The clearest early read on the 2026 US holiday season is not on the demand side at all. It sits on retailer balance sheets. The pattern in the last four weeks of data points to an October-quarter earnings round, running from roughly November 17 to December 4, 2026, in which inventory growth outpaces sales growth at a majority of the large US general-merchandise bellwethers, and in which “inventory” and “markdown” re-enter guidance language for the first time since the 2022 glut. The base case here is a managed overhang rather than a repeat of 2022, but the direction is likely set: goods arrived early, and the sales forecasts that would absorb them are being carried by price, not units.
Three independent signals anchor that call, all datable to the last 14–30 days. The National Retail Federation’s port tracker reversed its own slowdown forecast and now expects September to be the busiest import month of the year. Second-quarter filings from Walmart and Best Buy already show inventory growing roughly twice as fast as sales. And the two most-cited holiday forecasts, from Bain and Deloitte, both describe a season in which most of the nominal growth is price and most of the unit growth is online. A future observer can check the prediction against the Q3 balance sheets and the NRF’s November import actuals, both public by mid-January 2027.
In short
- The prediction: at the October-quarter earnings round (November 17 to December 4, 2026), inventory growth is likely to exceed sales growth at a majority of an eight-name panel of US general-merchandise bellwethers (Walmart US, Target, Home Depot, Lowe’s, Best Buy, Dick’s Sporting Goods, Kohl’s, Macy’s), with at least two of them naming inventory or markdown pressure in guidance.
- Timeframe: checkable by December 4, 2026 on the earnings side, and by mid-January 2027 on the import side (NRF’s November actual likely lands at or below November 2025’s roughly 2.02 million TEU).
- Signal 1: the NRF Global Port Tracker (September 9) walked back its early-peak call: September is now forecast at 2.31 million TEU, up 9.6%, while November is forecast down 0.9%. The Port of Los Angeles posted its strongest three months on record through August, with the executive director citing “early holiday shipments.”
- Signal 2: Walmart’s Q2 (August 20) showed US inventory up 6.3% against US net sales up 3.5%; Best Buy’s Q2 (August 27) showed inventories up 8.3% against revenue up 3.6%. Target (August 19) was the exception at 3% inventory growth against 3.8% comps.
- Signal 3: Bain (September 3) attributes more than half of its 4.5% holiday growth forecast to price, with in-store sales up only 2.5%; Deloitte (September 10) frames the season around “value-seeking” and “using promotions to manage spending.” Neither forecast implies the unit demand needed to clear early-arriving stock at full price.
Why this matters now
Every holiday season has a supply-side story and a demand-side story, and they usually run on different calendars. The demand story gets written in October and November, when Adobe, Mastercard and the NRF publish their forecasts and Black Friday coverage begins. The supply story is written in June through September, when containers land and inventory is booked at cost. In most years the two never conflict, because retailers order to a forecast and the forecast is roughly right.
The years that go wrong are the years in which the supply calendar moves and the demand calendar does not. That happened in 2022, when goods ordered during the 2021 shortage arrived into a consumer who had already shifted spending to services. It happened in milder form in 2019, when tariff-driven pull-forward in late 2018 left retail inventory-to-sales ratios elevated through the following spring. The 2026 setup rhymes with both. Importers shipped early to get ahead of tariff changes and fuel costs, and the resulting inventory is now sitting on balance sheets that were already growing faster than sales at the end of July.
The reason this matters in mid-September, rather than in December, is that the October quarter closes in about six weeks. Inventory booked by the end of October is inventory that must be sold, marked down or carried into 2027, and the earnings calls in the second half of November are where management teams will first have to describe it. That makes the Q3 round the natural checkpoint for a prediction about inventory, and it is why this piece is framed around what those calls are likely to say, rather than around how much shoppers will spend.
Signal 1: the import peak moved to September, and November is now forecast down
The NRF and Hackett Associates publish a monthly Global Port Tracker that forecasts containerized imports at the major US ports six months out. The September 9 edition is unusual because it reverses the forecasters’ own prior call. In August the tracker had described an early peak that would be “mostly behind us” by the fall. The new edition puts September at 2.31 million TEU, up 9.6% year over year and above May’s 2.24 million, which had been the year’s high point. Jonathan Gold, the NRF’s vice president for supply chain, was direct about it: “We thought the peak season would be mostly behind us by now, but that’s not the case.”
The shape of the rest of the forecast is what matters for inventory. After September, the tracker has October at 2.11 million TEU (up 1.7%), November at 2.0 million (down 0.9%), December at 2.03 million (up 1.1%) and January 2027 at 2.09 million (down 1.0%). Full-year 2026 lands at 25.7 million TEU, up just 1% on 2025. Read together, the numbers describe a year in which total import volume barely grows but its timing shifts forward by one to two months. That is not more goods; it is the same goods, earlier. As we noted when the September forecast was first lifted to 2.31 million TEU, the pattern looked like a late peak. The more consequential reading is that it is a front-loaded peak with a hollow behind it.
The Port of Los Angeles data, released September 9 and 10, confirms the same pattern from the terminal side. The port handled 955,907 TEU in August, its best three-month stretch on record across June, July and August at more than 2.9 million TEU. Loaded imports in August were 500,302 TEU, flat year over year but 7% above the five-year August average. Executive Director Gene Seroka credited “resilient consumer demand, early holiday shipments and a broad mix of cargo,” and the Retail Industry Leaders Association’s president, Brian Dodge, added that “a large share of holiday merchandise is already in the U.S.” because retailers had been managing tariff and supply-chain risk. Ben Hackett of Hackett Associates cited tariff increases, inflation and fuel costs tied to Iran tensions as the hurdles importers shipped around. The primary source for the monthly forecast is the NRF Global Port Tracker release.
Two things follow. First, if holiday merchandise is largely onshore by the end of September, it is on retailer balance sheets by the end of the October quarter, valued at a cost basis that includes whatever tariff was in force at import. Second, the November import decline is a mechanical consequence of the pull-forward, not a demand signal, and it will likely be reported in mid-December as a soft number just as retailers are trying to describe Q3 inventory as “well positioned.” The tension between those two headlines is the story of the season.
Signal 2: Q2 balance sheets were already running ahead of sales
The second signal is that the inventory build did not start in September. It was visible at the end of July, before the record import quarter had fully landed. Walmart’s second quarter of fiscal 2027, reported August 20, showed total revenue up 5.9% but Walmart US net sales up 3.5% and US comparable sales up 2.6% excluding fuel. Against that, US inventory rose 6.3% and global inventory 6% in constant currency. Chief Financial Officer John David Rainey attributed the increase to “cost inflation as well as higher inventory to support strategic initiatives in the U.S., including the optimization of inventory across fulfillment nodes.” Chief Executive John Furner said most categories were running between 1% and 4%, with consumables at the high end.
Best Buy’s second quarter, reported August 27, is a cleaner illustration because the company does not have a grocery mix to blur the picture. Revenue rose 3.6% to $9.78 billion and comparable sales rose 4.1%, a strong quarter by the company’s own guidance. Merchandise inventories, per the balance sheet in the earnings release, stood at $6.30 billion on August 1, 2026, against $5.82 billion a year earlier, an increase of $480 million or 8.3%. The release notes roughly $34 million of IEEPA tariff refunds contributing to the domestic gross profit rate, which is relevant because those refunds lower the effective cost basis of goods already sold, not of goods still on the shelf.
Target is the counterexample and deserves to be read as one. Its second quarter, reported August 19, showed comparable sales up 3.8% and inventory of $13.2 billion, up about 3%. That is a balance sheet growing slower than sales, which is the healthy configuration, and it reflects a company that spent 2024 and 2025 working inventory down after its own 2022 overhang. The prediction here does not require Target to flip; it requires a majority of the panel to be above the line. Two of the three names that have reported are already there, before the September import surge is booked.
| Retailer (Q2 report date) | Sales growth | Inventory growth | Gap (inventory minus sales) | Management framing |
|---|---|---|---|---|
| Walmart US (August 20) | +3.5% net sales; +2.6% comp | +6.3% US; +6% global (cc) | +2.8 pts | Cost inflation, fulfillment-node optimization |
| Best Buy (August 27) | +3.6% revenue; +4.1% comp | +8.3% ($5.82bn to $6.30bn) | +4.7 pts | No inventory commentary in release |
| Target (August 19) | +3.8% comp | +3% ($13.2bn) | roughly −0.8 pts | Post-2022 discipline holding |
Three caveats on the table. The Walmart figure carries a cost-inflation component, so some of the 6.3% is price rather than units. Best Buy’s growth partly reflects a computing cycle that management has described as strong, which argues for deliberate stocking. And a two-to-five point gap at the end of Q2 is not by itself a problem. It becomes a problem only if the September and October receipts widen it, and if the fourth-quarter demand that is supposed to close it turns out to be price-led rather than unit-led. That is where the third signal comes in.
Signal 3: the holiday forecasts are built on price, not units
The two forecasts published in the last two weeks agree on the headline and diverge only in scope. Bain’s September 3 outlook, covering November and December, projects US holiday sales above $1 trillion for the first time, up 4.5% nominally against 3.5% in 2025. Deloitte’s September 10 forecast, covering November through January, projects $1.70–1.71 trillion, up 4.0% to 4.8%, against 4.1% last year. Both are constructive. Neither, read carefully, implies the unit demand needed to clear a front-loaded inventory position at full margin.
Bain is explicit that more than half of its nominal growth comes from price increases. It splits the season into in-store sales growing 2.5% and nonstore sales growing 9%, with online generating about 60% of total dollar growth, up from roughly 50% in 2025. Take price out of the in-store number and the implied unit growth in physical stores is close to zero. Bain’s Aaron Cheris framed the balance retailers must strike as one “on price and promotions,” and the release lists high gasoline prices, tariff impacts, labor participation at a five-year low, credit card delinquency above its ten-year average and lower personal savings as headwinds. Our earlier coverage of the Bain trillion-dollar forecast read it as an inflation story; for inventory, the relevant point is that inflation lifts sales dollars and inventory dollars together, and does nothing to move boxes.
Deloitte’s language is softer but points the same way. Its economist, Akrur Barua, leans on disposable personal income growth of 4.5% to 5.2% and on digital comparison tools. Its retail vice chair, Natalie Martini, describes consumers “making deliberate choices about how they spend,” showing “value-seeking behaviors across income levels, including brand and retailer switching” and “using promotions to manage spending.” Deloitte projects e-commerce up 7.5% to 8.4%, which again places the unit growth online. A season in which the marginal unit is sold online and the marginal dollar is inflation is a season in which store-based inventory, the bulk of what arrived through the ports this summer, turns slowly.
There is a fourth, supporting data point that is not quite a signal because it is a plan rather than an outcome. Walmart said in August it is deploying roughly $2.9 billion of IEEPA tariff refunds into price, with the effect landing in Q3, and Target and Kohl’s have listed lower prices among the uses of their own refunds. Our analysis of the tariff refund flow into core goods prices expects that money to show up as goods deflation by the November CPI. For inventory, refund-funded price cuts are double-edged: they can move units faster, which helps clear stock, but they also lower the realized value of inventory that was booked at pre-refund cost, which is the accounting definition of a markdown.
What the pattern suggests
Put the three signals in sequence and a fairly specific chain emerges. Goods that would normally arrive in October and November arrived in July, August and September, for reasons that had nothing to do with demand. They landed on balance sheets that were already growing faster than sales at two of the three largest bellwethers to have reported. And the forecasts that would need to absorb them describe a season in which most growth is price and most unit growth is online. Each link is individually mild. Together they point to an October quarter in which inventory grows faster than sales at most large general-merchandise retailers, and in which management teams spend part of the November calls explaining why that is fine.
The specific, checkable form of the prediction is this. On an eight-name panel of Walmart US, Target, Home Depot, Lowe’s, Best Buy, Dick’s Sporting Goods, Kohl’s and Macy’s, at least five are likely to report October-quarter inventory growth above their own sales growth for the same period. At least two are likely to use “inventory,” “markdown” or “promotional” language in guidance or on the call in a way that implies fourth-quarter margin pressure. And the NRF’s November import actual, published in mid-December or January, is likely to land at or below November 2025’s roughly 2.02 million TEU, confirming that the summer surge was timing rather than volume. If all three hold, the pattern is confirmed. If fewer than four panel names are above the line and none flag markdowns, the prediction has failed.
The precedents matter because they set the range of outcomes. The most-cited one, 2022, was a genuine glut: inventory growth in the 30–40% range at the largest discounters, followed by a margin reset and a full year of clearance. Nothing in the 2026 data approaches that scale. Q2 gaps of three to five points, against a backdrop of 1% full-year import growth, look more like the 2019 episode, when tariff pull-forward in late 2018 lifted the Census retail inventory-to-sales ratio for a few quarters without forcing a broad reset. The 2026 base case sits between the two: visible, discussed on calls, absorbed with heavier promotion, but not a crisis.
| Precedent | What pulled imports forward | Inventory outcome | How it resolved | Read-across to 2026 |
|---|---|---|---|---|
| Late 2018 to 2019 | Section 301 tariff tranches on China | Retail inventory-to-sales ratio elevated for several quarters | Slow burn; promotional but no broad reset | Closest analogue: tariff-timed, modest scale |
| 2021 to 2022 | Post-shortage over-ordering, port congestion | Inventory up 30–40% at big discounters in early 2022 | Margin reset, cancellations, year of clearance | Scale not comparable; useful as the bear case |
| Spring 2025 | Pre-tariff front-loading, then a trough | Short overhang, absorbed by summer | Faded within two quarters | Shows retailers now manage pull-forward deliberately |
| Summer 2026 | Tariff changes, fuel costs, Iran-linked freight risk | Q2 gaps of 3–5 points at Walmart, Best Buy; record LA summer | Open | Base case: managed overhang, discussed on Q3 calls |
Wider context: the plateau season and the cost of carrying it
This piece is best read alongside two earlier calls. In July we argued that the 2026 holiday peak would flatten away from Black Friday into an October-to-December plateau, and that Amazon, Walmart and Target would converge on early October. Amazon confirmed Prime Big Deal Days for October 6 and 7 on September 15, inside the window that call anticipated. A flatter, earlier demand curve is good news for an inventory overhang, up to a point: it gives retailers ten weeks rather than five to move stock. It is also, however, a curve that is flattened by promotion, and promotion is what converts a balance-sheet overhang into an income-statement markdown.
The carrying cost has also changed. All four major parcel networks announced higher peak surcharges in the last month, with UPS starting September 27, FedEx September 28, USPS October 4 and Amazon Shipping’s highest tier running November 22 to December 26. UPS, in its late-August announcement, said it expects a 24% jump in volume from the third to the fourth quarter. Our analysis of why peak parcel costs likely outrun the headline rate increase lands on the same conclusion from the other direction: the unit that clears an overhang online, which is where the forecasts put the growth, costs more to deliver this year than last. Store inventory that has to be sold online to move is inventory whose margin is squeezed twice.
Finally, the tariff environment cuts both ways. Refunds of IEEPA duties, which Customs and Border Protection reported at roughly $107 billion completed by late August, are being pushed into price by the largest retailers. That helps velocity. But the goods that arrived this summer were imported under whatever schedule was live at the time, and the refund lowers the cost of goods already sold more than it lowers the book value of goods on hand. A retailer with a heavy September receipt position and a refund-funded price cut in October is, in accounting terms, choosing a markdown on the shelf to protect share. That is a rational choice. It is also exactly the configuration the prediction expects to hear described on the November calls.
Implications for retailers, suppliers and investors
For retailers the operational implication is timing. Inventory that is onshore in September and booked by October 31 will be judged on the Q3 call, not on Black Friday results. Management teams that plan to promote heavily in October, in line with the early-kickoff calendar, should expect to be asked why inventory grew faster than sales and should have the “deliberate early receipt” answer ready. The teams most exposed are those whose Q2 gap was already wide and whose category mix is store-heavy: consumer electronics outside the computing cycle, home and seasonal decor, and apparel at the department-store end of the panel.
For suppliers and brands the implication is order-book risk in the first quarter of 2027. If the November and January import numbers land where the NRF has them, down about 1% each, the pull-forward hangover is already in the forecast. Brands that shipped early into retailer distribution centers should expect slower replenishment orders in December and January, and should model the possibility that retailers push spring 2027 receipts out rather than in. The 2019 precedent suggests the hangover lasts two to three quarters, not one.
For investors the useful frame is the gap, not the level. Inventory growth of 6% to 8% is unremarkable in isolation; inventory growth of 6% to 8% against sales growth of 3% to 4%, sustained for two consecutive quarters, is the pattern that preceded margin guidance cuts in 2022 and the pattern that did not in 2019. The distinguishing variable in those two episodes was unit demand. The Bain and Deloitte forecasts describe unit demand that is flat in stores and healthy online. That argues for the 2019 outcome, with the caveat that the promotional intensity required to get there is likely to show up in fourth-quarter gross margin guidance rather than in a headline miss.
| Scenario | Rough probability | Panel names with inventory above sales at Q3 | Guidance language | NRF November actual | What it would mean |
|---|---|---|---|---|---|
| Managed overhang (base case) | ~55% | 5 to 6 of 8 | “Well positioned,” with margin caution | At or below 2.02m TEU | Prediction confirmed; promotion absorbs stock |
| Clean absorption | ~25% | 3 to 4 of 8 | No markdown language | Above 2.05m TEU | Prediction fails; units stronger than forecasts imply |
| 2022 echo | ~20% | 7 to 8 of 8 | Explicit Q4 margin cut at two or more names | Well below 2.0m TEU | Prediction confirmed, but for a worse reason |
Caveats: what could go wrong
The most important counter-signal is that the retailers have seen this movie. Target’s discipline at 3% inventory growth against 3.8% comps is the clearest evidence that the 2022 episode changed behavior, and Walmart’s CFO framed its own build as a deliberate fulfillment-network decision rather than a demand miss. If the September and October receipts were planned against an early-kickoff promotional calendar, the inventory position at October 31 may already be lower than the summer import numbers suggest, because the goods will have started selling through Prime Big Deal Days and the Walmart and Target October events before the quarter closes. That would shrink the gap at exactly the moment this prediction expects it to widen.
A second counter-signal is composition. A meaningful share of the record summer imports at Los Angeles was empties, transshipment and non-retail cargo, and loaded imports in August were flat year over year. The retail-specific pull-forward may be smaller than the headline TEU count implies. If the inventory that arrived early is concentrated in a few importers that are not on the panel, the panel could come in below the five-of-eight line even while the aggregate story holds.
Third, the demand forecasts could simply be too cautious. Bain’s own release notes tax refunds up $43 billion and an equity market up 23% year over year; Deloitte leans on disposable income growth of 4.5% to 5.2%. If unit demand in stores surprises to the upside, front-loaded inventory looks prescient rather than heavy, and the November calls will describe strong in-stock positions rather than markdowns. The August retail sales print, due September 16, is the first test of that. Our related call that holiday electronics discounts are likely to shrink is a partial hedge against this piece: if discount depth compresses in electronics, at least one large category is clearing stock on pricing power rather than promotion, which would argue against a broad markdown narrative.
Fourth, the inventory-versus-sales comparison is sensitive to cost inflation and to tariff accounting. Inventory is carried at cost; if tariff refunds are booked as reductions to cost of goods sold rather than to inventory, the balance-sheet figure stays high while the margin looks better, and management may reasonably describe that as a non-issue. A reader checking this prediction in December should look at the inventory line and the gross margin guidance together, and treat a wide inventory gap paired with unchanged margin guidance as a partial rather than a full confirmation.
FAQ
What exactly is being predicted, and by when?
That at the October-quarter earnings round (roughly November 17 to December 4, 2026), at least five of eight US general-merchandise bellwethers (Walmart US, Target, Home Depot, Lowe’s, Best Buy, Dick’s Sporting Goods, Kohl’s, Macy’s) report inventory growth above their own sales growth, at least two flag inventory or markdown pressure in guidance, and the NRF’s November import actual lands at or below November 2025’s roughly 2.02 million TEU.
Isn’t this the same as saying the holiday peak flattens into October?
No. The plateau call is about when shoppers spend. This call is about what retailers are holding when they report, and what they say about it. A flatter demand curve is one of the ways an overhang gets absorbed, so the two are linked, but they are checked against different data on different dates.
Why use inventory-versus-sales growth rather than inventory-to-sales ratios?
Because it is what the companies themselves report on the day, without waiting for Census data or normalizing for mix. The Census retail inventory-to-sales ratio is a useful cross-check but lags by six to eight weeks and blends auto dealers and grocers into the picture.
Walmart said its inventory build was deliberate. Doesn’t that undercut the prediction?
Partly. Deliberate builds are still builds, and they still have to be sold. The prediction is about the gap widening and being discussed, not about management admitting a mistake. The 2019 precedent was also deliberate, tariff-timed stocking, and it still left ratios elevated for several quarters.
Could tariff refunds make the overhang disappear?
They can make it cheaper to clear. Walmart is putting roughly $2.9 billion into price with the effect landing in Q3. Refund-funded price cuts move units, which helps, but they also lower the realized value of stock booked at pre-refund cost, which is what a markdown is. The likelier effect is that the overhang shows up in gross margin guidance rather than in a stranded balance sheet.
What if the NRF is wrong again and November imports are strong?
Then the import leg of the prediction fails, and the overhang thesis weakens because it would suggest retailers were adding, not shifting, volume into Q4. The tracker has already revised once this season, so this is a real risk. The balance-sheet leg would still be checkable on its own.
Which retailers are most exposed and least exposed?
Most exposed on current data: Best Buy (8.3% inventory growth against 3.6% revenue) and Walmart US (6.3% against 3.5%), with department stores and home retailers to be confirmed when they report. Least exposed: Target, which is growing inventory more slowly than sales, and off-price chains, which treat other people’s overhangs as supply.
Does a strong August retail sales print on September 16 change the call?
It softens it. Strong in-store unit demand is the single variable that separates the 2019 outcome from the 2022 one, and a surprise on the upside would argue for the clean-absorption scenario. It would not by itself reverse a Q2 gap that has already been booked.
How would a reader falsify this in December?
Count the panel. Fewer than four of eight names above the line, no markdown or promotional language in guidance, and an NRF November actual above 2.05 million TEU together mean the pattern did not hold. Any two of those three would be enough to call the prediction wrong.