The 30-year fixed mortgage averaged 6.71% in the first week of September, up from 6.66% the week before and from 6.50% a year ago. The 15-year sits at 6.04%. Rates have been drifting higher all summer, pulled along by a long end of the Treasury curve that touched its highest level in nineteen years last month.

Existing home sales in July ran at an annual rate of 4.06 million, down 1.7% from June and up 0.7% from a year earlier. That is the striking part. A market that supposedly hinges on borrowing costs has now absorbed a full percentage point of rate volatility with sales essentially flat.

Stability at a very low level is not the same as health

Four million existing home sales a year is close to the weakest pace in three decades, in a country with considerably more households than it had the last time volumes were this low. The market is not stable because it is functioning well. It is stable because it has already emptied out.

The mechanism is lock-in. Millions of owners hold mortgages issued at rates far below today's, and moving means surrendering that rate. The financial penalty for selling is not the price of the next house alone, it is the higher monthly cost of financing it. So people stay, listings stay scarce, and the transaction count settles at a floor that further rate moves barely disturb.

Inventory in July was 1.54 million units, about 4.6 months of supply. That is tighter than the six months usually considered balanced, which is why prices have not fallen. The median existing home sold for $434,100, up 2.0% from a year earlier and the 37th consecutive month of annual gains.

Where the pressure actually shows up

Whitmore's Watchlist:
XHB (SPDR S&P Homebuilders ETF): Builders benefit from the resale freeze, since a scarce existing market pushes buyers toward new construction.
MBB (iShares MBS ETF): Agency mortgage-backed securities, where slower prepayment extends duration exactly when rates are rising.
REZ (iShares Residential and Multisector Real Estate ETF): Residential landlords, on the other side of a market where fewer renters can afford to buy.

A frozen resale market is not uniformly bad news. Homebuilders have spent two years taking share from a supply of existing homes that will not come to market, and they can do something no individual seller can: buy down a buyer's mortgage rate as part of the sale.

Mortgage investors face the opposite problem. When nobody refinances, the bonds backing those loans keep paying their original coupons for far longer than the models assumed. Duration extends precisely when rates are rising, which is the least convenient moment for it. That extension risk is a structural feature of this market rather than a temporary condition.

For households, the number that matters is not the rate but the payment. Median days on market held at 29 in July and first-time buyers made up 29% of sales, both figures showing a market where demand exists and affordability rations it.

What would actually unfreeze it

The obvious answer is lower rates, and it is worth being precise about what that requires. Mortgage rates track the 10-year Treasury plus a spread, not the federal funds rate. The Fed could cut and mortgages could still stay high if the long end refuses to follow, which is close to what happened during the last easing episode.

Supply is the other lever, and it moves even more slowly. Building enough homes to loosen a 4.6-month inventory takes years of permits, labour and land, and the same high rates that freeze buyers also raise the cost of financing construction. That is the loop the market has been stuck in: high rates suppress both the demand that would clear inventory and the supply that would ease prices.

The other answer is time. Lock-in weakens as life events accumulate: jobs move, families grow, people age out of houses. Every year at these rates converts a few more locked-in owners into sellers regardless of the arithmetic, which is why forecasters keep describing a gradual thaw rather than a turn.

Whitmore's Take: A housing market that ignores a percentage point of rate movement is telling you the binding constraint is not the rate. That distinction matters when deciding whether a housing-linked position is really a rate bet.

Written by Daniel Whitmore
Millionaire Insiders