Independent newsroom The Wyre News Network OpEd desk

Perspective 7 min read

110 Basis Points

The 30-year mortgage rate last week was 6.74%. With the same 10-year Treasury yield and 2023's spread conditions it would have been 7.84%. That 110 basis points is the difference between a housing market posting year-over-year gains and one posting none, and nobody in real estate produced a single point of it. Spreads are wide because the Fed stopped buying mortgage-backed securities and private investors want paying for the risk, which means the best thing that happened to housing this year is a bond-market condition that can be withdrawn without notice. Meanwhile the same weekly data set publishes a documented slowdown threshold at 6.64%, and four in ten listings took a price cut last week. One is a tripwire nobody has written down. The other is a campaign trigger almost nobody automates.

By Élise Tremblay for the Wyre OpEd desk

The 30-year mortgage rate last week was 6.74%. With the exact same 10-year Treasury yield, under the worst spread conditions of 2023, it would have been 7.84%.

That gap is 110 basis points. It is the difference between a housing market posting small year-over-year gains and one posting none at all. And not one basis point of it was produced by anything anyone in real estate did.

Logan Mohtashami, HousingWire's lead analyst, called mortgage spreads "our friendly neighborhood housing hero" in his August 9 tracker and suggested we all go hug one. He is not being cute. He is naming the single load-bearing variable of the year, and it lives in the bond market, several layers of abstraction away from anyone who has ever hosted an open house.

A mortgage rate is not one number

It is the 10-year Treasury yield plus a stack of spreads, and the stack is where the story is.

Fannie Mae's decomposition, published December 2024, splits that stack in two. The primary-secondary spread covers the cost of originating the loan: servicing, guarantee fees, lender cost and profit. It averaged roughly half a percentage point from 1995 to 2005 and about 1.01 points after 2008. The secondary spread is what investors demand for holding mortgage-backed securities instead of Treasuries. It averaged 1.17 points from 1995 to 2005, compressed to 0.71 points from 2012 to 2019, and ran about 1.4 points from January 2022 through November 2024.

Mohtashami tracks the whole spread differently and puts it at 2.01% last week, against a historical range of 1.60% to 1.80%. So by his measure spreads are still wide. They are simply not 2023 wide. In 2023 they crossed 3%, a level last seen in 1986, and that is the only reason mortgage rates ever printed an 8 handle.

Nobody decided to help housing

This is the part worth sitting with.

Fannie Mae is blunt about the cause: "the Federal Reserve's balance sheet actions are the main factor driving the difference in the secondary spread." When the Fed stops buying mortgage-backed securities and lets its holdings roll off, private investors have to absorb the supply. Private investors are rate-sensitive in a way the Fed never was. They price prepayment risk. They want more yield to take the paper. The spread widens.

What improved in 2026 is not a policy or a program. It is the absence of chaos. 2023 had a banking crisis and an aggressive hiking cycle detonating the spread on top of everything else. 2026, so far, does not. That is the entire mechanism behind the best housing news of the year.

An absence of chaos can be withdrawn without notice, and it does not send a calendar invite. Mohtashami notes the 10-year has been driven lately by the Iran conflict, with Fed hawks openly discussing hikes and the September meeting genuinely up in the air.

The tripwire nobody has written down

Here is the number that belongs on a whiteboard.

Mohtashami's tracker shows the same threshold three years running: housing slows when mortgage rates get above 6.64%. Not "when rates are high." At 6.64%.

Last week rates were 6.74%. Right on the wrong side of it. Purchase applications, which look 30 to 90 days ahead, have now printed two negative year-over-year weeks after a year of nearly uninterrupted growth. His 2026 running tally: 25 weeks of positive year-over-year prints, 10 of them double-digit, against 4 negative. Weekly pending sales still eked out a gain, 67,026 against 66,347 a year ago. Total pending sales are still up, 381,302 against 374,025.

Read that honestly and it says the market is slowing without yet producing a big negative number. That is a demand curve with a documented inflection point, published weekly, for free. Most marketing plans in this industry do not contain a single number like it.

What the rest of the week actually said

Inventory grew 0.78% year over year, which is not a wave. New listings ran 67,301 last week against 66,341 a year ago, and 2026 has cleared 80,000 in a week four separate times, better than recent years and still the low end of the 80,000 to 100,000 range that counts as normal. For scale, the bubble years ran 250,000 to 400,000 new listings a week.

And 41.44% of listings took a price cut. Against 42% a year ago. Roughly four in ten, in a market everyone agrees is tight.

That last one is not distress. It is the ordinary weekly condition of the business. It is also a campaign trigger, and almost nobody automates it.

The uncomfortable part

If demand this year is being propped up by a spread, then a meaningful share of 2026 production was not earned. It was borrowed from a bond market that is going to want it back.

That is not an insult to anyone's hustle. The listings were still won, the deals were still closed, the weekends were still worked. But the reason those deals were available to be closed at all is that a technical condition in the mortgage-backed securities market held the rate 110 basis points below where a worse year would have put it. Nobody in the industry voted for that, negotiated it, or can renew it.

There are about 1.44 million NAR members as of late June 2026. When the spread hands the market back, they will not be sorted by who had the best market. They will be sorted by who built a cost of acquisition that survives a worse one.

The playbook

1. Put 6.64% on the wall

Mohtashami has handed the industry a free leading indicator and published it weekly for years. Rates above 6.64% have preceded a slowdown three years running. That is a budget trigger, a staffing trigger, and a creative-rotation trigger. Write it down, decide now what happens when it trips, and stop rediscovering it every cycle.

2. Automate the price cut

Four in ten listings get a reduction. In most shops that is a phone call, a manual edit, and a new photo if anyone remembers. It should be an event: creative swaps, budget re-weights, the "improved price" campaign launches itself, and the old creative stops serving the same hour. If your systems cannot react to a status change without a person noticing it first, the price cut is not the problem.

3. Measure cost per closed transaction, not cost per lead

Cost per lead is the number vendors quote because it is the number that flatters them. It is also the number that moves the least when the market turns. When volume contracts, cost per lead can hold perfectly steady while cost per closing doubles. Only one of those two numbers tells you whether you can still afford your marketing next year.

4. Stop paying for demand you were going to get anyway

A rate-driven market produces buyers who were always going to transact. Spend attributed to them looks spectacular and proves nothing. The honest question is what your programs produce at the margin, and the honest way to find out is to turn something off deliberately and watch what happens, while the market is still forgiving enough to absorb the answer.

5. Build the machine while the market is kind

Governance, tracking, creative pipelines and automation all get built badly under pressure and cheaply under calm. Right now is calm. The spread is buying an interval, and intervals are for building.

6. Watch the spread, not the rate

The rate is the output. The spread is the mechanism, and it moves first. Anyone whose plan reacts to the mortgage rate is reacting to a lagging indicator of their own business.

Our position

Mohtashami's joke is the right instinct and the wrong conclusion. Hug the spread, certainly. Then go and act like it is leaving.

Every market has a hero nobody hired. In 2023 the same variable was the villain and took mortgage rates to 8%. It is the identical mechanism in both years, pointed in opposite directions, and at no point did anybody in real estate get a vote.

What you get a vote on is the machine. Whether a price cut is an event or a chore. Whether you know your cost per closing or only your cost per lead. Whether the tripwire is on the wall or in somebody's memory. None of that is glamorous, and all of it is yours, and it is the only part of this that still works at 7.84%.

Sources

Logan Mohtashami, HousingWire, "Mortgage spreads keeping housing demand intact for now" (2026-08-09), for the 6.74% rate, the 7.84% / 7.46% / 7.27% counterfactuals, spreads at 2.01% against a 1.60% to 1.80% historical range, the 2023 move above 3% and the 1986 comparison, the 6.64% threshold, weekly and total pending sales, the purchase application tally, inventory, new listings, and the price-cut share.

Fannie Mae Housing Insights, "What Determines the Rate on a 30-Year Mortgage?" (analysis dated 2024-12-11), for the two-spread decomposition, the historical spread averages, and the quoted finding that Federal Reserve balance sheet actions are the main factor driving the secondary spread.

National Association of REALTORS membership, approximately 1.44 million as of late June 2026.

A note on what is not here. Every real estate cost-per-lead benchmark we located in research traced back to SEO content mills rather than a primary source, and the widely repeated agent-attrition figures could not be reconciled with NAR's own reporting. Both were left out rather than hedged. We have also deliberately omitted a peak-membership figure, because the sources disagree and we could not settle it.

Related Reading

More Opinion

From the same desk

Perspective

A Rounding Error

Over twenty-four hours the single largest AI crawler on one consultant's website arrived roughly 1,500 times under a nonprofit research archive's name, sent back nothing, and asked for his SSH keys, his AWS config and his Firebase service account key. Across a hundred such paths: 1,028 requests, 6.7 megabytes, zero referrals, and requests for anything he had actually written rounding to nothing. It appeared in no security log, because security logs record rule trips and he was not blocking it. It appeared in exactly one place: the AI crawler dashboard, counted as an audience. Cloudflare's CFO says humans will be a rounding error on the internet within five years and admits he has been wrong every time, always low. Machine traffic passed human traffic in May 2026, two years earlier than the company forecast. Why robots.txt is a request that OpenAI's own documentation says may not apply, why blocking the obvious way costs you the visibility and keeps the fetches, what it means that agent config files are now on the standard secret-scanning wordlist, and the afternoon of work almost no company has done.

13 min

Perspective

Cut The Price Last

When demand softens, the first lever everybody reaches for is price. It is also the most expensive lever on the board, the hardest to reverse, and almost never the one that was actually broken. Redfin reports buyers have negotiating power across most of the country, the median housing payment fell to $2,575 because sellers cut asking prices to a one-year low, and pending sales hit their lowest since early April. Underneath it, July payrolls came in at minus 23,000 against an expected plus 83,000 with another 103,000 revised away. So the softening is real, and the question is which lever you pull and in what order. Why cutting price assumes a diagnosis you have usually not earned, the cheaper levers ranked, why the gap between the decision and the execution is where the margin leaks, and the measurement problem underneath all of it: cost per lead holds beautifully steady while cost per closed sale doubles.

6 min

Perspective

The Layer Nobody Looks At

A man in Connecticut typed instructions into his own court filing in three-point white text on a white background, where no judge would see them and any software would. They told the system to agree with him and to disregard the court's earlier denials. He was caught, warned, did it again, and lost his e-filing privileges. In the same seven days Google made the visible watermark on its AI images optional while keeping the invisible one, Anthropic shipped an API letting third parties detect text its models probably wrote, a new benchmark found that no frontier model clears sixty percent on basic visual perception with counting, localization and OCR among the weakest skills, and OpenAI started turning Mac users' clicks and keystrokes into a searchable memory their assistant can read, while warning in its own documentation about prompt injection. Five stories, one story. Every document, image and screen you produce now has two readers and they do not see the same thing. What the judge worked out on his own about where these attacks get caught, why provenance just moved into the layer humans cannot check, what a watermark hit actually proves and why a clean result proves nothing at all, and the ten-line diff almost nobody runs.

18 min