A beat on the margin line, a billion dollars cut from revenue guidance, and a 20% cancellation rate — all in the same release.
By FinancialMarkets.com · July 22, 2026
Homebuilder earnings in a soft market usually tell one of two stories. Either the builder chases volume and bleeds margin, or it protects margin and admits the buyers are not there. On Tuesday, D.R. Horton did something rarer. It told both stories at once, in the same release, and left investors to decide which one to trade on.
The flattering story first. America''s largest builder earned $3.20 a share against a FactSet consensus of $2.97. Revenue of $9.23 billion beat the $9.14 billion estimate. And the number housing analysts watch most closely, gross margin on home sales, came in at 20.7%, above the top of management''s own guidance. Then the other story: orders were flat, cancellations jumped, and Horton cut a billion dollars and change out of its full-year revenue outlook. Shares nudged up about 1.5% before the open and then spent the day trying to make up their mind.
The Margin Was Earned the Hard Way
It would be easy to assume a builder hits a 20.7% margin in this market by quietly juicing incentives and calling it discipline. That is not what the numbers show. Horton closed 23,983 homes, up 4% and at the high end of its plan, while average prices held near $362,000, down only 2% from a year ago. The margin help came from the factory floor, not the sales office. Construction costs per home fell 5% year over year and 2% from the prior quarter. Build times shrank by about three weeks. The machine is running faster and cheaper.
Executive Chairman David Auld summed up the operating philosophy: teams are "managing each community with discipline, balancing pace, price, incentives and inventory levels to maximize returns." Strip the corporate phrasing and it means this. Horton will not buy volume it has to pay for twice, once in incentives and again in resale values. This quarter, that restraint worked.
What restraint cannot do is manufacture growth. Net income still fell 12% to $904.9 million. Per-share earnings fell just 5%, and the difference is arithmetic: Horton bought back $615.7 million of stock in the quarter, $2.2 billion for the fiscal year so far, and the diluted share count is down 7%. Shareholders feel a softer landing. The underlying business is still descending.
The Part of the Story Management Could Not Dress Up
Orders came in at 23,084 homes, flat with a year ago. The cancellation rate rose to 20%, up from 16% just last quarter and 17% a year earlier. Chief Executive Paul Romanowski said the spring selling season started out "relatively in line with normal seasonality," then softened after the company''s April earnings call. Auld named the culprit without hedging: "Affordability constraints and cautious consumer sentiment continue to impact new home demand." Incentives, he added, will stay elevated through the fourth quarter.
The guidance did the rest of the talking. Full-year revenue is now expected at $32.5 billion to $33 billion, down from the $33.5 billion to $34.5 billion range set in April and well below the roughly $33.8 billion analysts had penciled in. Closings guidance fell to 83,800 to 84,300 homes from 86,000 to 87,500. Note what is absent from the explanation. No charge. No supply problem. No accounting noise. The company planned for more buyers than showed up, and it said so plainly. In this market, that candor is worth something even when the content stings.
One detail most of the coverage walked past: the order backlog ended June at 15,983 homes worth $6.2 billion, up 14% in units from a year ago. A weak prior-year base helps that comparison, and a 20% cancellation rate erodes the quality of any backlog. Still, this is not what freefall looks like. It looks like stagnation at enormous scale, which is a different problem with different remedies.
The Defense Holds for Now. The Question Is Next Year.
For the fourth quarter, management guided to a 20.5% to 21% gross margin, and finance chief Bill Wheat said he expects a "relatively stable margin," with the honest caveat that many homes are sold and closed inside the same quarter, so visibility is limited. Near term, believe them. The cost machine is genuinely working.
The further out you look, the thinner the cushion gets. The easy construction savings are being harvested now. Lot costs are up 5% from a year ago and still climbing, and management flagged lumber as a modest headwind for fiscal 2027. Overhead is drifting the wrong way too. Selling, general and administrative expense rose 8% and now eats 8.3% of homebuilding revenue, up from 7.8%. Horton spent years planting infrastructure in new markets. With orders flat, all that footprint is a fixed cost waiting for a recovery to justify it.
Then there is the inventory math, which deserves more attention than it got. Horton ended the quarter with 38,000 homes in inventory, up from 29,600 at its September fiscal year-end. Some 23,300 are unsold and 7,600 are finished, though only 600 have been done for more than six months, so the stock is fresh. Here is the uncomfortable logic: when you get faster at building and no faster at selling, finished homes accumulate. Management sees it, started 23,900 homes in the quarter, and signaled fewer starts ahead. Smart inventory control. Also, unavoidably, a smaller pipeline of closings next year.
Built to Wait, Priced to Wait
If this downturn runs long, Horton is constructed for the siege. It controls 568,500 lots but owns just 22% of them; the rest sit under purchase contracts it can walk away from cheaply. Owned lots have fallen 14% since the fiscal year began. Two-thirds of its closings this year came on lots developed by Forestar or other third parties, which keeps capital out of dirt and options open. Liquidity stands at $6.1 billion, debt to capital at 23%, and the company still intends to spend roughly $2.5 billion on buybacks and $500 million on dividends this fiscal year. It handed shareholders $742.8 million this quarter alone.
The most honest number in the whole release might be the return profile. Homebuilding return on inventory is running at 17% and return on equity at 12.8%. Respectable, and better than most peers will manage in this tape. But management itself says current returns sit below its long-term expectations. Translation: the platform is fine. The era is not.
So what actually changed on Tuesday? The old bull case said Horton''s cheap-house focus would let it keep growing straight through the affordability squeeze. That case is now gone, retired by the company''s own guidance. The new bull case is narrower and more patient: Horton defends margins and returns better than anyone, waits out the drought, then converts its footprint into growth when buyers return. Everything in this quarter supported that case except the one thing it cannot supply, a date. Two-thirds of Horton''s mortgage closings go to first-time buyers, the customers most exposed to rates, jobs and nerves. They are the moat and the wound at the same time.
Horton is idling its growth engine on purpose, because revving it would torch margin for nothing. Respect the discipline. But a company this good at operating just told you, with a billion dollars of deleted revenue guidance, that the housing downturn has not finished clearing. The margin is proof of management. It is not proof of a bottom.