MHO — Q3 2026 Financial Report Analysis
Q3 · Fiscal year 2026 · Published by Pham Hop
M/I Homes Q3 2026: net income fell 37.5% to $66.6M ($2.56/share) as mortgage rate buydowns and lot costs cut gross margin 3 points to 20.9%; orders slipped 2.5% and backlog value fell 8.9%.
At a glance
- Gross margin fell 3.0 points to 20.9% (homebuilding alone: 18.4%, down from 21.5%). That is the share of each sale left after paying for the land and the construction, and it is where almost all of the profit decline came from.
- Incentives cut revenue by $73.2 million, up from $53.5 million a year earlier. That is roughly $32,700 per home delivered versus about $23,300, and $23.2 million of the increase was mortgage rate buydowns alone.
- Backlog is down 6.5% in homes and 8.9% in dollars. M/I ended the quarter with 2,046 sold-but-not-yet-closed homes worth $1.10 billion, fewer than it delivered in the quarter itself (2,241), while net new orders slipped 2.5%.
- Revenue
- $1.1B
- -4.9% YoY
- Net income
- $67M
- -37.5% YoY
- Diluted EPS
- $2.56
- -34.7% YoY
- Operating margin
- 7.7%
This period vs a year ago
- Same period last year
- This period
Year-ago figures (≈) are worked out from the growth rate the company reported. Each row has its own scale.
Overview
M/I Homes' third quarter of 2026 (three months to September 30) showed what happens to a homebuilder when mortgage rates go back above 7%: it can keep selling roughly the same number of houses, but only by paying more to bring buyers' monthly payments down. Revenue fell 4.9% to $1.08 billion and net income fell 37.5% to $66.6 million, or $2.56 per diluted share against $3.92 a year earlier. Management's own explanation in the 10-Q is direct: profitability fell "primarily as a result of higher lot costs and the increased cost of mortgage interest rate buydown incentives due to the elevated mortgage interest rates," with rates that "exceeded 7% by the end of the quarter."
The numbers
| Metric | Q3 2026 | Q3 2025 | YoY change |
|---|---|---|---|
| Total revenue | $1,076.1M | $1,131.8M | -4.9% |
| Housing revenue | $1,031.4M | $1,095.4M | -5.8% |
| Gross margin (total) | 20.9% | 23.9% | -3.0 pts |
| Homebuilding gross margin | 18.4% | 21.5% | -3.1 pts |
| Operating margin | 7.7% | 12.0% | -4.3 pts |
| Net income | $66.6M | $106.5M | -37.5% |
| Diluted EPS | $2.56 | $3.92 | -34.7% |
| Homes delivered | 2,241 | 2,296 | -2.4% |
| Average sales price, homes delivered | $460,000 | $477,000 | -3.6% |
| New contracts (net) | 1,861 | 1,908 | -2.5% |
| Cancellation rate | 12.3% | 12.2% | +0.1 pts |
| Backlog (homes) | 2,046 | 2,189 | -6.5% |
| Backlog value | $1,103.0M | $1,210.8M | -8.9% |
| Active communities, end of period | 246 | 233 | +5.6% |
Operating margin here is revenue minus land and housing costs, inventory charges, selling and general and administrative expense, divided by revenue; it excludes net interest income ($3.6 million in Q3 2026, $4.5 million in Q3 2025).
Takeaway: M/I is holding volume by paying for it. Deliveries fell only 2.4%, but the price of each home fell 3.6% and incentives per home rose about 40%. With rates above 7% at quarter-end and the backlog smaller than one quarter's deliveries, the margin pressure is still building heading into Q4, not fading.
Why margins fell: buydowns first, then land
A mortgage rate buydown is when the builder pays the lender upfront so the buyer gets a lower interest rate, and with it a lower monthly payment. M/I books buydowns and closing-cost help as a cut to the selling price, so they show up directly in revenue and in the reported average price. In Q3 those incentives reduced revenue by $73.2 million, versus $53.5 million in Q3 2025.
The 10-Q breaks down the $43.7 million drop in homebuilding gross profit: the lower average price "included a $23.2 million increase in mortgage interest rate buydowns," plus "a $6.0 million increase in lot costs, a 2% decrease in homes delivered and $1.7 million in unusual warranty charges," partly offset by $2.9 million less in inventory charges (write-downs on land and homes that are worth less than their book value). The extra buydown spending alone equals about 2.2% of homebuilding revenue, so it accounts for most of the 3.1-point fall in homebuilding gross margin. Management also points to "mix of inventory homes delivered" — inventory homes (also called spec homes) are homes started before a buyer signed, and they are typically the ones builders discount most to move.
The two regions felt it differently:
| Region | Q3 2026 gross margin | Q3 2025 gross margin | New contracts YoY | Backlog homes YoY |
|---|---|---|---|---|
| Northern (Midwest, Mid-Atlantic) | 20.1% | 23.0% | +1% | -5% |
| Southern (Texas, Florida, Carolinas, Tennessee) | 16.9% | 20.2% | -4% | -8% |
In the North, the cause was cost: incentives plus an $8.1 million rise in lot costs, and incentive costs cut the average price by $13,800 per home. The South is weaker on demand. Orders fell 4% and the average price of homes in backlog dropped from $544,000 to $501,000. The 10-Q attributes the smaller backlog to "a shift in demand to inventory homes which offer incentives." Southern operating income fell from $62.9 million to $36.1 million. The South also carried $4.1 million of the quarter's $4.7 million inventory charges and the $1.7 million Florida warranty charge (attic ventilation problems in two communities, with $2.0 million still accrued).
Overhead made the drop steeper. Selling, general and administrative expense rose $6.4 million to 13.1% of revenue (from 11.9%), mainly from more model homes and sales offices as the community count grows, higher headcount and IT costs. Realtor and sales commissions fell $2.9 million.
What the headline numbers hide
- Orders vs. backlog: the warning sign is here, but mild. This is the pattern to watch at a homebuilder: orders shrinking while a big backlog hides it for a quarter or two. At M/I, net new orders fell 2.5% even though the company had 5.6% more active communities. A community is a subdivision where it is actively selling, so more communities should mean more sales. Sales per community fell to 2.6 a month from 2.7 (South: 2.6 from 2.9). Year-to-date orders are still up 5% (6,598 vs. 6,278), so this is a third-quarter slowdown, not a collapse. Cancellations did not get worse (12.3% vs. 12.2%).
- The backlog will not carry Q4 by itself. 2,046 homes in backlog is less than the 2,241 delivered in Q3. Q4 deliveries depend on selling and closing inventory homes within the quarter, and those are the homes that need the biggest incentives. Backlog homes are priced about 2.5% lower than a year ago ($539,000 vs. $553,000).
- Cash conversion improved, but only because the company spent less on land and construction. Operating cash flow for the nine months was $159.2 million, 75% of the $213.5 million net income, up from 43% a year earlier. The difference is mostly a smaller inventory build ($171.4 million vs. $314.5 million), not stronger earnings.
- Buybacks softened the EPS decline a little. Diluted shares fell 4.6% to 25.9 million. That is why EPS fell 34.7% while net income fell 37.5%. M/I spent $50.1 million on buybacks in Q3 and $150.2 million in nine months, about as much as all its operating cash flow, so cash fell $68.6 million to $620.6 million. The tax rate (23.4% vs. 23.8%) made almost no difference.
- One-offs are small. Inventory charges ($4.7 million) and the Florida warranty charge ($1.7 million) together cost about $6.4 million before tax, roughly 7% of pre-tax income. Excluding them would not change the picture. Interest income fell to $3.6 million from $4.5 million because the cash balance is smaller.
- The mortgage arm is steady. Financial services revenue fell 5% to $32.7 million on fewer loans and lower margins on loans sold. Operating income fell $3.9 million. M/I Financial's capture rate (the share of buyers with a mortgage who used M/I's own lender) rose to 96% from 93%, which matters because that is the channel the buydowns run through. After quarter-end, the company replaced its $200 million mortgage warehouse line with a $300 million one.
- The balance sheet is not a concern. Nothing was drawn on the $900 million credit facility ($808.9 million available after letters of credit). Homebuilding debt-to-capital was 18%, and book value reached a record $130 per share.
- No numerical guidance. The 10-Q gives no revenue or margin outlook. The one concrete target is year-end 2026 community count about 5% above 2025, which the 246 communities already track.
Outlook
Management's tone is cautious. It says conditions, "particularly the increasing mortgage interest rate environment, may persist through the remainder of 2026 and into 2027." Its stated plan is more incentives and tight spending: "the selective use of sales incentives, including mortgage interest rate buydowns, to improve affordability, support demand, increase sales activity, and reduce contract cancellations," alongside "disciplined overhead management" and "prudent land acquisition and development spending." Land spending is in fact down (land purchases $345.5 million vs. $363.1 million in nine months), and lots under control fell 3% to about 48,900, around five years' supply.
My read: a buydown's cost rises with the market rate. Rates were higher at the end of the quarter than the average for it, so the per-home incentive cost going into Q4 is probably higher than Q3's roughly $32,700. Unless rates ease, Q4 gross margin is more likely to stay near or below Q3's 20.9% than to recover. The things to watch in the Q4 report:
- Incentives per home. Is the revenue reduction for incentives still rising faster than deliveries?
- Sales per community in the South. It fell from 2.9 to 2.6 a month. If it keeps falling while community count grows, the extra communities are adding cost faster than sales.
- Inventory charges. They are small now ($8.9 million in nine months). A jump would signal that land bought at earlier prices no longer pays off at today's selling prices.
The balance sheet gives M/I room to keep buying sales this way for a while. The question is how much margin it gives up to do it.