Retail Sales “Up” 0.5%—But the 90% Confidence Interval Says It Could Be Zero

On May 14, 2026, the official retail and food services release led with a clean +0.5% month-over-month for April. The fine print immediately undercut it: the 90% confidence interval includes zero, meaning there’s “insufficient statistical evidence” that sales actually rose. That’s not a rounding error—that’s a headline with an asterisk the size of the chart.

Here’s what the release actually says, stripped of the spin:

• April 2026 retail and food services sales were $757.1 billion, up 0.5% month-over-month (±0.4%) and up 4.9% year-over-year (±0.5%)—all figures are explicitly “not for price changes.”

• The release states: “The 90 percent confidence interval includes zero. There is insufficient statistical evidence to conclude that the actual change is different from zero.”

• February–April 2026 sales rose 4.4% versus the same period a year ago (±0.4%), nominal.

• Nonstore retailers were up 11.1% year-over-year (±1.8%); Food services and drinking places were up only 2.7% year-over-year (±1.8%).

• The February–March change was revised to +1.6% (±0.2%) from +1.7% (±0.4%)—a small trim that matters for momentum.

• A Special Notice flags that the Annual Revision Report is delayed due to the ARTS-to-AIES transition, adding structural uncertainty.

• Footnote (1) confirms a coverage break from April 2025: estimates now include only businesses with paid employees, excluding nonemployers.

Here’s what the data reveals:

• The +0.5% MoM headline is statistically fragile; the true change may be zero.

• The ~5% YoY pace is nominal, not real; volume growth is unknown.

• Strength is concentrated online (Nonstore +11.1%) while restaurants lag (+2.7%), challenging broad-strength narratives.

• Revisions are active (March nudged down), and methodological changes raise the noise floor.

The 0.5% That Might Be Nothing

The release’s top line touts +0.5% for April, but the same document states there’s insufficient evidence to say it’s different from zero. This isn’t just semantics. A ±0.4% standard error on a 0.5% move implies a very wide tent for interpretation—especially after March was revised down to +1.6% from +1.7%. The market loves clean narratives; the statistics don’t.

Retail trade—excluding food services—was also up 0.5% (±0.4%). That parallel move doesn’t scream acceleration; it confirms a picture of weakly positive nominal trend that could be flat in real terms once prices are accounted for. The release doesn’t adjust for inflation, and it tells you so repeatedly.

For investors parsing signal from noise: set your confidence bands wide. The advance read uses a relatively small sample, and in many cases doesn’t impute for nonresponse unless the nonrespondent is highly influential. Monthly changes around half a percent are exactly where this design’s noise starts crowding the music.

Nominal Mirage: No Prices, No Real Story

The April YoY number rounds to +4.9%, matching the familiar ~5% pace we saw in last year’s late-summer print. But the release is explicit: these estimates are not price-adjusted. In other words, we do not know if volumes rose, fell, or stood still.

This matters now more than usual. If price levels for retail categories are still elevated or uneven, nominal growth can mask flat or negative real demand in specific segments. The three-month comparison (+4.4% for Feb–Apr year-over-year) tells the same story: nominal momentum, real ambiguity. Without the GDP-relevant control group in this advance release, we’re guessing at goods consumption’s direct contribution to Q2 growth.

The punchline: Nominal ≠ real, and the release won’t help you bridge the gap.

E-Commerce Pops, Restaurants Lag—So Much for “Broad-Based”

The headline narrative hints at broad resilience. The details do not.

• Nonstore retailers: +11.1% YoY (±1.8%). Solid double-digit nominal growth. That’s your eye-catcher.

• Food services and drinking places: +2.7% YoY (±1.8%). Barely above the margin of sampling error.

That split is not cosmetic. It implies wallet share shifting toward online and away from discretionary dining, a typical late-cycle or budget-conscious behavior. If restaurant spending is lagging nominally, real volumes could be meaningfully softer, given persistent price increases in food away from home over the past two years. Even if Nonstore is genuinely hot, it doesn’t rescue categories tied to services-heavy cost structures and sticky wages.

The release’s selective emphasis on the strongest segment inadvertently overstates breadth. Investors should assume rotation, not resurgence.

Methods Matter: Noise, Revisions, and a Structural Break

This isn’t the usual statistical housekeeping—there are real investment implications in the footnotes:

• Sample and estimator: The advance estimates use a subsample of roughly 4,800 firms with a link-relative approach. Limited imputation for nonrespondents means results can swing on month-to-month response shifts.

• Delayed benchmark: The Annual Revision Report is delayed by the ARTS-to-AIES transition. Translation: we’ll be living with higher level and trend uncertainty for longer than normal.

• Coverage break (April 2025): Estimates now include only businesses with paid employees; nonemployers were included before the benchmark. That’s a quiet but meaningful structural break that can under-represent micro-sellers and complicate comparisons to earlier vintages.

• Suppressed/Unavailable: Some subseries are marked (S) due to high variability; GAFO and other aggregates may be not available (). The missing pieces increase reliance on partial exclusions rather than a clean control group* proxy for GDP.

Bottom line: treat small monthly changes as provisional. The distribution of revisions is not your friend when the measurement framework is in transition.

What This Means for Markets

Rates and the Fed: Don’t Lean on This Print
- A statistically ambiguous +0.5% MoM does not signal overheating. With the data nominal and fragile, it won’t force the Fed’s hand.
- Expect rates to trade the next inflation and labor prints, not this release. If anything, the lack of clear real demand strength tempers hawkish impulses.

Equities: Rotation > Rising Tide
- E-commerce and logistics: The +11.1% YoY for Nonstore supports selective strength in parcel, last-mile, and warehouse automation. Favor platforms showing unit economics improving alongside volume.
- Restaurants and services discretionary: +2.7% YoY nominal is soft. Margin pressure remains a risk if wage and food input costs outpace tepid traffic. Emphasize operators with pricing power, mix upgrades, and disciplined unit growth.
- Consumer staples: In a nominal-only world, companies with steady price realization and inelastic demand look comparatively defensive, especially where private-label trade-down has stabilized.

Macro and Positioning
- GDP trackers should wait for the full monthly report with the control group, then triangulate with card data and company commentary. Don’t overfit the advance print.
- Given revision risk and structural uncertainty, run wider scenario bands around Q2 consumption. Keep exposure balanced between growth beneficiaries (e-commerce/logistics) and defensives that can weather flat-to-negative real volumes.
- Consider barbell positioning: quality growth tied to online penetration gains on one end; cash-generative defensives with pricing power on the other. For cyclicals, be selective—favor balance sheets that can handle a slower traffic tape.

Looking Ahead

• Watch the next vintage for whether the April +0.5% survives revision—especially as the AIES transition continues to ripple through benchmarks.

• Track category breadth: if restaurants remain stuck near +2–3% nominal while Nonstore stays double-digit, the consumer narrative shifts from “strong” to “selective and cautious.”

• Pair this release with upcoming PCE deflators to infer real goods consumption. Without price adjustment, today’s “growth” may be just inflation’s shadow.

April’s retail “increase” reads more like a statistical shrug than a surge. The momentum the headline implies isn’t confirmed by the confidence intervals, the revisions, or the category mix. For investors, the edge lies in ignoring the headline drumroll and trading the composition, the caveats, and the coming revisions. Favor the parts of consumer spending that are actually growing—and be paid to wait on the rest.