The US Census Bureau publishes its advance estimate of July retail and food services sales on Friday, August 14, 2026 at 8:30 a.m. Eastern Time. It is the last broad read on American consumer demand before a fresh 50% tariff wave on Canadian goods takes effect five days later.
The print arrives with an unusually messy set of crosscurrents. June was flattered by an Amazon Prime Day event that fell inside the reference month, gasoline prices have been falling in monthly terms while remaining far above year-ago levels, and the National Retail Federation is forecasting a record back-to-school season on top of a consumer base that is splitting sharply by income.
In short
- Release date: the July 2026 Advance Monthly Retail Trade Report lands Friday, August 14 at 8:30 a.m. ET, per the Census Bureau schedule.
- The baseline: June sales were $768.6 billion, up 0.2% on the month and 6.7% year over year.
- The comp problem: June’s 1.9% jump in nonstore retailers was helped by a Prime Day event running June 23–26, which sets up a possible payback in July.
- Inflation context: July CPI, released August 12, rose 0.1% on the month and 3.4% year over year, with core at 2.5%, so most of any nominal sales gain is real volume rather than price.
- The tariff clock: Section 338 duties of 50% on roughly $20 billion of Canadian goods take effect August 19, making this the last pre-shock reading on categories such as furniture and apparel.
What the Census Bureau actually publishes on August 14
The Advance Monthly Retail Trade Survey, usually shortened to MARTS, is the first official estimate of how much Americans spent at retailers and restaurants in the prior month. It is a sample-based advance figure, which means it is revised twice before it settles.
The Census Bureau confirms the July reference month lands on August 14 at 8:30 a.m., with August data following on September 16 and September data on October 15. The quarterly retail e-commerce report, which gives a cleaner online-only split, is not due until November 19.
Because MARTS is nominal, it is not adjusted for inflation. A 0.4% monthly gain in a month when goods prices rose 0.3% is a very different story from the same gain when prices were flat. That distinction matters more than usual in 2026.
Which numbers actually move markets
Three figures do most of the work. The headline number captures all thirteen major categories including autos and gasoline, and it is the one that leads wire coverage. It is also the noisiest.
The ex-autos measure strips out motor vehicle and parts dealers, which alone represent about 20% of total sales and can swing the headline on their own. Ex-autos-and-gas removes the second big distortion, the price-driven movement at filling stations.
The most watched figure among economists is the control group, which excludes autos, gasoline, building materials and food services. It feeds directly into the consumption line of GDP. TD Economics put June’s control group at 0.5% on the month, notably stronger than the 0.2% headline.
Why revisions deserve attention this month
Advance estimates carry meaningful confidence intervals. Census reported June at up 0.2% with a margin of plus or minus 0.4 percentage points, which is wide enough that the sign of the change was not statistically certain.
Revisions to May and June will be published alongside the July figure. In a month with as many one-off distortions as this one, the revision to June’s Prime Day-inflated online number may end up mattering as much as the July headline itself.
Why June set an awkward baseline for the July print
June looked soft on the surface and firmer underneath. Total sales rose 0.2% to $768.6 billion, but the modest gain was dragged down by a 5.3% monthly drop at gasoline stations, a category that is roughly 7% of the total and moves on pump prices rather than volume.
Strip that out and the picture improves. Our coverage of the June 2026 retail sales report noted that online spending led the gains, and the inflation-adjusted change was around 0.6%, three times the nominal headline.
The gasoline distortion cuts both ways
Gasoline station receipts fell 5.3% from May but were still up 19.8% against June 2025. That combination is unusual and it means the category is subtracting from monthly prints while flattering the annual comparison.
July CPI showed gasoline down another 2.9% on the month while still 24.6% higher than a year earlier. If that pattern held in the sales data, gasoline stations likely subtracted from the July headline again.
The Prime Day payback problem
Nonstore retailers, the category that captures most e-commerce and carries a 17% weight, rose 1.9% in June. That was the largest monthly gain among the major categories, and reporting attributed a good share of it to an Amazon Prime Day event that ran June 23–26.
Pulling demand forward into late June creates an arithmetic problem for July. Households that bought a laptop or a vacuum during the event are unlikely to buy another three weeks later, so the online category faces a hard comparison even if underlying demand is stable.
The counterweight is timing. A late-June event pushes some deliveries and some back-to-school purchasing into July, and Census records sales when the transaction occurs rather than when the parcel arrives. The net effect is genuinely ambiguous.
What July inflation data already told us
The July consumer price index, published on August 12, gives a useful head start on interpreting Friday’s sales figure. Headline CPI rose 0.1% on the month and 3.4% year over year, easing from 3.5% in June. Core CPI, excluding food and energy, rose 0.2% on the month and 2.5% annually.
Both readings were broadly in line with economist expectations. That matters for the retail sales print because it means a strong nominal number in July would mostly reflect real volume rather than price increases.
| CPI category (July 2026) | Month over month | Year over year |
|---|---|---|
| Headline CPI | +0.1% | +3.4% |
| Core CPI (ex food and energy) | +0.2% | +2.5% |
| Energy | -1.5% | +14.7% |
| Gasoline | -2.9% | +24.6% |
| Electricity | +0.1% | +4.2% |
| Food, all | +0.1% | +3.0% |
| Food at home | -0.1% | +2.7% |
| Food away from home | +0.3% | +3.4% |
| Shelter | +0.1% | +3.2% |
| Airline fares | +2.2% | +25.5% |
The gap between core goods and core services
The composition is as informative as the level. New vehicle prices rose 0.1% in July while used cars and trucks rose 0.4%, and medical care was up 0.4%. Airline fares accelerated 2.2% on the month.
Food at home actually fell 0.1%, which is a meaningful relief for grocers who have spent two years defending basket sizes. Food away from home rose 0.3%, continuing the long-running pattern of restaurant inflation outpacing supermarket inflation.
Analysts reading the July report described tariff pass-through as uneven rather than absent. Some categories appear to have worked through most of the adjustment, while others are only beginning to move surcharges onto shelf prices. Household furnishings and supplies, which includes furniture and floor coverings, was cited as a category showing exactly that mixed picture.
How far tariff costs have actually reached the shelf
The central question hanging over Friday’s release is whether the 2025 and 2026 tariff rounds are showing up as lower volumes, higher prices, or compressed retailer margins. So far the evidence points to all three in different categories.
The structural change is larger than any single duty rate. The United States eliminated its $800 de minimis threshold for all countries on August 29, 2025, meaning every inbound shipment now goes through formal or informal customs entry and is assessed duty on its declared value. We covered the follow-on step when the last parcel loophole closed on July 24, which raised the postal prepaid-duty threshold to $2,500 and ended the Section 122 surcharge on postal shipments.
Why the pass-through has been slower than expected
Importers front-loaded. Retailers spent much of the first half of 2026 pulling forward inventory ahead of announced effective dates, which meant summer shelves were partly stocked with pre-tariff goods bought at pre-tariff landed cost.
That buffer is finite. As pre-tariff inventory sells through, the replacement cost resets at the new duty rate, and the price effect shows up with a lag of one or two quarters rather than immediately.
Duty drawback and foreign trade zone strategies absorbed some of the remainder. Neither is a permanent shield, and both add compliance cost that eventually lands somewhere in the price.
The global de minimis picture behind the US numbers
The US is no longer an outlier. The low-value import exemption that underpinned a decade of cross-border direct-to-consumer shipping is being dismantled across the major consumer markets, and the sequencing matters for anyone reading American e-commerce data.
The EU removed its EUR 150 customs duty exemption from July 1, 2026, following a Council decision in November 2025. In its place sits an interim flat-rate duty of EUR 3 per item category on small parcels shipped from businesses outside the bloc to EU consumers.
A separate EU-wide handling fee intended to cover customs processing costs has been proposed but not finalized. The Commission has indicated the amount and start date will be determined in autumn 2026, with industry expectation pointing to November. The UK has confirmed its GBP 135 threshold will be removed, though on a much longer timeline running to March 2029.
| Market | Old threshold | Status | Replacement charge |
|---|---|---|---|
| United States | USD 800 | Removed August 29, 2025 | Standard duty by HTSUS code, formal or informal entry |
| European Union | EUR 150 | Removed July 1, 2026 | EUR 3 interim flat-rate duty per item category |
| EU handling fee | Not applicable | Proposed, amount and date due autumn 2026 | To be determined |
| United Kingdom | GBP 135 | Confirmed for removal by March 2029 | Not yet specified |
Nonstore retailers is the MARTS category that captures most e-commerce, and it is the one where cross-border economics bite hardest. When every parcel carries duty and a customs entry, the cost advantage of shipping direct from overseas narrows against domestic fulfillment.
The observable response has been a shift toward holding inventory inside the destination market. That converts a variable per-parcel cost into a fixed warehousing cost, which favors larger sellers and squeezes the long tail of small cross-border merchants.
For the July data specifically, the effect is more likely to appear as a change in who is selling than as a change in how much is sold. Domestic marketplaces and US-warehoused sellers absorb volume that previously arrived as direct international parcels, and both land in the same nonstore line.
Why the August 19 tariff date makes this a pre-shock reading
Five days after the July sales figure, a new and legally novel tariff takes effect. On July 20, 2026, three presidential proclamations imposed an additional 50% ad valorem duty on a range of Canadian goods under Section 338 of the Tariff Act of 1930, effective 12:01 a.m. ET on August 19.
It is the first use of that statute in nearly a century. Section 338 permits duties of up to 50% on imports from a country judged to discriminate against US commerce, and unlike several other tariff authorities it carries no built-in expiry.
The retail exposure is wider than the headline sectors suggest. The named triggers were motor vehicles, alcoholic beverages and dairy, but the annexes reach furniture, clothing and textiles, cement, plywood, paper, cosmetics, fishing rods, hockey sticks and swimming pools. Our earlier analysis of what the 50% tariffs on Canadian goods mean for retailers set out the category detail.
| Feature | Section 338 (Canada, from Aug 19) | De minimis repeal (from Aug 2025) |
|---|---|---|
| Rate | 50% additional ad valorem | Standard HTSUS rate by 10-digit code |
| Scope | Roughly $20bn of Canadian imports | All inbound shipments, any value or origin |
| Share of source-country imports | About 5% of $382bn from Canada in 2025 | Not applicable, universal |
| Trade-agreement relief | USMCA origin does not exempt | Not applicable |
| Stacking | Does not stack with Section 232; reportedly stacks with Section 301 | Stacks with all applicable duties |
| Duration | No statutory expiry; revocable by proclamation | Permanent under current rules |
| Notable exclusions | Oil, gas, critical minerals, potash, sector-tariffed goods | Limited |
The legal question that will not resolve quickly
Trade lawyers have flagged genuine uncertainty. Analysis published by CSIS notes that no court has yet interpreted Section 338, that the statute contemplates a US International Trade Commission investigation which the proclamations do not confirm took place, and that some analysts argue the provision was implicitly repealed by later legislation.
Prime Minister Mark Carney has called the measures a violation of the trade agreement and said Canada stands ready to engage intensively. Talks have not been cut off, and the 30-day window between signature and effective date has been widely read as a negotiating deadline rather than a fixed schedule.
For retailers the practical implication is planning under uncertainty. A duty that could be suspended by proclamation at any point is difficult to price into a holiday assortment that was committed months ago.
Back-to-school is the swing factor in the July number
July is when back-to-school purchasing begins in earnest, and the National Retail Federation is forecasting a record season. Total back-to-school and back-to-college spending is projected at $146.8 billion, which would make it the second-largest US consumer spending event behind the winter holidays.
| Back-to-school segment | 2025 | 2026 forecast | Change |
|---|---|---|---|
| K-12 total spending | $39.4bn | $43.3bn | +9.9% |
| College total spending | $88.8bn | $103.5bn | +16.6% |
| Combined total | $128.2bn | $146.8bn | +14.5% |
| Average K-12 household spend | $858.07 | $863.86 | +0.7% |
The record hides a widening split
The two rows in that table tell different stories. Aggregate K-12 spending is forecast to rise about 9.9%, while the average household plans to spend only 0.7% more, at $863.86 against $858.07 last year.
That gap implies the total is being driven by participation and by price rather than by families choosing to buy more. Commentary on the forecast has described the pattern as a record on the surface and K-shaped underneath.
Consumer expectations reinforce that reading. NRF survey work found 64% of back-to-school shoppers expect to see higher prices this year, and shoppers started earlier and leaned harder on promotions than in prior seasons.
What the individual categories are likely to show
Reading MARTS well means reading it by weight. A large percentage move in a small category barely registers, while a modest move in motor vehicles can define the headline.
| Category | Share of total sales | June 2026 month over month |
|---|---|---|
| Motor vehicle and parts dealers | 20% | +1.9% |
| Nonstore retailers (mostly e-commerce) | 17% | +1.9% |
| Food services and drinking places | 14% | +0.1% |
| Food and beverage stores | 12% | Little changed |
| General merchandise stores | 10% | Little changed |
| Gasoline stations | 7% | -5.3% |
| Building material and garden equipment | 6% | Little changed |
| Health and personal care stores | 5% | Little changed |
| Clothing and accessories stores | 3% | Small decline |
| Furniture and home furnishings | 2% | Flat |
| Miscellaneous store retailers | 2% | Small decline |
| Sporting goods, hobby, music and books | 1% | Little changed |
| Electronics and appliance stores | 1% | Little changed |
Categories most exposed to the tariff story
Furniture and home furnishings is small in the index at roughly 2% but disproportionately important as a signal. It is directly in the path of the Section 338 annexes, it was flat in June, and July CPI showed household furnishings as a category with visible but uneven tariff effects.
Clothing and accessories carries a 3% weight and posted a small monthly decline in June despite being up 4.8% against June 2025. Canadian apparel and textile lines across Harmonized Tariff Schedule chapters 50 through 63 are covered by the new duties, though Canada is a modest share of total US apparel imports.
Nonstore retailers is where the de minimis repeal shows up most directly, because cross-border direct-to-consumer parcels now carry duty and formal entry costs that did not exist two years ago.
Where a positive surprise would most likely come from
Food services and drinking places is the category to watch for a genuine upside surprise. It carries a 14% weight, it managed only 0.1% in June, and restaurant demand tends to hold up when households feel secure about employment.
General merchandise, at 10%, is the other candidate. It captures the discount and mass channels that gain share when consumers trade down, so strength there can coincide with weakness elsewhere. That has been a recurring pattern this cycle, and it also appeared during the spending surge that pushed US retail spending to a four-year high earlier this summer.
What counts as a beat, a miss and a warning
Market commentary ahead of the release has settled on a rough set of thresholds for the headline monthly change. They are useful as a framework even though consensus forecasts for this particular print were still forming in the days before publication.
A reading above 0.6% on the month would be read as clear consumer resilience, particularly given the Prime Day comparison working against it. That would imply back-to-school demand more than offset the June pull-forward.
A reading in the 0.2–0.5% range would signal modest growth in line with the recent trend. Given June came in at 0.2%, anything in that band represents continuity rather than a turn.
A flat or negative headline would raise genuine questions about consumer health. The important qualifier is that a negative headline driven entirely by gasoline stations, which fell 5.3% in June and saw pump prices decline again in July, would be a price effect rather than a demand signal.
The three tables to open first
Census publishes the release with monthly percent changes, year-over-year changes and revised prior months in the same document. Reading them in the wrong order produces the wrong conclusion.
Start with the not-adjusted year-over-year column for a sanity check on trend, because it is immune to the seasonal adjustment factors that generate most monthly noise. June’s 6.7% annual gain was the more reliable signal that month, not the 0.2% monthly change.
Then read the revisions. If June’s 0.2% is revised up materially, the Prime Day effect was larger than first estimated and the July comparison is correspondingly harder. If it is revised down, the opposite applies and a soft July print carries less information.
What the print means for GDP and the Federal Reserve
The control group is the transmission channel. TD Economics estimated after the June release that consumer spending was tracking about 2% annualized in the second quarter, a clear pick-up from roughly 0.5% in the first quarter.
A July control group at or above 0.4% would keep third-quarter consumption tracking near that pace. A flat or negative reading would raise questions about whether the second-quarter acceleration was a tax-refund and equity-market effect rather than a durable improvement.
On rates, the July CPI report has already shaped expectations. Ellen Zentner of Morgan Stanley said in-line inflation “will keep the ‘no need to hike rates’ narrative intact,” while Lindsay Rosner of Goldman Sachs said contained core inflation “adds to encouraging signs” consistent with a September hold.
Retail sales rarely override CPI in that calculus, but a large miss in either direction would be read as evidence about how much room the consumer has left as the next tariff tranche lands.
What retailers, sellers and marketplaces should watch
For operators the useful signal is not the headline but the composition. Three things are worth extracting from Friday’s tables.
First, whether nonstore retailers hold positive despite the Prime Day comparison. A flat or slightly negative July after a 1.9% June would be a normal payback. A sharp decline would suggest the online channel is genuinely losing momentum as landed costs rise.
Second, whether the gap between general merchandise and discretionary categories widens further. Trade-down behavior is the clearest available proxy for household stress, and it shows up in relative category growth before it shows up in the aggregate.
Third, whether restaurant spending diverges from grocery. Food away from home inflation at 3.4% against food at home at 2.7% means any nominal restaurant strength needs to be discounted before it is read as real demand.
Practical steps before August 19
Importers with Canadian-origin goods have a hard deadline. Goods admitted to a foreign trade zone after August 19 must generally enter under privileged foreign status, and goods still in a warehouse will pick up the duty when withdrawn for consumption.
Assortment decisions for the fourth quarter are the second pressure point. Categories covered by the annexes will reprice, and retailers that committed holiday buys before July 20 are absorbing a cost that was not in the plan.
Labor planning is the third. The cost pressure arrives in the same quarter that seasonal staffing decisions are made, which is part of why holiday 2026 retail hiring is tracking toward a new low as automation absorbs volume that used to require temporary headcount.
How to read Friday’s number without overreacting
Advance retail sales are a first draft. The margin of error on the monthly change is wide, the series is nominal, and two of the thirteen categories can move the headline on their own.
The disciplined approach is to look at the control group first, then the composition, then the revisions to prior months, and only then the headline. On a month with a Prime Day distortion in the base and a tariff deadline five days out, that ordering matters more than usual.
The full release schedule and the underlying tables are published by the Census Bureau on its economic indicator release calendar.
Frequently asked questions
When exactly is the July 2026 US retail sales report released?
Friday, August 14, 2026 at 8:30 a.m. Eastern Time. The Census Bureau publishes the Advance Monthly Retail Trade Report for July, with August data due September 16 and September data due October 15.
What were June 2026 retail sales?
Advance estimates put June retail and food services sales at $768.6 billion, up 0.2% from May with a margin of plus or minus 0.4 percentage points, and up 6.7% against June 2025. Sales for the April to June quarter were up 6.4% year over year.
Why might July online sales look weak?
Nonstore retailers rose 1.9% in June, helped by an Amazon Prime Day event that ran June 23–26. Demand pulled forward into late June creates a difficult comparison for July even if underlying online demand is unchanged.
What is the control group and why do economists prefer it?
The control group excludes motor vehicles, gasoline, building materials and food services, removing the most volatile and most price-driven categories. It feeds directly into the consumption component of GDP. TD Economics put June’s control group at 0.5%, against a 0.2% headline.
Has tariff pass-through shown up in consumer prices yet?
Partially and unevenly. July CPI rose 0.1% on the month and 3.4% annually, with core at 2.5%. Analysts reading the report described some categories as having worked through most of the adjustment while others are only starting to pass surcharges to consumers, with household furnishings cited as a mixed case.
What happens to Canadian imports on August 19, 2026?
An additional 50% ad valorem duty takes effect at 12:01 a.m. ET under Section 338 of the Tariff Act of 1930, covering roughly $20 billion of Canadian goods across hundreds of tariff classifications. USMCA origin does not exempt covered goods.
Which retail categories are covered by the Section 338 tariffs?
Beyond the named motor vehicle, alcoholic beverage and dairy sectors, the annexes reach furniture, clothing and textiles, cement, plywood, paper, cosmetics, fishing rods, hockey sticks, seeds, wigs and swimming pools. Oil, gas, critical minerals, potash and goods already under sector-specific tariffs are excluded.
How much are Americans expected to spend on back-to-school in 2026?
The National Retail Federation forecasts a record $146.8 billion in total, split between roughly $43.3 billion for K-12 and $103.5 billion for college. Average K-12 household spending is projected at $863.86, only marginally above $858.07 in 2025, and 64% of shoppers expect higher prices.
Will the retail sales report change Federal Reserve expectations for September?
Probably only at the extremes. July CPI came in close to forecasts, and commentary from Morgan Stanley and Goldman Sachs pointed toward a September hold. A large surprise in the retail control group would be read as evidence about remaining consumer capacity rather than as a direct policy trigger.