Rates Hit 7.12%. Your Lender Skips the ARM Conversation.

Why Aren't More Buyers Being Offered an ARM Right Now?

The 30-year fixed rate hit 7.12% for the week ending September 18, 2026, the highest level since May 2024, according to the Mortgage Bankers Association's weekly survey. In that same report, the 5/1 ARM rate sat at 6.10%, more than a full point lower, and adjustable-rate applications climbed to 9.8% of total mortgage volume. Most buyers never hear that the option exists, because the loan officer handling their referral has no reason to spend the extra thirty minutes explaining it.

That's the real gap right now. It isn't a rate gap. It's an attention gap.

The Math Nobody's Making for Your Buyers

Walk through what a full point actually does. On a $450,000 loan, the difference between 7.12% and 6.10% runs somewhere around $330 a month, close to $4,000 a year, for the first five years of the loan. That is not a rounding error. That's a buyer who got priced out at 7.12% and qualifies comfortably at 6.10%, or a buyer who can now afford the house two streets over instead of settling for the smaller one.

The data on the ground backs this up. NAR's existing-home sales figures put inventory at 4.9 months of supply in August, the highest level since November 2015, and first-time buyers made up 30% of transactions, the largest share in over a year. There's more product sitting on the market and more buyers who need every basis point they can get to make the numbers work. A well-structured ARM conversation is one of the few levers left that can move a "not right now" buyer into a "let's write the offer" buyer this quarter, the same way we broke down the builder buydown structures that are quietly winning buyer-side deals right now.

Run the same math on a $650,000 loan, which is closer to what a lot of move-up buyers are financing this fall. The gap between 7.12% and 6.10% works out to roughly $475 a month, or about $5,700 a year, for the first five years. For a buyer who's already stretched, that's the difference between a debt-to-income ratio that clears underwriting and one that doesn't. It's the difference between waiting another two quarters for rates to move and writing an offer this weekend.

Here's the problem. Most of the loan officers your agents refer buyers to are not having that conversation. They quote the 30-year fixed because everybody already understands it, it closes faster, and it doesn't require walking a nervous first-time buyer through rate caps, adjustment periods, and refinance timing. An ARM takes more time to originate and more time to explain. Time is the one thing a referral LO with forty other files doesn't have for a deal they may never see again.

Why Your Referral Lender Won't Make This Call

This isn't a knock on individual loan officers. It's math. Marina Walsh at the MBA put it plainly earlier this year: the industry has never gone through such a long stretch of compressed margins, with average net production profit down to 25 basis points in the second quarter, compared to 89 basis points back in early 2021. When a lender's margin per loan has dropped by roughly two-thirds, the only way to stay profitable is volume, and volume rewards the fastest, most standardized product on the shelf. That's the 30-year fixed. It is not a custom ARM structure built around one buyer's five-year plan.

Layer on top of that the fact that a referral LO usually has no relationship with your team beyond that single file. They didn't build your buyer's pipeline, they don't know the buyer's timeline, and they have no real incentive to invest extra structuring time in a transaction that came from a loose sponsorship instead of an actual partnership. We wrote about this exact dynamic in why the reactive loan officer model is dying: a referral relationship gets you a name on a list, not a producer who treats your pipeline like their own.

What an Embedded LO Seat Actually Fixes

This is exactly the gap an embedded loan officer is built to close. A dedicated LO who sits inside your team's operation, who sees your buyer leads before they ever become a referral, has a completely different incentive structure. Your volume is their volume. Structuring an ARM for a buyer five years from a planned move, building a temporary buydown into a builder deal, or running the numbers on a fixed option next to an adjustable one side by side isn't extra work that eats into their margin. It's the job.

I've watched this play out with teams doing 150 to 600-plus transactions a year. The ones giving away buyer-side lending to outside LOs are also the ones whose buyers get the generic quote and the generic product, every time. The ones with an embedded seat are the ones whose loan officer can say "here's what a 5/1 ARM does to your payment for the next five years, here's what happens if you refinance when rates come down, here's what happens if they don't," and mean it, because they've run that exact math for the last twenty buyers on that team too. It's the same reason 76% of buyers end up with whatever lender their agent recommends: that recommendation is worth almost nothing if the lender on the other end isn't built to earn it.

Most teams never build that seat because they still think of mortgage as something to refer out, not something to own. Every point rates climb makes that a more expensive habit. This week's 7.12% print is one more data point that the fixed-rate-only conversation is leaving real, closeable buyers on the table, and with the loan officer pool shrinking industry-wide, the outside lenders left standing have even less bandwidth to make that call for you.

How to Get This Conversation Happening on Your Team This Month

You don't need to overhaul your whole lending model to fix this. You need three things in place, in this order.

First, pull your last 90 days of buyer files and check how many actually got shown more than one rate structure. If the answer is close to zero, you already have your baseline problem, and it has nothing to do with the market.

Second, ask whoever is handling your buyer-side lending right now, in-house or referred, how they decide when to bring up an ARM or a temporary buydown instead of defaulting to the 30-year fixed. If the honest answer is "we don't really," that's not a training gap. That's an incentive gap, and training won't fix it.

Third, run the numbers on what it actually costs you to keep routing buyer-side business to a lender who has no reason to work that hard for a one-off file, versus what it costs to build or bring on a dedicated seat whose comp is tied to how well your pipeline converts, not how fast they can churn a standard file. Most teams have never run that comparison with real numbers.

Rates crossing 7.12% isn't the real story here. The story is how much buyer-side business is sitting untouched because nobody on the deal has the incentive to make the ARM call, the buydown call, or the creative-financing call that actually gets a hesitant buyer to the table. That's exactly the conversation we have on a partnership call: how to build an embedded lending seat that treats your pipeline like its own, not a referral to hand off and hope for the best.