We closed the week with the 30-year fixed sitting at 6.65% and the 10-year Treasury at 4.74%, which is roughly four basis points higher than where the benchmark sat when Wednesday's FOMC statement crossed the wire. Freddie Mac's Primary Mortgage Market Survey printed 6.66% on Thursday, up eight basis points from 6.58% the prior week, and the 15-year moved to 6.04% from 5.96%. The interesting wrinkle is the compression at the short end: the 5/1 ARM came in around 6.43%, roughly 22 basis points inside the 30-year, after gapping out above the fixed rate two weeks ago when lenders pulled short-duration pricing ahead of the Fed. That relationship has now normalized, and it matters, because a positively sloped rate sheet is the difference between an ARM being a real tool and an ARM being a rate sheet decoration.
The macro picture that produced those numbers is genuinely split. The Fed held at 3.50% to 3.75% on Wednesday for the fifth consecutive meeting, but the vote was 9 to 3 with Hammack, Kashkari and Logan all dissenting in favor of a quarter-point increase. That is a hawkish dissent bloc, not a dovish one, and it is the reason the long end has refused to rally despite softening labor data. Thursday's June core PCE then landed at 3.3% year over year, down from 3.4% in May, with headline PCE at 3.7% versus 4.1% prior. Inflation is decelerating, but it is decelerating from a level that is still 130 basis points above target, and the median 2026 dot now clusters between 3.6% and 4.1%, which leaves the door open to a hike rather than a cut. Meanwhile June payrolls came in at just 57,000 against a 115,000 consensus with April and May revised down a combined 74,000. That is a labor market losing altitude while inflation is still elevated, and bond traders have no clean way to price it.
For brokers, the practical takeaway heading into this week is that the entire rate story now runs through Friday's July employment report. If payrolls confirm the June weakness, the 10-year likely breaks lower and rate sheets improve into the following Monday. If payrolls surprise to the upside, the three hawkish dissenters get validated and you should expect the 30-year to test 6.80% quickly. Either way you are looking at a week where your locked pipeline is fine and your floating pipeline is a coin flip, so the conversation to have with borrowers before Friday is about their own tolerance, not your forecast. MBA application volume already fell 6.4% for the week ending July 24, so the marginal borrower is clearly rate-sensitive right now, and that is exactly the borrower who benefits from a non-QM structure that gets qualified on something other than a payment shock.