S&P 500 7,656.98 +0.86%Nasdaq 26,333.04 +0.96%Dow 52,573.29 +0.98%Russell 2000 2,903.94 +0.45%as of 2026-09-11 close
◢ The SmallCap Signal
Small & mid-cap intelligence for the next market cycle
Analysis

Tariff refunds are making some consumer gross margins unreadable, and Lovesac's Thursday print is the cleanest example yet

Lovesac reported a 1,200 basis point jump in second-quarter gross margin and swung to a GAAP profit while its adjusted EBITDA turned negative — a divergence driven by roughly $21.0 million of IEEPA tariff refunds and interest, $20.0 million of which ran through cost of merchandise sold. Oxford Industries and YETI have posted the same distortion at different magnitudes, and each has drawn the GAAP-versus-adjusted line in a different place.
Illustrative photograph: the United States Capitol building.

There is a version of Lovesac's second quarter, reported Thursday morning, that reads as a turnaround: gross margin of 68.4 percent against 56.4 percent a year earlier, a 1,200 basis point improvement, and GAAP net income of $7.4 million and diluted earnings per share of $0.51 against a year-earlier net loss of $6.7 million and a loss of $0.45 per share. There is another version, in the same release, in which adjusted EBITDA was negative $1.3 million against positive $0.8 million a year earlier. Both are accurate. The difference between them is a tariff refund.

According to the company's Sept. 10 release, Lovesac received approximately $21.0 million of refunds of tariffs levied under the International Emergency Economic Powers Act during the quarter, and related interest. The release specifies the accounting treatment: "Of this amount, $20.0 million was recognized through cost of merchandise sold, $0.3 million reduced inventory, and $0.7 million was recognized as interest income within interest and other income, net." It also quantifies the margin effect itself, rather than leaving it to be inferred: "Gross margin increased primarily due to recoveries of IEEPA tariffs, which contributed 1,240 basis points, and a 250 basis point improvement in product margin driven primarily by price increases." On that arithmetic the refund more than accounts for the whole of the reported 1,200 basis point expansion, with other, unspecified drivers working the other way.

The refund is excluded from adjusted EBITDA. The reconciliation in the release backs out $20.0 million as an IEEPA tariff refund line item, and that single decision is what produces the strange result of a company reporting positive GAAP net income and negative adjusted EBITDA in the same quarter. It is not a red flag; if anything it is the conservative choice, since it keeps a non-recurring customs recovery out of the measure management points investors toward. But it does mean that the headline GAAP figures and the headline non-GAAP figures are describing two different businesses this quarter.

Strip the refund out and the operating picture is flat to soft. Net sales were $161.2 million against $160.5 million, an increase of $0.7 million, or 0.4 percent, which the release says was "primarily driven by 14 net new showrooms." Omnichannel comparable net sales were negative 1.9 percent. Selling, general and administrative expense was $72.3 million, or 44.8 percent of net sales, against $72.1 million, or 44.9 percent. Merchandise inventory was $130.2 million at Aug. 2, 2026 against $124.0 million a year earlier. The company ended the quarter with 284 showrooms, having opened five and closed two during the period.

Chief Executive Shawn David Nelson described the quarter in the release as one in which "second quarter results reflected a record Q2 performance and came in within our guidance range against a choppy category backdrop," adding that "the high end of our business remained a clear source of strength, with resilience building as customers configure larger setups and add value enhancers like Reclining Seat, Lovesoft and Storage." He also said the company enters the back half "in a position of real financial strength, with no debt and a strong cash position," and that it is "appropriately measured" in its outlook for the balance of the year.

The cash is real, and that is the part of the refund story that is not an accounting artefact. Lovesac ended the quarter with $68.8 million in cash and cash equivalents, against $34.2 million as of Aug. 3, 2025. The release states that "there was no balance on the Company's line of credit as of August 2, 2026 and August 3, 2025," with $34.0 million of availability. The company has repurchased $7.2 million of common stock year to date. A customs refund is a genuine transfer of cash to the balance sheet even when it tells you nothing about whether sofas are selling.

On guidance, the release states that the company expects for the full year of fiscal 2027 "net sales in the range of $690 million to $710 million," "net income in the range of $14.5 million to $18.5 million," "adjusted EBITDA in the range of $31.5 million to $35.5 million," and "diluted income per common share in the range of $0.98 to $1.26 on approximately 14.6 million estimated diluted weighted average shares outstanding." For the third quarter it guided to net sales of $140 million to $150 million, a net loss of $9 million to $12 million, an adjusted EBITDA loss of $7 million to $10 million, and a basic loss per common share of $0.62 to $0.83 on approximately 14.5 million estimated basic weighted average shares. The framing sentence carries the qualifier that matters: "The Company's outlook continues to reflect the latest backdrop for tariffs for the remainder of the year, without speculating as to incremental changes that might arise."

Lovesac is not alone. It is the third US consumer name in four weeks, among those examined here, whose reported gross margin has been reshaped by IEEPA refunds, and the three have handled the disclosure differently enough that comparing them is instructive.

Oxford Industries, the owner of Tommy Bahama, Lilly Pulitzer and Johnny Was, reported on Sept. 3 that its gross margin rose to 73.8 percent from 61.4 percent a year earlier, and said the increase was "primarily due to (1) the favorable impact of recognizing $42 million of tariff refund claims as a reduction of cost of goods sold." On second-quarter net sales of $394.4 million, down from $403.1 million, the company reported GAAP earnings per share of $3.25 and adjusted earnings per share of $1.34 — a spread of nearly $2. Its full-year outlook makes the split explicit: for fiscal 2026 ending Jan. 30, 2027, it expects net sales of $1.430 billion to $1.470 billion, GAAP earnings per share of $3.07 to $3.47, "which includes $2.07 of tariff refund receivables and related interest," and adjusted earnings per share of $1.60 to $2.00.

YETI drew the line somewhere else again. In its Aug. 13 release the company stated that "the net tariff benefit consisted of IEEPA tariff refunds, of amounts expensed in 2025 and 2026, recorded as a reduction of cost of goods sold during the quarter, which favorably impacted gross profit by $42.6 million, and gross margin by 890 basis points." On adjusted figures the benefit shrank to 170 basis points, because the company excluded $34.4 million related to the 2025 net tariff impact, plus $2.9 million of interest income, from adjusted results while leaving the portion tied to 2026 tariff expense in. Second-quarter net sales were $483.9 million, up 9 percent, with GAAP gross margin of 66.7 percent against adjusted gross margin of 59.5 percent, GAAP earnings per share of $0.94 including approximately $0.40 of net tariff benefit, and adjusted earnings per share of $0.67 including approximately $0.03.

The practical consequence for anyone reading these three companies' numbers is that their GAAP gross margins and GAAP earnings per share are, this quarter, poor guides to underlying trading. Each has made a defensible but different choice about which portion of a refund belongs in adjusted results — YETI splitting by the year the tariff was originally expensed, Lovesac excluding the refund from adjusted EBITDA entirely, Oxford quantifying the tariff contribution inside its GAAP guidance range. None of these is wrong. All of them break period-over-period comparability at these three companies, and they will break it again in the opposite direction once the refunds stop. How widespread the pattern is across the broader consumer sector is not something these three releases establish.

The distortion is arriving into a tape that has been unkind to smaller companies. TheStreet's market wrap, timestamped 4:33 p.m. ET on Sept. 9, reported that the Russell 2000 closed Wednesday down 1.36 percent at 2,919.86, a steeper fall than the S&P 500's 0.48 percent decline to 7,636.36, the Nasdaq Composite's 0.64 percent decline to 26,253.34 or the Dow's 0.77 percent decline to 52,380.66. It attributed the session to escalating US-Iran tensions, Brent crude settling at $101 a barrel and Treasury yields setting fresh 52-week highs, with the 10-year reaching 4.843 percent. Higher input costs and higher funding costs both bear disproportionately on smaller balance sheets.

The discipline this demands is unglamorous. Read the reconciliation table before the headline, check which year's tariff expense a refund relates to, and treat any gross margin that has moved by several hundred basis points in a single quarter as a question rather than an answer. The refunds are cash and they are welcome; they are also, by construction, a one-off, and the quarters in which they lapse will look worse than the underlying business for exactly the same reason these look better.

This article is for general information only and is not investment advice. Figures are as reported by the cited sources at time of writing.

Related coverage