There's a sentence that gets repeated every week on financial television, and it's wrong in a way that costs people real money.
The sentence is some version of "when the Fed cuts, mortgage rates will come down." It sounds obvious. The Fed sets interest rates, a mortgage is an interest rate, therefore the Fed sets mortgage rates. Except that isn't how the machinery works, and if you're sitting on the sidelines waiting for the Federal Open Market Committee to hand you a cheaper house, you're waiting at the wrong door.
The Fed sets one rate. That's it. It sets the target range for the federal funds rate, which is the overnight cost of banks lending reserves to each other. Right now that range is 3.50 to 3.75 percent, with the effective rate printing around 3.63. Overnight money. Money that gets borrowed at four in the afternoon and paid back the next morning.
Your mortgage is a thirty year obligation. Nothing about an overnight rate tells you what a thirty year commitment should cost.
What actually prices your mortgage is the 10-year Treasury yield, plus a spread. The 10-year is the market's price for lending money to the United States government over roughly the same horizon that a typical mortgage actually lives, because most mortgages get paid off or refinanced long before year thirty. The spread on top is what investors demand for taking your loan instead of the government's.
As of last week, Freddie Mac's weekly survey put the average 30-year fixed at 6.67 percent, down a basis point or two from the week before. The 15-year sat at 5.96. The 10-year Treasury has been hovering around 4.65. Do the subtraction and you get a spread of roughly two full percentage points.
Historically, that spread runs closer to 1.5 or 1.7. Which means somewhere between thirty and fifty basis points of what you're paying has nothing to do with the Fed, nothing to do with Treasury yields, and nothing to do with your credit score. It's a premium the market is charging for uncertainty, and it is the single most overlooked line item in American household finance.
Here's what's inside that spread, because it isn't arbitrary.
The largest component is prepayment risk. When you take a 30-year fixed mortgage, you get something remarkable: the right to walk away from the contract whenever it suits you. Rates drop, you refinance, and the investor who owned your loan gets their capital back at exactly the moment reinvesting it became less attractive. Rates rise and you sit tight, leaving that investor stuck holding a below-market asset. You hold a free option. Someone has to price it, and that someone is you, embedded in your rate.
The second component is volatility. Prepayment risk gets more expensive when rates move around a lot, because the option you hold becomes more valuable and therefore more costly to sell you. We have spent this year with a war affecting energy prices, an inflation rate that went from a negative monthly print in June to a positive one in July, and a rates market that flipped from pricing cuts to pricing hikes. That's an environment where mortgage investors demand to be paid more.
The third is who's buying. For years, the Federal Reserve was a massive purchaser of mortgage-backed securities. It isn't anymore. Banks, which were also large buyers, have their own reasons to be more careful about duration. When the two biggest structural buyers step back, the remaining buyers set the price, and they set it wider.
Now connect this to what's happening in housing this week, because the calendar is dense.
The Census Bureau releases housing starts and building permits for July on Tuesday morning. This series has been genuinely wild. May printed 1.177 million annualized, the lowest since 2020. June rebounded 19 percent to 1.427 million. Multifamily has been swinging in forty and seventy percent increments month to month. Single-family starts, meanwhile, have been drifting down for three straight months. That divergence tells you something: builders are still willing to put up apartments, and increasingly unwilling to speculate on houses nobody has signed for.
On the existing side, the National Association of Realtors reported July sales down 1.7 percent from June. June ran at a 4.09 million annual pace with a median price of $440,600 and 4.6 months of inventory. Year to date, sales are up 2.4 percent. NAR's chief economist has been direct about the gap: a normal market is closer to five million units a year, and we're running around four.
That missing million transactions a year isn't demand disappearing. It's lock-in. Tens of millions of households hold mortgages at three and four percent. Moving means surrendering that rate and re-entering at 6.67. So they don't move. The house doesn't list, inventory stays thin, and prices stay stubborn even as sales volume sits a fifth below normal. It's the closest thing modern housing has to a frozen order book.
You'll get the corporate read on Tuesday and Wednesday when Home Depot, Toll Brothers, and Lowe's report. Watch what the builders say about incentives, because builders have a tool homeowners don't: they can buy down a buyer's rate and bury the cost in the price. When rate buydowns get more aggressive, that's a builder telling you demand is weaker than the headline sales number suggests.
So what do you actually do with this?
Start by accepting that the spread, not the Fed, is your variable. If the Fed hikes in September, the 10-year may barely flinch, because a hike aimed at controlling inflation can lower long-term inflation expectations and push long yields down. That has already happened this year. Conversely, if the Fed cuts into an economy the market thinks is overheating, long yields could rise. The front end and the back end of the curve are separate animals with separate drivers, and your mortgage lives at the back.
If you're waiting to buy, stop watching FOMC dates and start watching two numbers: the 10-year yield and the Freddie Mac survey rate. Subtract one from the other every Thursday when the survey drops. When that gap starts compressing back toward 1.7, you're getting relief that has nothing to do with policy and everything to do with volatility calming down. That's the signal most buyers never see coming because they're staring at the wrong screen.
If you already own, your refinance trigger should be built on math, not on a feeling. The rough rule is that refinancing makes sense when your monthly savings recover your closing costs inside the time you actually plan to stay. If closing costs run $6,000 and a refi saves you $180 a month, that's a 33 month breakeven. If you might move in two years, it's a bad trade regardless of how good the rate looks. Write your trigger rate down now, while you're calm, so you're not making the decision in a week when everyone's excited.
If you're being offered points, run the same arithmetic. A point is one percent of the loan amount paid up front to reduce your rate. On a $400,000 loan that's $4,000 for perhaps a quarter point of rate, which is roughly $60 a month. Sixty-six months to break even. That's a long time to be right about staying put.
And if you're weighing an adjustable rate mortgage, notice that ARMs have recently priced around 6.36 percent against the 6.67 fixed. Roughly thirty basis points of savings for taking on all of the future rate risk yourself. In a market pricing a possible hike, that is a thin premium for a large transfer of uncertainty onto your balance sheet. Sometimes it's the right call. Just be clear you're being paid very little to accept it.
There's one more piece of arithmetic worth running, and it's the one that gets skipped most often: the rent versus buy comparison, done properly.
Most people compare a rent payment to a mortgage payment. That comparison is close to meaningless, because a mortgage payment isn't a cost. Part of it is principal, which is money moving from one pocket to another. What you actually spend to own is the interest, plus property taxes, plus insurance, plus maintenance, plus the transaction costs of eventually selling, amortized over how long you stay.
On a $400,000 loan at 6.67 percent, first-year interest runs a bit over $26,000. Property taxes vary enormously by state but call it 1.1 percent of value nationally. Homeowners insurance has been rising faster than almost any other household line item, and in several states it has doubled inside three years. Maintenance is conventionally estimated at one percent of value annually, and anyone who owns a house knows that estimate is generous in a good year. Selling costs run five to six percent of the sale price.
Add those together and the true annual carrying cost on a mid-priced home is frequently forty to fifty thousand dollars before a single dollar of principal gets paid. That's the number to compare against rent, not the mortgage payment. Sometimes buying still wins comfortably. Sometimes it doesn't, and the person who ran the honest version of the math is the one who finds out before signing rather than after.
The insurance line deserves particular attention right now because it's quietly become the fastest-moving cost in homeownership, it isn't fixed by your mortgage rate, and it doesn't stop rising when you lock a rate. A household that fixed its mortgage at three percent in 2021 and considers itself insulated has still watched its total housing cost climb, because two of the four major components were never fixed at all.
The bigger point underneath all of this is one I keep coming back to. For most American households, the house is the single largest position they hold, it's the most leveraged, it's the least liquid, and it's the only major asset they can never rebalance. You can trim an overweight stock in thirty seconds. You cannot sell nineteen percent of your kitchen.
Which means the discipline has to happen everywhere else. If your home equity is now sixty percent of your net worth, that's a concentration decision you made by accident, and the only place to correct it is in the accounts you can actually control. Most people don't know their real number because the mortgage lives at one institution, the equity estimate lives on a real estate app, and the investment accounts live somewhere else entirely.
This is exactly the problem Empower's free financial dashboard was built to solve. It links your mortgage, your property value, and your investment accounts onto one net worth statement so you can actually see what percentage of your wealth is sitting in a single illiquid, leveraged, undiversifiable asset. It costs nothing to connect, and for most people the first look is genuinely uncomfortable in a useful way.
This week gives you a real test. Tuesday's starts number, Wednesday's builder commentary, and the Thursday mortgage survey all land inside seventy-two hours. Watch the spread, not the headline. The headline will tell you what happened. The spread will tell you what it costs you.
I put together a one-page tracker for exactly this: the four numbers to log each week, where to find each one for free, the refinance breakeven formula with worked examples, the points calculation, and a simple rule for when spread compression is real versus noise. It takes about four minutes a week to maintain and it will keep you from making a six-figure decision based on a television segment.
Reply to this email with the word SPREAD and I'll send you The Mortgage Spread Tracker.
Own the system,
Taylor Voss
Money Systems Lab
DISCLOSURE: This newsletter is educational content and does not constitute financial, investment, tax, or legal advice. Money Systems Lab is not a registered investment advisor. All figures cited are drawn from public sources as of the date of publication and are subject to change. This issue contains affiliate links, which means Money Systems Lab may earn a commission at no additional cost to you if you open an account through them. We only reference tools we would use ourselves. Always do your own research and consult a qualified professional before making financial decisions.