Stewart's New Lower Bar: When the Share-Gain Machine Meets a 7.5% Mortgage

For roughly three years, Fred Eppinger has sold Stewart Information Services as a share-gainer that could outrun a broken housing market. In the third quarter of 2026, the housing market finally outran Stewart.

The mechanics were simple and brutal. The 30-year mortgage pushed to roughly 7.5% by October, buyers "very quickly move to the sidelines" when rates clear 7%, and existing home sales printed below 4 million for two of the quarter's three months. Adjusted EPS came in at $1.33 against $1.64 a year earlier; net income fell to $30 million from $44 million. Year-to-date revenue growth of 22% decelerated to 13% in the quarter. The stock has already voted — down 25.5% over ninety days and 31.8% below its July high of $72.23.

What makes this report more than a rate victim story is the language shift. residential market is now the single loudest theme in Stewart's own keyword set — it simply was not before, not with this force. And the CEO's framing has changed from offense to defense: “the market began to set a new lower bar. For residential housing.” — Frederick Henry Eppinger Jr., Chief Executive Officer · 2026-10-22

The guidance that didn't survive the quarter

Three months ago the tone was entirely different. On the July call, Eppinger framed the year with a specific promise and a confident verdict on his own execution:

I believe that we'll grow revenue probably 20% and earnings 30%. That's kind of the range, I think... I'm right on track. It's right where I wanted us to be.

Instead, earnings went backwards on a 13% revenue quarter. That is the tension in this report: a company whose strategic narrative (share shift, commercial doubling, Real Estate Solutions scale) was validated through the first half, now meeting a volume environment where the operating leverage runs the wrong way. Stewart's operating margin of 6.1% sits less than half its 2021Q4 peak of 13.1%, and the company carries 500 locations of fixed cost against 4 million existing home sales.

Commercial: from engine to question mark

The one business that has carried Stewart for eight quarters — commercial, which doubled in two years and grew 40-50% at times — slowed to 9% overall growth. Closed domestic commercial orders actually rose 18%, but the average fee per file fell 8% to $16.3 thousand because the mix shifted away from large projects. Note which keyword fell hardest in Stewart's own lexicon this quarter: data centers, previously a growth boast, now a headwind to deal size, alongside energy.

There was also a timing hit, and management was unusually candid about it: “it is multiple millions that pushed. So it was meaningful. To the quarter.” — Frederick Henry Eppinger Jr., Chief Executive Officer · 2026-10-22 This is where the prior quarters' vocabulary becomes prophecy. bumpy and big deal were the top two themes on the July call; a year earlier Eppinger had already warned “do I think we're going to sustain 45% growth? No.” — Frederick Henry Eppinger, Chief Executive Officer (CEO) · 2025-07-24 The lumpiness he described — deals that slip a quarter on financing or carrying costs — is now the reported number rather than a caveat.

To his credit, he did not retreat on the pipeline (record open commercial orders, no data-center slowdown yet), and he flagged something the market hasn't priced: chatter of a slowdown in small commercial that has not yet shown up in Stewart's own books.

What management stopped saying

The most telling change is the verb. Through the spring and summer, the story was reinvestment — “we've reinvested a bunch of that because I want to sustain it.” — Fred Eppinger, Chief Executive Officer (CEO) · 2026-07-23 This quarter: “we will be very focused on managing our business investments and expenses.” — Frederick Henry Eppinger Jr., Chief Executive Officer · 2026-10-22 That is a pivot in posture, and it comes with a pledge to sharpen M&A criteria so deals "fit" the long-term model.

Ironically, the promised capital deployment finally happened — excess capital went to work on Nationwide Appraisal Network, ProTitleUSA, Tower Title and Rattikin, roughly $150 million at 4-6x EBITDA. The balance sheet shows it: effective net cash of $282 million, down from $610 million a year earlier. The $300 million above statutory reserve requirements that David Hisey cited still leaves room, and the seventh dividend raise in under six years ($2.10 to $2.20) is a real signal of confidence.

Two other themes got quieter. The title-waiver pilot that consumed whole Q&A sessions in 2025 is now dismissed as “kind of old news” — Frederick Henry Eppinger Jr., Chief Executive Officer · 2026-10-22 — defensible, given Stewart's minimal exposure to the refi market, but a notable de-escalation of a bear case that once moved the whole group. And Real Estate Solutions margin guidance wobbled within a single answer: low-to-mid teens in two to three quarters, then “for the next couple of quarters, we are going to see... kind of at this level of margin.” — Frederick Henry Eppinger Jr., Chief Executive Officer · 2026-10-22 The two statements aren't identical, and the market will notice the gap.

AI, meanwhile, is the one place management got more expansive rather than less: from roughly 75 initiatives in February to more than 100 now, with about 50 already "penciling" returns — but with an explicit 18-month lag before it moves the P&L. That is a cost story disguised as a technology story, and it matters most in exactly the scenario Stewart is now in.

The wrong side of a market-wide tide

Stewart's problem is not idiosyncratic. The global keyword board for this quarter is dominated by Fed rate hike expectations and the mortgage universe — and the regional banks reporting alongside Stewart (SSB, BWB, HWC, SMBK) keep referencing a September hike and the resulting deposit and loan repricing. Homebuilders are describing the same demand air-pocket in different words: lower absorption pace, weaker capture rates. The housing complex is collectively re-pricing a higher-for-longer curve, and Stewart simply has the most operating leverage to the transaction count.

The valuation has partly caught up: 13.4x price to net income, down 25% year-over-year, and 0.6x revenue. Note the fundamentals here are Q2 (period ending July 28), so they understate the Q3 deterioration.

The bull case is not silly: a dividend raise, $300 million of deployable capital, a scaled credit-and-appraisal franchise serving the top-300 lenders, and a commercial pipeline management insists is at record levels. The bear case is arithmetic: every quarter that 30-year stays near 7.5%, the fixed cost base grinds the margin. Eppinger's own summary is the fairest one — “all we need is a normalized market. I do not need it to go all the way back. I just need it to stabilize.” — Frederick Henry Eppinger Jr., Chief Executive Officer · 2026-10-22 That is a modest ask. In this rate environment, it is also the whole thesis.