Macy's Beat and Raised Guidance, Yet Its Stock Fell. A One-Time Tariff Refund Explains Why

At 6:56 a.m. Eastern on September 10, Macy's printed a second-quarter report it could read as a win, and the tape said otherwise. Net sales of $4.9 billion, up 1.1%. Comparable sales up 2.7%. Adjusted earnings of $0.63 a share against a street of roughly $0.35. Full-year guidance raised on every line. And by premarket the stock was down more than 3%. It is a common confusion, and it is worth a few minutes to get the record straight.

The report is not the number. The report is a story the company tells about the number, and Macy's story has a hole in it.

The beat leans on money that will not come back.

The $0.63 figure is the flattering one. Macy's also told investors that this quarter's adjusted profit carried a $0.23 net benefit from tariff refunds — import duties the company overpaid on goods that entered before new tariffs took effect, repaid to it by the government. Take that out and adjusted earnings land near $0.40, against $0.35 a year ago. The same story plays out in the gross margin: it expanded 180 basis points to 41.5%, and all 180 of those basis points came from the tariff refund; strip the refund out and the underlying margin moved about 10 basis points.

This matters because a one-time refund is not a run-rate. It is money the company already has — $116 million total, received in full — and most of it is gone as we speak. Roughly $20 million is guided into this year's earnings; the other $96 million is being reinvested back into the business. A shareholder looking at "$0.63 a share, up 14%" and a shareholder looking at "adjusted profit barely beat last year before a government refund" are looking at the same quarter. The market priced it as the second sentence.

The "raise" undershot the room.

To be fair, Macy's did raise its own forecast. Full-year adjusted EPS goes to $2.15 to $2.35, from $2.00 to $2.20. Net sales to $21.675 billion to $21.825 billion. Comparable sales growth to 1.0% to 1.5%. On its own scale, a genuine improvement.

But "raised guidance" and "beat expectations" are different claims, and here they point in opposite directions. The midpoint of the new sales range, about $21.76 billion, still came in below the roughly $21.82 billion analysts had already penciled in. In other words, Wall Street had priced in more sales than the company — even after raising — was willing to promise. The earnings midpoint matched consensus rather than beating it. So the upgraded outlook did not clear the height of the bar; it came up just short of a bar the market had already set. That is why the headline "beat and raise" did not move the stock up.

There is a second reason the guide lands with a thud. The higher profit range embeds only about $0.05 a share from the tariff money. The rest of the refund is reinvestment — spending on the turnaround, not profit. So the "raise" is, in part, the company saying it found a one-time gift and handed most of it back to the strategy instead of to the bottom line.

Where the growth actually lives.

Strip the accounting and ask what shoppers are doing, and the picture is sharper, and more comfortable for Macy's leadership to talk about. Comparable sales were up 2.7% overall, but that is a blend of very different stores. Bloomingdale's comped up 11.3% — its highest second-quarter volume ever. Bluemercury rose 6.2%. The Macy's nameplate itself, the department stores whose name is on the door, comped up just 1.1%, and the renovated "Reimagine 200" locations within it did 1.9%.

This is a portfolio growing fastest where the company is not. The premium — attention, mindshare, and the spending that follows it — is migrating from the legacy department-store banner toward the younger, faster luxury brands, and the company is spending the $96 million of refund money to reinforce that migration: more into Reimagine 200 and the luxury nameplates, less into defending the old floor plan. It is a deliberate bet that the Macy's brand can shrink its way to health while Bloomingdale's and Bluemercury carry the growth. That bet has been winning on the luxury side and only barely winning on the flagship door.

What the low price already knows.

This is where the valuation matters, and it is the same low price either way. The stock trades in the low-$20s, below its 50-day average, with a relative-strength reading in the low 30s — it has been fading for a while, and the earnings-day fall was a continuation, not a shock. On trailing earnings near $2.42 a share, the multiple is about nine times. Macy's carries $1.3 billion in cash against $2.4 billion in debt with no material maturities until 2030, and it pays a dividend of about $0.77 a year — a regular quarterly dividend of 19.15 cents, roughly a 3.5% yield at that price.

A single-digit multiple and a growing dividend on a retailer that just beat are not free money. They are the market's verdict that the number on the screen is not the number underneath: that the run-rate is thinner than the headline, that the flagship brand is still the laggard, and that the "raise" only half-cleared the bar the market had already set. Paying nine times for tariff-refund-flattered earnings with a sales guide below consensus is not obviously a bargain; it is a price that already discounts doubt.

The record, then: a real quarter, comfortably above the estimate, genuinely raised — and the selloff was rational because the durable earnings power behind it is thinner than the headline, and the top line still undershot what the room wanted. The two prints that would update this record are mundane and specific. First: does full-year comparable sales of 1.0% to 1.5% get delivered without more refund assistance, or does the Macy's banner comp start closing the gap to Bloomingdale's and Bluemercury? Second: does the reinvested tariff money — the $96 million handed back to the strategy — show up as durable comp growth, or quietly as a reason the real beat was always smaller than it looked? One of those tells you whether the low multiple is a value, and the other tells you whether it is a trap. Neither is answered yet; the tape is still moving.