Conventional about 7.36% — a basis point of drift on Fed-decision eve
Tuesday morning, the day before the Fed, and the number barely moved. Here it is source by source so you can audit it instead of taking my word for it. Bankrate's national average, time-stamped 6:30am Eastern this morning, reads 7.38% on the 30-year fixed — up 2 basis points from yesterday's 7.36%, and still down 9 basis points from a week ago. Mortgage News Daily posted its Monday, October 26 close at 7.33%, unchanged on the session, with no new print until this afternoon. The Mortgage Reports' panel carries 7.359%. Average those three and you get 7.356%, so I'm carrying the headline at about 7.36% — one basis point off where I had it yesterday. Real talk: that is not a move. That is noise, and it doesn't touch your payment. Now the honest wrinkle I'd rather hand you than hide, because it's the same one as yesterday and it hasn't closed. There's a fourth panel reading materially lower. Mortgage Research Center — the data behind Fortune's Tuesday figures — has conventional at 7.231%, roughly 13 basis points under my three-source average, and it shows that number down 7 basis points on the week from 7.297%. I'm not folding it into the headline, because I've been averaging the same three sources all month and I want the number comparable day over day. But I'm telling you it exists, because it's the cleanest illustration of the thing I say constantly: a published average is a survey of somebody's lender panel, not an offer to you. On the weekly side, Freddie Mac's survey still reads 7.32% on the 30-year for the week ending October 22, down from 7.40% and its first weekly decline since August, with the 15-year at 6.68%. The next survey publishes Thursday morning — the same morning as PCE. All of these are illustrative averages, not a rate quote and not a commitment to lend.
FHA near 7.10% and VA near 7.09% — the 50-basis-point spread between sources is still open
FHA is running near 7.10% and VA near 7.09% on the 30-year fixed, both a basis point or two better than yesterday. Here's the panel so you can see the seams. On FHA: Mortgage News Daily 6.96% at Monday's close, down 2 basis points on the session; Bankrate 7.16% as of 6:30am Eastern this morning; The Mortgage Reports 7.192%. Average 7.104%, with a 23-basis-point spread low to high. On VA: Mortgage News Daily 6.98% Monday, down a basis point; Bankrate 7.12% this morning; The Mortgage Reports 7.162%. Average 7.087%, spread of 18 basis points. Now add that fourth panel and watch what happens. Mortgage Research Center has FHA at 6.692% and VA at 6.743%, both down 9 and 10 basis points respectively on the week. Take the FHA numbers alone: 6.692% at the low end against 7.192% at the high end. That's a full 50-basis-point spread between published averages, on the same program, on the same morning — essentially unchanged from the gap I showed you yesterday. Most lenders won't tell you this, so I will: that gap isn't anybody lying. Different outlets poll different lender panels, with different credit-score and down-payment assumptions, different points baked in, and different refresh schedules — some overnight, some from the prior close. Which is exactly why no headline, mine included, is a quote. Your actual number comes off your file: credit, down payment, debt-to-income, occupancy, property type, and which of the 160+ investors I shop turns out to be sharpest on your particular profile that day. The honest use for all of this is direction and ballpark. The quote comes from the file.
The 10-year is holding just under 5% — and oil is easing for a second day
This is the mechanism to understand before tomorrow afternoon, so let me walk it one more time. Your mortgage rate tracks the 10-year Treasury, not the fed funds rate. As of this morning the 10-year sits at about 4.98%, down a basis point or two on the day and essentially parked where it's been since Thursday's rally. For the shape rather than one dot: roughly 5.13% on Monday October 19, 5.12% Tuesday the 20th, about 5.06% Wednesday the 21st, then Thursday the 22nd's drop to 4.96% — down about 17 basis points, the largest one-day decline of 2026 — Friday giving roughly 5 of that back, and the two sessions since going almost nowhere. Worth remembering what caused Thursday: not the Fed and not inflation, but a rotation out of AI and technology shares into Treasuries after Meta's results revived questions about how it's funding its buildout. And what reversed part of it Friday: oil. Which brings us to the small piece of good news this week. Brent settled Friday at $116.72, its highest close since April 29, after Houthi strikes on Saudi targets including energy infrastructure. It's come off that high two sessions running — this morning it reads about $114.84, down roughly 0.8% on the day. The reasons given: profit-taking after that run, real questions about whether the conflict has actually cut physical supply enough to justify the price, Middle Eastern crude exports recovering toward pre-war levels, and markets waiting on what the administration decides about Iran. Why it matters to you: cheaper oil means less inflation premium, and less inflation premium eventually means a better 10-year. Emphasis on eventually, because the 10-year still hasn't moved on it. Here's the part worth sitting with. In the five sessions behind us your rate was pushed around by a tech selloff, an oil spike, and a geopolitical headline — and not once by a Fed announcement. That's the whole argument against waiting for a headline to tell you it's time.
Today's data, then the Fed tomorrow at 2pm — this is the loaded stretch
Knowing what's on the calendar is most of what separates feeling blindsided from feeling prepared, and the next three days are dense. Today, Tuesday October 27, brings four prints: September durable goods orders at 8:30am Eastern, the Case-Shiller home price index for August at 9:00, then the Conference Board's consumer confidence read and September new home sales both at 10:00. Case-Shiller and new home sales are the two that speak to housing directly, and they land while the Fed is in its pre-meeting blackout — so there's no official commentary coming to frame them. Tomorrow, Wednesday October 28, opens with the September trade balance and wholesale inventories at 8:30, then the main event: statement at 2:00pm Eastern and Chair Kevin Warsh's press conference at 2:30. There are no updated economic projections and no dot plot at this meeting, which means the statement language and that press conference are the entire story. Thursday, October 29 is the heavy one, and it's all at 8:30am — weekly jobless claims, the first estimate of third-quarter GDP with consensus around 3%, and the September PCE price index including core PCE, where the crowd is looking for 0.3% on the month. PCE is the Fed's own preferred inflation gauge. Freddie Mac's weekly survey publishes that same morning. Friday, October 30 has the third-quarter Employment Cost Index at 8:30, Chicago PMI at 9:45, and two Fed speakers in the afternoon — San Francisco's Mary Daly at noon and St. Louis's Alberto Musalem at 12:30. Here's the piece most mortgage coverage will skip, and after what happened last Thursday I think it matters to you. Microsoft and Alphabet report tomorrow after the close — hours after the Fed — and Amazon and Intel follow Thursday after the close. Hyperscaler capital spending is running somewhere around $800 billion this year, and last week proved that when those names sell off, money runs into Treasuries and your rate improves for reasons that have nothing to do with housing or monetary policy. So there are two separate things that could move your number in the next 48 hours, and only one of them is the Fed.
The Fed decides tomorrow — a hold is about 90%, and December is the real question
Start with where we are. September's meeting raised the target range to 3.75%–4.00% — a quarter point, and the first increase in three years, the opening move of Kevin Warsh's time as Chair. Then core inflation came in hot on Wednesday the 14th and October hike bets spiked. Then Warsh, speaking at the IMF and World Bank annual meetings in Bangkok, declined to tip his hand at all: 'I don't prejudge meetings.' The markets took 'no signal' as 'no hike,' and the hold odds have been firming ever since. For tomorrow's announcement at 2pm Eastern, a hold is the strong base case — CME FedWatch has a hike at roughly 10%, which puts no-change near 90%, and the broad read across coverage this morning is that a hold is largely priced in. Two weeks ago this was closer to 70/30, so the move has been large. December is where the conviction actually lives. Futures are showing better than 80% odds of another hike in December. The analyst reads worth knowing: Deutsche Bank's Brett Ryan expects Warsh to say more hikes will probably be needed over the coming months, and argues that skipping October buys officials more time to assess the data. Evercore ISI's Krishna Guha reads Warsh's recent remarks as fitting a baseline of roughly one hike per quarter. Warsh has also said he's wary of previewing policy and would prefer to end the Fed's forward guidance altogether — which is why, with no dot plot at this meeting, his 2:30 press conference tone is the dot plot. Here's what I'd want you to take from all of it. The fed funds rate is not your mortgage rate. It's an overnight bank rate, and the 10-year Treasury is what prices your loan — which is why rates can rise on a Fed cut and fall on a Fed hold. And real talk: nobody writes a check for an interest rate. They write a check for a payment. The rate dictates the payment, so the payment is what we build, and we build it out of program, assistance, and who pays for what at the closing table — not out of guessing what Warsh says tomorrow afternoon.
Friday's flash PMI was the hawkish surprise — selling prices rose the most since 2022
This landed Friday, October 23, and it's still the most important recent number in this briefing, because it speaks directly to the inflation question the Fed is sitting with right now. S&P Global's flash October PMI showed the economy running hot. The composite output index came in at 57.6, right on consensus and down from September's 58.4 — a two-month low, but still firmly in expansion. Manufacturing hit 56.0 against a 55.5 consensus, which is a 53-month high. Services came in at 57.9, a shade under the 58.0 expected and down from 58.8. Growth, in other words, is not the problem. Prices are. Selling prices rose at their fastest pace since August 2022, and the increase was broad-based across both goods and services. Input costs rose at their fastest pace since November 2022 outside of September, with respondents pointing at oil-based inputs, energy, fuel and freight, plus metals like steel, aluminum and copper on tariffs and shortages. Supplier delivery times lengthened the most since July 2022. Chris Williamson, S&P Global's chief business economist, didn't soften it: 'October's flash PMI data indicate that the US economy continues to boom at the start of the fourth quarter,' he said, adding that excluding the post-COVID reopening, 'recent months have seen the best growth spell for US businesses since early 2015.' Then the line that matters for your rate: 'The result is the sharpest increase in selling prices for over four years,' which he said points to consumer price inflation rising further above the Fed's target in the coming months. Also Friday, the University of Michigan's final October consumer sentiment read came in at 48, basically flat against September's 48.1 and sitting near the bottom of that series' history. One-year inflation expectations ticked up to 4.7% from 4.6%, and five-year expectations to 3.5% from 3.4%. Here's my honest translation. A booming economy with accelerating selling prices and rising inflation expectations is not the mix that gets you lower mortgage rates. It's the mix that keeps the Fed cautious and keeps the long end of the curve sticky. That's the real reason Friday's bond move went the wrong way, and it's why I'd treat last Thursday's rally as a gift rather than a trend. It also raises the stakes on Thursday's PCE report considerably. None of that changes what I can actually go to work on for you, which is structure — program, assistance, and who's paying for what at the closing table.
Last week's data: claims at a three-month low, leading indicators flat, applications down 5.3%
Three things from Thursday the 22nd are still worth your time. First, weekly jobless claims came in at 194,000 — a three-month low, with the four-week average sitting just under 200,000. The wrinkle underneath it: continuing claims rose to 1.728 million, up about 1.5% on the week. Read those two together and you get what economists keep calling a low-hire, low-fire labor market. Employers aren't laying people off, but they also aren't hiring, so the people who do lose a job are staying unemployed longer. Second, the Conference Board's Leading Economic Index was unchanged in September — a flat 0.0% on the month against an expected 0.1% decline, leaving the index at 99.6. Better than forecast, and flat is not falling, but flat isn't acceleration either. Third, and this is the one that speaks directly to the housing market: the Mortgage Bankers Association reported mortgage applications fell 5.3% for the week ending October 16. Purchase applications were running about 18% below last year's pace, and adjustable-rate mortgages made up 10.5% of applications — a tell that borrowers are reaching for anything that lowers a payment. MBA's Joel Kan put it plainly, and the follow-on quote this week was blunter still: lower rates were not enough to boost application activity. Here's how I'd translate that last one, because it reads like bad news and for a buyer it mostly isn't. Purchase applications 18% below a year ago means roughly one in five of the people who would have been competing with you last October isn't in the market this October. That shows up as less competition on the house you want and more willingness across the table. On the ARM share: 10.5% is high, and I'd rather you understand why people are reaching for them than reach for one yourself without a conversation. An ARM can be the right tool in a specific situation. It is not a free lunch, and the fixed-versus-ARM decision is one of the places I see people make an expensive choice fast.
Two housing reports came in weak — pending sales hit a record low
These two landed Tuesday, October 20, and both came in on the soft side. First, September new residential construction from the Census Bureau. Housing starts ran at a 1,221,000 seasonally adjusted annual rate, down 5.2% from August's revised 1,288,000 and down 7.4% from September 2025's 1,319,000. Single-family starts were 859,000, down 4.6% from August's revised 900,000. Building permits — the forward-looking piece — came in at 1,400,000, down 0.2% on the month and down 3.0% from a year ago. Completions were 1,192,000, up 2.2% on the month but down 14.0% from a year ago. Fair warning on those confidence intervals: Census publishes wide margins on starts, so month-to-month moves in this series are noisier than they look. Second, and this is the bigger number: NAR's September Pending Home Sales Index fell 7.3% from August and 11.3% from a year ago — a record low for the series. The West was the weakest of the four regions, down 9.2% on the month and 15.0% on the year. Lawrence Yun, NAR's chief economist, tied it straight to rates: 'Market conditions weakened in September due to fast-rising mortgage rates.' He added the framing I'd underline for anyone shopping right now — 'Although job growth and income gains can support home purchases,' he said, 'housing affordability is more affected by changes in mortgage rates.' Two more lines from him worth your time. On where rates go: 'Given that mortgage rates have already risen much more than the Federal Reserve's short-term interest rate hikes,' he said, 'mortgage rates could be topping out.' That's his read, not a forecast you should bank on, and I'm not going to dress it up as one. And on prices: 'Despite the difficulty of obtaining a mortgage, the limited number of distressed properties is ensuring steady home prices.' His closing thought was that buyers and sellers may simply be holding back until mortgage rates move favorably. Now let me translate all of that, because a record low in pending sales sounds like a reason to stay home and it mostly isn't. Pending sales are contracts signed — it's a measure of how many people are competing with you right now, and the answer is the fewest on record. Builders are cutting prices and buying down rates because starts and traffic are falling. Sellers are conceding because contracts aren't getting signed. Every one of those weak numbers is a line item in your negotiating position. The buyers who do transact in a market like this are the ones who got their financing squared away first and could move when the right house showed up. That's the part I can help with today.
Builder confidence at a 14-year low — and 43% of builders are cutting prices
This one landed Monday, October 19. The NAHB/Wells Fargo Housing Market Index — builder confidence — fell three points in October to 29. That's the lowest reading since June 2012, fourteen years back. Under the hood: current sales conditions 33, down 2 points; sales expectations for the next six months 32, down 5; prospective buyer traffic 18, down 5. That traffic number is the weakest since April 2012 if you set aside the April 2020 pandemic shutdown. NAHB Chairman Bill Owens tied it to mortgage rates having 'climbed well above 7% in recent weeks,' plus higher gas prices and inflation weighing on traffic. Chief Economist Robert Dietz said demand is still there, but higher rates and elevated construction costs are keeping more buyers on the sidelines. The starts and permits data above is the same story showing up in actual construction activity rather than sentiment — and today's new home sales print at 10:00am Eastern is the next read on whether it's still going that direction. Now here's why I'm handing you a builder sentiment number instead of skipping it, because on its face it sounds like bad news and for you it mostly isn't. Two lines from that release matter more than the headline index. First, 43% of builders cut prices in October — the highest share anywhere in the 2020 through 2026 stretch — with the average cut running 6%, which is now the seventh month in a row at that same figure. Second, 65% of builders used sales incentives, the 19th straight month at or above 60%. Read that the way a buyer should read it. Builders can't move inventory at these rates, so they're discounting and they're buying down rates to make payments work. New construction is one of the few corners of this market where the seller has an actual budget line for exactly that. If you've only been looking at resale, it's worth walking a couple of builder communities and asking flat out what they'll put toward financing. Just bring your own lender to that conversation. The builder's preferred lender is not your only option, and comparing the whole package — the rate, the costs, and what the incentive is actually tied to — is the entire ballgame. That's a conversation I have with people every week, and it costs you nothing.
September CPI and PPI both ran hot — the reason December still looks the way it does
Let's go back to the numbers that set up this stretch, because they're the reason the December odds sit above 80%. The Bureau of Labor Statistics published September CPI on Wednesday, October 14. Headline inflation rose 0.6% on the month, which is exactly what forecasters expected, and that took the annual rate to 3.7% from 3.4% in August — a tenth hotter than the 3.6% the crowd was looking for. Core, which strips out food and energy and is the number the Fed actually watches, rose 0.3% on the month against a 0.2% consensus, pushing the annual core rate to 2.6% from 2.4%. Both core readings came in a tenth above expectations. That's what 'hot' means in practice — not a disaster, but the wrong direction in the one series that drives Fed policy. Two details underneath the headline worth your time. First, gasoline did most of the damage on the headline number, same story as August — and with Brent still up near $115 after last week's spike, keep that mechanism in mind for the October report, because oil at these levels doesn't stay out of the data forever. Second, and this is the one that actually matters, core services rose on shelter costs jumping, and shelter is the sticky part. Then Thursday the 15th brought two more. Producer prices, the wholesale version of inflation, came in up 1.3% on the month and 6.3% over the year. That is a big number, and it matters because wholesale prices feed into consumer prices a month or two later. Retail sales beat too — up 0.8% on the month against a 0.6% consensus. But here's the detail most coverage skipped, and it's the useful one: strip out gasoline and retail sales rose only 0.2%. So the headline says the consumer is spending; the guts say the consumer is mostly paying more at the pump. Nothing in this paragraph is a prediction. It's the backdrop Thursday's PCE report gets read against, and PCE is the gauge the Fed actually prefers.
NAR's September existing-home sales: sales ticked up, and buyers got more room
NAR published September existing-home sales on Tuesday, October 13, and here's the picture. Sales ran at a 3.99 million annual pace, up 1.5% from August but down 2.2% from a year ago. Inventory climbed to 1.60 million homes — up 1.9% from August and 4.6% from September 2025 — which works out to a 4.8-month supply, unchanged from August and up from 4.5 months a year ago. The median existing-home price was $417,500, up 1.3% year over year from $412,300. Homes took a median of 33 days to sell, up from 31 in August and flat against a year ago. Worth holding this one next to the pending sales number above, because they're measuring different moments: existing-home sales are deals that closed in September, pending sales are contracts signed in September. Closings held up; new contracts hit a record low. That's the pipeline thinning out in real time, and it's why I read the leverage as still shifting toward buyers rather than away. Now the part I actually care about for the people I work with. First-time buyers were 31% of all sales in September, up from 30% in August and 30% a year ago. All-cash buyers were 27%, unchanged on the month and down from 30% a year ago. Investors and second-home buyers were 20%, up from 15% in August. Read those three together, because they're better news for a first-time buyer than the headline is. Your share of the market is going up while the overall pace sits near the slowest of the year, and cash buyers are down three full points from a year ago — which means fewer of the offers you simply cannot compete with. Lawrence Yun, NAR's chief economist, said the pace part plainly: 'August and September marked the two slowest months for home sales this year.' And on why, he pointed to recent mortgage rate increases meaning fewer buyers can qualify to purchase a home. For the West specifically, sales ran a 730,000 pace, up 2.8% on the month and down 5.2% from a year ago, with a median price of $613,600 — down 0.3% year over year. Next up on the national calendar: October existing-home sales on Thursday, November 12.
