On Thursday evening Lululemon reported quarterly earnings of $2.92 a share. Analysts had modelled $1.82. That is a 60 percent beat — the kind of number that normally carries a stock double digits higher. On Friday the shares fell 17.38 percent to $100.61, the lowest level in eight years. Roughly $2.4 billion of market value disappeared in a single session.
The resolution sits in one line of the release: $134.5 million of the quarter’s result was a tariff refund. Of the $2.92 per share, $0.86 — very nearly 30 percent of reported earnings — came not from customers but from the United States government, after a court struck down the duties that had been collected. Strip that item out and Lululemon earned $2.06 a share against $3.11 in the year-ago quarter. That is not up 60 percent. That is down 34 percent.
What the quarter actually contained
Revenue fell 4 percent to $2.415 billion, down 5 percent in constant currency, missing the $2.46 billion consensus. Comparable sales — the measure that strips out new stores and shows what the existing base did — dropped 9 percent, or 10 percent in constant dollars. In the Americas, the home market that still carries most of the business, comparable sales fell 12 percent.
Reported gross margin rose to 60.5 percent, up 200 basis points. That is the refund again: $134.5 million on $2.415 billion of revenue is roughly 557 basis points. Without it, gross margin would have been about 54.9 percent against 58.5 percent a year earlier — down roughly 360 basis points. The cross-check confirms it. In the first quarter, which booked no refund, gross margin was 54.2 percent. The 60.5 percent is the outlier, not the run rate.
At the operating line the distortion is larger still. Reported operating income was $453.7 million, down 13 percent, an 18.8 percent margin. Excluding the refund, $319.2 million remains against $521.5 million a year ago — down 38.8 percent, at an operating margin near 13.2 percent rather than 20.7. A company whose revenue slips 4 percent while operating profit falls almost 39 percent does not have a soft-patch problem. It has an operating leverage problem, and leverage cuts in both directions.
The money came from a court, not from the sales floor
Where the $134.5 million came from is the most interesting part of the story. From February 2025 the administration levied tariffs under the International Emergency Economic Powers Act — first on Chinese imports, then on Canada and Mexico, and from April 2025 on most trading partners. On 20 February 2026 the Supreme Court held those tariffs invalid. Lululemon then sued in the Court of International Trade to recover the duties it had paid. The $134.5 million is the result.
That makes it a line item with three characteristics that disqualify it as a source of earnings at once: it is non-recurring, it is political, and it is contested. Contested because a class action filed in June 2026 alleges the company passed roughly $240 million of those tariff costs to shoppers through higher prices while simultaneously demanding a full refund from the government, with no commitment to return any of it to the people who paid. Nike faces a comparable suit over roughly $1 billion of refunds. Across the United States, tariff receipts of more than $133 billion sit under court-ordered refund risk.
The scale explains the size of the item. Going into this year the company had guided to roughly $380 million of tariff costs. Part of that was pushed through in price, and that is exactly where the claim bites: prices stayed up when the tariffs came down. For earnings quality it means the $134.5 million is not only non-recurring but possibly not final. A partial win for the plaintiffs would create a provision that claws back part of the refund — in a fiscal year whose operating result no longer has the cushion to absorb it.
The general lesson runs past the specific case. When the only strongly growing line in a quarter originates in litigation, earnings per share has stopped measuring the operating business for that quarter. The market took exactly one session to do that arithmetic, then destroyed 17.8 times as much market value as the refund was worth.
The guidance was the real news
Anyone reading only the headline number misses the reason for the collapse. Lululemon cut its full-year outlook for the third time in six months, and by an amount that changes the character of the revision.
In March, alongside fiscal 2025 results, the company guided fiscal 2026 revenue to $11.35–$11.50 billion — growth of 2 to 4 percent — with earnings of $12.10 to $12.30 a share. By June that had become $11.00–$11.15 billion of revenue and $10.95 to $11.15 of earnings. The guidance now reads $10.35–$10.50 billion of revenue, a decline of 5 to 7 percent, with earnings of $9.48 to $9.73.
At the midpoints: the revenue outlook has fallen from $11.425 billion to $10.425 billion. That is exactly one billion dollars removed in two quarters, 7.4 times the refund that generated the headline beat. The earnings outlook went from $12.20 to $9.605, down 21 percent against the company’s own March forecast and down 27.6 percent against the $13.26 actually delivered last year. Street consensus still stood at $11.03 billion of revenue; the new midpoint is $605 million below it.
The direction inside the year matters more than the level. For the third quarter now under way the company guides to $2.29–$2.32 billion, a decline of 10 to 11 percent. The second quarter was down 4 percent. Backing the full-year number into the fourth quarter implies roughly $3.2 billion against $3.6 billion last year — also around minus 10 percent. Management is not forecasting that the decline flattens in the second half. It is forecasting that the decline holds at the worse level.
A brand problem, not a consumer problem
It would be convenient to blame the economy. The calendar forbids it. That same Friday the Labor Department reported 162,000 nonfarm payroll additions for August against 55,000 expected. Unemployment held at 4.1 percent, June and July were revised up by a combined 55,000, and the household survey showed 683,000 people entering the labour force. The American consumer has a job. They are simply not buying this brand.
The detail lines say the same thing. Inventory is disciplined: down 1 percent in dollars and down 7 percent in units. There is no pile of goods that has to be cleared through markdowns — the margin is not suffering on price, it is suffering on volume. The interim co-chief executives named traffic explicitly as the driver, along with negative commentary in the media and social channels, which impacted traffic and a softer-than-planned response to new product launches. Leggings, the foundation of the franchise, fell roughly 20 percent as demand shifted toward looser, away-from-body silhouettes.
That is the most uncomfortable diagnosis a consumer brand can receive. A price problem is solved with price, a cost problem with costs, an inventory problem with markdowns. A traffic and taste problem is solved only with an assortment that reaches stores two to four quarters from now. There is a partial offset: outside the Americas the brand still works. China Mainland revenue rose 4 percent as reported, though down 2 percent in constant currency, and the rest of the world grew 5 percent, or 6 percent in constant dollars.
What the buyback says about management’s own read
One line from the cash flow statement deserves more attention than it gets. During the second quarter Lululemon repurchased $330 million of its own stock — 2.7 million shares, an average price near $122. On Friday the stock closed at $100.61. The capital deployed into buybacks during the quarter is therefore down roughly 17.7 percent, almost exactly the size of the one-day fall.
This is not hindsight criticism, it is information about the information set. The same leadership that spent $330 million on its own shares during the quarter cut the annual forecast for the third time a few weeks later. A repurchase is a statement about intrinsic value. When a single reporting period contains both the purchase and the cut, intrinsic value was not knowable at the moment of purchase. For judging future capital allocation, that matters more than any guidance line.
What this means for a US portfolio
Lululemon pays no dividend, so the case is a pure price story with no withholding question attached. Gains realised on a position held more than a year fall under long-term capital gains at 0, 15 or 20 percent depending on taxable income, plus the 3.8 percent net investment income tax above the applicable thresholds; positions held a year or less are taxed at ordinary rates. Anyone harvesting a loss here should watch the wash-sale rule — repurchasing the same or a substantially identical security within 30 days before or after the sale disallows the loss and rolls it into the new basis, and a same-sector ETF is usually the cleaner substitute than an option position on the same name.
The more useful angle is the read-across. Nike is the obvious comparison and now the direct one, since the incoming chief executive spent 26 years there. Deckers Outdoor, through HOKA, and On Holding are on the other side of the same taste cycle that is currently working against Lululemon. Under Armour is the cautionary tale of what happens when a brand loses the frequency and tries to buy it back with distribution. Gap, through Athleta, competes directly at a lower price. And the two brands taking the most share — Alo Yoga and Vuori — are private, which means the damage shows up in Lululemon’s comparable sales long before it shows up in anyone’s reported revenue. The metric to track across all of them is never the earnings beat; it is comparable sales and the traffic underneath.
The counter-arguments worth knowing
The case is not as one-sided as a 17 percent decline suggests. First, the balance sheet is clean: $1.39 billion of cash, no meaningful net debt, and inventory shrinking faster in units than revenue is shrinking in dollars. Second, an 18.8 percent operating margin — or even the 13.2 percent that remains after removing the refund — is not a distress number in apparel; it is above what most competitors reach in good years. Third, the international business is still growing, and 825 stores are a distribution network that cannot be replicated quickly.
Fourth, valuation: at $100.61 against guided earnings of $9.605, the stock trades near 10.5 times this year’s expected profit, a multiple that already discounts a considerable part of the decline. And fifth, the refund is real cash sitting in the bank, whatever one thinks of where it belongs in the income statement.
Outlook: the test is written into the company’s own guidance
On 8 September, three days after this drop, Heidi O’Neill takes over as chief executive. She arrives from Nike, where she spent 26 years and latterly served as president of consumer, product and brand, and she succeeds an interim pairing of the chief financial officer and the chief commercial officer. New chief executives who inherit a decline have a structural incentive to reset expectations early and thoroughly — which means a fourth guidance cut would not falsify the thesis, it would be a routine part of the sequence.
The testable proposition sits in the company’s own numbers. Guidance assumes minus 10 to 11 percent in the third quarter and roughly minus 10 percent in the holiday quarter: a decline that stabilises at a worse level rather than deepening. If third-quarter comparable sales fall by more than the 10 percent just reported, the annual figure fails again. If they hold, the assumption that this is an assortment cycle rather than an erosion of the brand becomes tenable for the first time in a year. The number to watch is not earnings per share — with no tariff item, the third quarter is guided to $0.93–$0.98 and will look thin regardless — but traffic. And the prior low of $97.99, which after Friday sits barely below the close.
Try TradingView Free for 30 Days
Plus get a $15 discount on your first subscription through this link.
Read more in our topic hub: Topic Hub: Quarterly Earnings Tracker 2026
