We told you the July minutes would answer one question: were the three dissenters outliers, or the visible edge of something bigger? We got our answer, and it wasn't the friendly one. The minutes, released Wednesday at 2pm, showed that many policymakers believe further rate increases could become necessary if inflation fails to moderate. Not three. Many.

Remember the setup: the July 29 vote was 9–3 to hold at 3.50–3.75%, with Hammack, Kashkari, and Logan all dissenting for a quarter-point hike — the first three-way dissent in one direction since September 2016, and the closest this committee has come to raising rates in a decade. The minutes tell us the hold was more grudging than the vote implied. There was also a genuinely unusual institutional note: Warsh raised the possibility of cutting the FOMC's calendar from eight meetings a year to six, arguing that meeting roughly every two months would let more data accumulate between decisions. No decision was made and 2026's schedule is unchanged — but it tells you how this chair thinks about policy: less frequent, less telegraphed, more deliberate.

The 10-year Treasury sits near 4.70% this morning, holding above its 50-day average around 4.60% after ticking up to 4.708% Thursday on an oil rally. That yield is the foundation of your mortgage rate — and it explains why two calm inflation reports and a shrinking job market haven't delivered the relief everyone expected.

But there's a second layer almost nobody discusses, and it's worth more to you than the next Fed meeting. Your mortgage rate isn't the 10-year — it's the 10-year plus a spread. Historically that spread runs about 1.5 percentage points. Right now it's running roughly 1.98 points. Do the arithmetic on that: at a normal spread, a 4.70% 10-year would produce a 30-year fixed near 6.20% instead of the ~6.65% we're actually looking at. There is close to half a percentage point sitting in the spread itself — not in the Fed's policy rate, not in inflation, but in the premium investors charge for taking mortgage risk in a volatile market.

Why does that matter practically? Because it's a second, independent path to lower rates that doesn't require the Fed to do anything. Spreads widen when volatility is high and the policy path is murky; they compress when things calm down. We don't need a rate cut to get to the low 6s — we need boring. A quieter Gulf, a predictable inflation trend, and a Fed that markets feel they understand would each compress that premium. On a $400,000 loan, closing half of that gap is worth roughly $60 a month, forever, with no help from Washington at all.

Freddie Mac's weekly survey came in at 6.65%, down two basis points — the third straight weekly decline, easing from 6.69% to 6.67% to 6.65%. It's real relief, and it's also modest: a year ago the same survey read 6.58%, so we're still seven basis points worse than last August despite everything the data has done for us since. The 15-year at 5.95% is the better story, though it too sits above last year's 5.69%.

Let's be straight about what we can and can't know. Nobody has a forecast worth betting your closing on. But the mechanics are knowable, and they sort into three paths:

Path	What Has to Happen	Where the 30-Yr Lands
The Grind most likely	Inflation keeps cooling slowly, the Fed holds in September, oil stays where it is. Nothing breaks, nothing resolves.	Mid-6s into the fall — roughly where we've lived all summer, drifting a few basis points a week
The Compression the good case	PCE confirms the cooling, Warsh sounds measured at Jackson Hole, volatility falls and the spread narrows toward normal — no Fed cut required.	Low 6s, potentially a 5-handle on the 15-year. This is the realistic upside
The Re-Rate the risk	PCE runs hot, energy spikes again, and the hawkish bloc in those minutes wins the September argument.	Back toward 7% — the same level that stalled this market in July

Notice what's not in that table: a scenario where the Fed cuts and rates fall a full point. That isn't on the menu this year. The realistic upside runs through calm — through the spread compressing and the term premium easing — not through a dramatic policy rescue. Which is why we keep saying the same unglamorous thing: control what you can control, and don't build a plan around a forecast.

Date	Event	Why It Matters
Wed Aug 26 8:30am ET	July PCE — the Fed's preferred gauge (BEA-verified)	Does it confirm the CPI's cooling, or give the hawks their evidence?
Fri Aug 28 morning	Warsh's first Jackson Hole keynote as Chair	With guidance abolished, a set-piece speech carries unusual weight. The tone is the guidance
Fri Sep 4	August Jobs Report	Does July's −23K become a trend? Flash edition
Fri Sep 11	August CPI · BLS-verified	Last inflation print before the decision · Flash edition
Sep 15–16	FOMC + fresh dot plot	First projections since June, when 9 of 18 penciled in a hike by year-end

Calendar correction: our last issue listed July PCE for Tuesday, August 25. The BEA's own release schedule puts it on Wednesday, August 26 at 8:30 a.m. ET. We verify every date against the primary source before printing it, and we flag it when we miss.

Closing in the next 45 days: lock, and lock before Wednesday. PCE lands the 26th and Warsh speaks the 28th — two events that can move the long end in either direction inside one week, with no forward guidance to soften the landing. Pair it with a float-down so the compression path still works for you. Locking into a week like this isn't pessimism; it's just refusing to gamble with a closing date.

Refinancing: your window is the compression scenario, not a Fed cut. Watch two numbers rather than the headlines — the 10-year and your own quoted rate. If the 10-year breaks meaningfully below 4.60% and stays there, that's your signal to move fast. Have your documents staged now so "fast" is actually possible.

Everyone: the single most valuable thing you can do this month has nothing to do with the Fed. Get multiple Loan Estimates on the same day, and compare them line by line. With spreads this wide, lender pricing varies more than the market does. That's a gap you can close yourself — this week, without permission from anybody in Washington or Wyoming.

The minutes told us the hawks are more numerous than the vote showed. The 10-year told us it doesn't care about two good inflation reports. And the spread told us something more useful than either: roughly half a point of your mortgage rate is being charged for uncertainty, not for inflation or Fed policy. That's the number we'd watch this fall, because it can improve without a single vote being cast.

Freddie's 30-year has now fallen three weeks running — 6.69, 6.67, 6.65. It's slow, it's unglamorous, and it's still worse than a year ago. But the direction is finally right, and the next eight days — PCE Wednesday, Warsh Friday — will tell us whether it continues or reverses. We'll be here for both.

Have a great weekend. And if you're closing soon, let's talk before Wednesday — that's what I'm here for. — Ryan