furnishings | $10.9B | $10.7B | Declined |
| Food and beverage stores | $85.2B | $85.1B | Edged down |
| Clothing and accessories | $26.6B | $26.4B | Declined |
| General merchandise | $81.7B | $81.8B | Little changed |
| Food services and drinking places | $92.7B | $92.9B | Increased |
All figures are seasonally adjusted sales in billions of dollars.
Auto and parts dealers’ sales fell about 1.6% from September to October, while gasoline-station sales rose about 1.5% and nonstore sales about 1.2%. Those are calculations from the reported dollar levels, not separate significance tests. The release cautions that cross-industry comparisons have not been tested for significance, so the category movements are best read as the shape of the point estimates—not a ranking of statistically confirmed winners and losers.
Nor should “nonstore” be used as a synonym for online shopping alone. The supplied figures do not isolate a shift from physical stores to the internet, explain the auto decline, or identify whether inventory changes played a role.
Nominal Dollars Leave the Real-Spending Question Open
The year-over-year headline is striking: October sales were 6.8% above October 2025. But both that figure and the monthly estimate measure dollars spent. The Census release seasonally adjusts sales but does not adjust them for price changes.
That distinction leaves the central household-demand question unanswered. If prices rose, some portion of higher dollar sales could reflect more expensive purchases rather than greater quantities. The release provides neither a suitable inflation measure nor the gasoline prices and volumes needed to separate price effects from fuel purchases. It therefore cannot establish whether real retail spending grew in October, or by how much.
The data also do not settle broader claims about consumer strain. The supplied release and historical excerpts contain no measures of household income, confidence, credit-card debt or borrowing. A weak monthly estimate is a reason to watch the next readings—not evidence, by itself, that consumers are running out of room.
What to Watch in the Next Release
The next assessment should focus on whether October’s softness persists, and whether revisions alter the picture.
- Total monthly sales: A subsequent estimate clearly above zero would strengthen the case that October marked a pause rather than a flat month. Another estimate whose confidence interval includes zero would leave the signal inconclusive.
- Autos versus the rest of retail: Watch whether motor-vehicle sales stabilize and whether the ex-auto measures continue to outpace the headline. The current release does not establish that those gaps are statistically meaningful.
- Prices and real activity: Pair retail dollars with price data before drawing conclusions about purchasing volumes. Gasoline sales particularly need price and volume context.
- Revisions and GDP-related detail: The provided material does not include enough earlier component data to evaluate revisions to the GDP-relevant control group or calculate an effect on quarterly consumption. Nor does it provide a forecast comparison, so the size of any surprise cannot be assessed.
The release describes an advance estimate based on a sample of about 4,800 firms, weighted and benchmarked to represent more than three million businesses. That method gives the figures broad coverage, but sampling and response uncertainty remain part of the story—especially when the headline change is small. The report also notes that estimates are seasonally adjusted and that some details may be suppressed when data quality or sampling variability is inadequate.
October’s record is therefore narrower than either a clean growth story or a consumer-stress alarm: nominal sales edged higher, the estimate could not rule out zero, and autos weakened while several other categories rose. The next release must show whether that softness persists—and price data will be needed to tell what it means for real spending.
Produced with AI assistance from public data releases. For information only, not investment advice. How we work