M/I Homes vs. PulteGroup Q3 2026: Volume vs. Margin as Rates Top 7%

M/I Homes (MHO) and PulteGroup (PHM) reported the same three months, July to September 2026, into the same market: mortgage rates climbing back above 7%. Both sold fewer homes and earned less. But the way each one absorbed the hit was different. M/I kept its deliveries close to last year's and paid for that with incentives, so its profit margin took the damage. PulteGroup held its margin near 25% and let volume fall instead. This comparison uses only the figures in our two published reports: M/I Homes Q3 2026 and PulteGroup Q3 2026.

At a glance

• Profit fell twice as hard at M/I. Net income dropped 37.5% at M/I Homes versus 18.8% at PulteGroup, even though M/I's revenue fell less (−4.9% vs −9.8%).
• The gross margin gap is wide. M/I's homebuilding gross margin was 18.4% (down 3.1 points); PulteGroup's home sale gross margin was 24.8% (down 1.4 points, about 0.7 of which was one land write-down).
• The backlog story is reversed. M/I's backlog shrank 8.9% in dollars and now holds fewer homes than it delivered in the quarter; PulteGroup's backlog is still 3.4% larger in dollars than a year ago, though it fell since June.

The numbers side by side

*The two margins are not measured the same way. M/I's is operating income before net interest income; PulteGroup reports no operating income line, so our PHM report uses income before taxes divided by revenue. Treat the gap as directional, not exact. **M/I: homebuilding gross margin (its total gross margin, including the mortgage arm, was 20.9%, from 23.9%). PulteGroup: home sale gross margin. ***M/I: SG&A as a share of total revenue; PulteGroup: as a share of home sale revenue.

EPS is close ($2.56 vs $2.53) only because PulteGroup has far more shares; it says nothing about which company is more profitable.

Two answers to the same rate problem

A homebuilder facing higher mortgage rates has two levers. It can cut the effective price, usually through a mortgage rate buydown (the builder pays the lender upfront so the buyer gets a lower rate and a lower monthly payment), and keep sales moving. Or it can hold price and accept selling fewer homes. These two quarters show one company leaning on each.

M/I Homes chose volume. Deliveries fell only 2.4%, but incentives cut revenue by $73.2 million, up from $53.5 million a year earlier, or roughly $32,700 per home versus about $23,300. M/I's 10-Q says profitability fell "primarily as a result of higher lot costs and the increased cost of mortgage interest rate buydown incentives," and breaks out a $23.2 million increase in buydowns alone, plus $6.0 million more in lot costs. Our M/I report estimates the extra buydown spending by itself accounts for about 2.2 points of the 3.1-point fall in homebuilding gross margin. The average price of a delivered home fell 3.6%.

PulteGroup let volume go. Closings fell 8.7% while the average selling price was essentially flat ($561,000 vs $564,000). The revenue drop is a volume story, not a price story. Its gross margin fell 1.4 points, and the release attributes about 0.7 points of that to a $26 million land impairment (a write-down of land worth less than its book value). Excluding it, our PHM report puts the margin at about 25.5%, slightly above Q2's 25.0%. One caveat: PulteGroup's report was written from the earnings release, before the 10-Q, so it does not split the remaining margin decline between incentives and land costs the way M/I's 10-Q does.

Why PulteGroup could hold price: its CEO says "more than 60% of our business" is now move-up and active-adult (Del Webb) buyers, who buy pricier homes and depend less on mortgage rates than first-time buyers. PulteGroup's orders fell 6.5% in units but only 3.9% in dollars, so the average new contract was worth more than a year ago (about $564,000 vs $548,000, a calculation in our PHM report). M/I's buyers look more rate-sensitive: its Southern region's backlog price fell from $544,000 to $501,000, and its 10-Q describes "a shift in demand to inventory homes which offer incentives" (homes built before a buyer signed, the ones builders discount most).

Where the slowdown shows up

Both companies opened more communities (subdivisions actively selling) and sold fewer homes from each one.

• M/I: 5.6% more communities, orders down 2.5%. Sales per community fell to 2.6 a month from 2.7, and in the South to 2.6 from 2.9. Year-to-date orders are still up 5%.
• PulteGroup: 7% more communities, orders down 6.5%, so each community sold roughly 12% fewer homes. Year-to-date orders are up 1.4%, after a 6% rise in Q2 turned into a 6% fall in Q3. Florida was its only region with higher orders (+5.9%); the Southeast, West and Northeast each fell about 15%.

On order trends alone, PulteGroup's quarter was weaker. The difference is in what comes next. M/I ended September with 2,046 homes in backlog, fewer than the 2,241 it delivered in Q3, so Q4 depends on selling and closing inventory homes within the quarter, which are the homes that need the biggest incentives. PulteGroup's 10,298-home backlog is still 4.1% larger than a year ago, which our PHM report reads as support for Q4 closings comparing better than Q3's, though it shrank by about 670 homes since June.

Both carried heavier overhead relative to sales, for the same reason: more communities to staff and run on fewer sales. M/I's SG&A (selling, general and administrative costs) rose to 13.1% of revenue from 11.9%; PulteGroup's rose to 10.3% of home sale revenue from 9.4%.

Cash and buybacks: opposite directions

Both companies cushioned EPS by buying back stock. M/I's diluted share count fell 4.6% and PulteGroup's 5.2%, which is why EPS fell less than net income at each.

How they paid for it differs:

• M/I converted more of its profit into cash: nine-month operating cash flow was $159.2 million, 75% of net income (43% a year earlier), mainly because it spent less building inventory. Its $150.2 million of nine-month buybacks roughly matched that cash flow; cash fell $68.6 million to $620.6 million. Nothing was drawn on its $900 million credit facility and homebuilding debt-to-capital was 18%.
• PulteGroup went the other way. Nine-month operating cash flow was $637 million against $1,295 million of net income, under half, because cash spent on inventories nearly doubled to $1.15 billion. It spent $1.03 billion on buybacks, more than its operating cash flow, issued $795 million of new notes while repaying $613 million, and moved from net cash of $378 million at year-end to net debt of $401 million (debt-to-capital 12.1%, still low).

PulteGroup also got an unexplained lift from "other income, net" ($19.7 million vs $1.8 million); without it, our PHM report estimates pre-tax income would have fallen about 21% rather than 19%. M/I's one-offs were small: $4.7 million of inventory charges and a $1.7 million Florida warranty charge, roughly 7% of pre-tax income.

What each is watching into Q4

Neither company gave numerical guidance.

• M/I says the rate environment "may persist through the remainder of 2026 and into 2027" and plans "the selective use of sales incentives, including mortgage interest rate buydowns." Our M/I report expects Q4 gross margin to stay near or below Q3's 20.9% unless rates ease, because buydowns cost more as market rates rise and rates were higher at quarter-end than the quarter's average.
• PulteGroup says it will "balance the price and pace of home sales, align production accordingly," which suggests fewer homes started if orders stay weak. The open questions are its cancellation rate and how much of the margin decline is incentives, both due in the 10-Q.

Takeaway: PulteGroup had the stronger quarter on profitability: its gross margin fell 1.4 points to 24.8% against M/I's 3.1-point drop to 18.4%, and its net income fell half as much (−18.8% vs −37.5%), because a buyer base tilted to move-up and active-adult customers let it hold price while M/I paid rising buydowns to keep volume. But PulteGroup's demand weakened more (orders −6.5% vs −2.5%) and it is funding buybacks partly with debt, while M/I has a cleaner balance sheet and better cash conversion. For Q4, M/I's thin backlog is the bigger risk to margins; for PulteGroup, it is whether orders keep falling while inventory spending stays high.

Figures come from our two published reports, which are based on M/I Homes' 10-Q and PulteGroup's earnings release (Exhibit 99.1) for the quarter ended September 30, 2026. This comparison is for information only and is not investment advice.