The Lock Desk

Freddie Mac put the thirty year fixed at 6.95 percent last Thursday. That is up from 6.76 percent the week before, and 6.26 percent a year ago. The fifteen year landed at 6.26 percent. It was the fourth straight weekly increase and the largest one week jump in about sixteen months.

If you are buying, refinancing, or sitting on a line of credit tied to your house, that number just changed your math.

Here is what almost everybody does in response. They open three tabs, get three quotes, pick the lowest one, and feel like they did the work.

That is not the work. That is the appetizer. The spread between three retail quotes on the same day is usually 15 to 25 basis points. The spread between a badly structured loan and a well structured one on the same property, with the same borrower, at the same lender, is routinely 75 to 150 basis points plus thousands of dollars in cost.

Today I am going to walk you through the five levers that live inside that second number. This is the stuff the lock desk knows and the loan officer does not always volunteer.

Your mortgage rate is not a number your lender invented. It is built from two pieces.

Piece one is the market. Mortgage pricing tracks the ten year Treasury, which has been climbing and sat around 4.86 percent heading into last week's Fed meeting. You have exactly zero influence over the ten year Treasury. Stop trying. Stop reading forecasts about it. Nobody knows.

Piece two is the spread, which is everything stacked on top of the market rate to account for you, your property, your loan structure, and the lender's margin. Historically that spread runs 1.5 to 2 percentage points over the ten year. Right now the gap between 4.86 and 6.95 is a little over two points.

That spread is the entire negotiable surface. Every lever below is a way to shave it.

Most borrowers think a rate lock is a yes or no. It is not. It is a menu, and the menu has prices.

A 15 day lock is cheaper than a 30 day lock. A 30 is cheaper than a 45. A 60 day lock can cost you 25 basis points in rate or the equivalent in points compared to a 15. The lender is selling you insurance against market movement, and longer coverage costs more. Obvious when you say it out loud, invisible on most rate sheets shown to consumers.

Three things to do with this.

Ask for the lock tier pricing in writing. Literally ask: "What is my rate at a 15, 30, 45, and 60 day lock?" If your loan officer cannot produce that in an hour, you are talking to someone who is not looking at the rate sheet.

Do not lock longer than your close. People lock 60 days out of anxiety on a deal that closes in 22. You paid for 38 days of coverage you are throwing away.

Ask about the float down. Many lenders offer a one time float down provision, where if the market improves by some threshold before closing, you get the better rate. Some include it free on longer locks. Some charge for it. Some pretend it does not exist until you ask by name. In a rising market a float down is worth less than it sounds, but it costs nothing to ask and it tells you immediately how sophisticated your lender is.

And know the extension fee before you need it. Lock extensions typically run somewhere around 2 to 6 basis points a day. If your closing slips a week, that is real. Know the number before you are three days out and desperate.

A discount point costs 1 percent of the loan and buys down your rate, usually somewhere between an eighth and a quarter of a point depending on the day.

The standard break-even math is straightforward. Divide the cost of the point by the monthly payment savings and you get the number of months it takes to recover.

On a 500,000-dollar loan, one point costs 5,000 dollars. Buying 6.95 down to 6.70 saves roughly 84 dollars a month. Break-even is about 60 months.

Now here is the part people skip. The relevant question is not "will I own this house for five years." It is "will I have this loan for five years."

Those are different questions and the second one is shorter, almost always. You might refinance. You might recast. You might move. The median mortgage does not survive anything close to its term. If you believe rates are near a local peak and you would refinance on a meaningful decline, every dollar you put into points is a dollar you are betting will not get refinanced away.

My rule: buy points only when your honest expected loan life is at least double the break-even, and only when you are not already stretching to close. In a rising rate environment with more hikes projected, points get more attractive, not less, because the refinance-away risk drops. But double the break-even is still the bar.

This is the most underused lever on the board right now, and with rates at a nineteen month high, sellers are increasingly open to it.

Say you negotiate 10,000 dollars off a 500,000-dollar purchase. On a 400,000-dollar loan at 6.95 percent, dropping the loan to 392,000 saves you about 53 dollars a month. Nice. Small.

Now take that same 10,000 dollars as a seller credit toward a rate buydown instead. Depending on pricing that day, 10,000 dollars buys roughly two discount points on that loan, which might take you from 6.95 to around 6.45. That saves roughly 130 dollars a month, every month, for as long as you hold the loan.

Same 10,000 dollars from the seller. Two and a half times the monthly benefit.

Or put it into a temporary buydown, the 2-1 structure, where your rate is two points lower in year one and one point lower in year two before settling at the note rate. On that loan, a 2-1 runs roughly 12,000 dollars in escrow and cuts your year one payment by something like 500 dollars a month. If your income is stepping up, or you genuinely expect to refinance, the front loaded relief can be worth more to you than a permanent quarter point.

Two cautions, because I am not going to sell you the fun version without the fine print.

Seller credits are capped by loan type and down payment. Conventional loans with 10 to 25 percent down generally cap seller contributions at 6 percent of price. Under 10 percent down it drops to 3 percent. Know your ceiling before you negotiate.

And the appraisal has to support the price. A credit only works if the house appraises at the higher number. Ask your agent to structure it that way in the offer, not as an afterthought during inspection.

Loan level price adjustments are the surcharges baked into your rate based on risk characteristics. They are published, they are mechanical, and they are stair stepped, which means they move in bands rather than smoothly.

Bands are the opportunity. Being one point of credit score below a threshold costs you the same as being forty points below it.

The bands that matter most:

Credit score. Pricing tiers commonly break at 620, 660, 680, 700, 720, 740, and 760. Moving from 738 to 741 can be worth an eighth to a quarter point in rate on the same loan. If you are within ten points of a break, paying down a revolving balance to drop utilization before you pull credit is the highest return financial move available to you that month. Not metaphorically. Actually.

Loan to value. Breaks at 80, 75, 70, and 60 percent. Crossing under 80 kills mortgage insurance on a conventional loan and improves your pricing at the same time. Sometimes an extra 4,000 dollars down saves you 180 dollars a month in PMI plus a pricing improvement. Run the LTV bands before you finalize your down payment, not after.

Property and occupancy. Second homes and investment properties carry material adjustments. Condos above 75 percent LTV carry one. These are not negotiable but they are knowable, and knowing them stops you from getting talked into a structure that quietly costs you a half point.

Escrow waiver. Waiving escrow usually costs you an eighth of a point or a small fee. People take the waiver for control and then lose the math. If you are not going to earn more than an eighth of a point on the float, do not waive.

Two structural notes.

The fifteen year sits at 6.26 percent against the thirty at 6.95. That is a 69 basis point discount for the shorter term, which is wide. If the payment fits and you would otherwise be prepaying a thirty, the fifteen gets you the lower rate as a feature rather than a sacrifice. If the payment does not fit comfortably, do not force it. Liquidity beats amortization speed almost every time.

And then the move almost no lender will bring up, because it earns them nothing: the recast.

If you already have a mortgage at a rate below today's market, refinancing is off the table. Refinancing a 5.5 percent loan into a 6.95 percent loan to access equity or reduce a payment is usually a bad trade dressed up as a solution.

A recast solves the payment problem without touching the rate. You make a lump sum principal payment, typically 5,000 to 10,000 dollars minimum, and the servicer re-amortizes the remaining balance over the remaining term. Your rate stays exactly where it is. Your term stays where it is. Your payment drops. The fee is usually 150 to 500 dollars.

Not every loan permits it and most servicers bury it. Call and ask for the recast department by name. If you have a low rate mortgage and cash you were considering throwing at the principal anyway, a recast converts that into monthly cash flow instead of just a shorter tail.

Two operational notes that will make all five levers work better.

Get everything on one day. Rate quotes decay. Comparing a Monday quote to a Thursday quote in a week where the thirty year moved 19 basis points is comparing nothing to nothing. Line up your calls inside a four hour window.

Record the calls. This is the single biggest upgrade I made to how I handle lender conversations. I run Fathom on every lender call, and afterward I have a searchable transcript of exactly what was promised on lock length, float down, credits, and fees. When the loan estimate arrives and something has quietly drifted, I am not arguing from memory. I am reading from the transcript.

And for tracking the market while you are in process, I run a small Make.com scenario that pulls the weekly Freddie Mac survey number and the ten year yield into a sheet every Friday morning. Three months of that history is what turns "rates feel high" into an actual lock decision.

If you are in a transaction right now, here is the sequence.

Pull your credit and check the band distance first, before anything else, because it takes weeks to fix and everything else takes days. Set your down payment against the LTV bands. Get same day quotes with full lock tier pricing. Negotiate seller credits toward buydown rather than price, up to your contribution cap. Then and only then decide on points, using double the break-even as your bar.

Do that and you will beat the person who opened three tabs by a wide margin, on a loan you will carry for years.

Want the worksheet?

I built The Lock Desk Worksheet to run this exact sequence. It has the lock tier comparison grid, the honest points break-even calculator with the loan life input, the credit versus price cut side by side, the LLPA band distance checker, the recast versus refinance decision tree, and the full set of questions to read to your loan officer in order.

Reply to this email with the word LOCK and it is yours.

See you Friday for the deep dive.

Alex Rivera, Wealth Architect at Wealth Grid