October 17, 2026 · 8 min read
ARM Share Just Hit 10.5%: What an Adjustable-Rate Mortgage Really Costs at 7.50%
The ARM share of mortgage applications reached 10.5%, its highest since October 2025. What the gap under the 30-year fixed buys, and what the reset can cost.

One number in this week's mortgage data is doing more work than the headline. Applications fell 2.7% for the week ending October 9, 2026, which is the kind of line that gets skimmed. Underneath it, the adjustable-rate mortgage share of total applications climbed to 10.5%, the highest level since October 2025, according to the Mortgage Bankers Association's Weekly Applications Survey. More than one in ten people applying for a mortgage right now is choosing a loan whose rate is not fixed for the life of the term. A year ago that was a fringe choice. It is now a mainstream one, and if you are buying in Rhode Island, Massachusetts, or Connecticut this fall, somebody is going to put an ARM in front of you.
What the 10.5% is actually telling you
It is telling you that payments, not rates, are driving decisions. The MBA's survey put the average contract rate on a 30-year fixed conforming loan at 7.50% for that week, up from 7.49%, which Joel Kan, the MBA's Vice President and Deputy Chief Economist, described as the highest since November 2023 and more than a percentage point above a year ago. The demand picture around it is thin: the Refinance Index was 56% below the same week last year, and the unadjusted Purchase Index was 16% below. When rates rise past the point where buyers can absorb them, the people still in the market do not stop buying. They start reaching for whatever lowers the monthly number, and the ARM is the most available lever on the shelf.
Why the ARM is cheaper this month
Because the short end of the curve is cheaper than the long end right now. In the same MBA survey, the average contract rate on a 5/1 ARM was 6.54%, up from 6.43% the week before. Against the 7.50% conforming fixed, that is a gap of 0.96 percentage points. For most of the last few years that gap was narrow enough that the ARM was not worth the complexity, which is exactly why the ARM share sat in the single digits. At nearly a full point, the arithmetic changes, and so does the sales pitch you are going to hear. Worth noting in the same breath: that 6.54% came with 1.13 points, while the 7.50% fixed came with 0.83 points, so part of the advertised gap was bought and paid for up front.
What the gap is worth in dollars
Take a $450,000 loan amount, which is a realistic figure across much of southern New England. At 7.50% over 30 years, principal and interest run about $3,146 a month. At 6.54%, the same loan runs about $2,856. That is roughly $290 a month, or about $17,400 across the first five years before the rate can move. There is a second, quieter benefit people forget: because more of the lower payment goes to principal, the ARM balance after five years would sit near $421,400 against roughly $425,800 on the fixed, so you would owe about $4,300 less. Those are illustrative calculations, not quoted figures, and they exclude taxes, insurance, and mortgage insurance. Your own numbers depend on your rate, your points, and your loan size.
What happens at the reset
This is the part the monthly savings tends to crowd out. After the fixed period ends, the rate becomes an index plus a margin written into your note, recalculated on a schedule and bounded by caps. A common structure limits the first adjustment to two percentage points, but caps, margins, indexes, and adjustment frequency vary from loan to loan, so the only caps that matter are the ones in your own documents. Run the bad case before you sign. If that $450,000 ARM reset two points higher, to 8.54%, the payment on the remaining balance over the remaining 25 years would be roughly $3,405 a month, which is about $259 more than the fixed you passed on. Again, illustrative. The point is not that the reset will happen that way. The point is that you should know the number before it is your problem rather than after.
The three questions that decide whether it fits
How long will you hold the loan, honestly? Not how long you plan to love the house, but how long the loan survives, which is a different question once you account for a move, a job change, or a refinance that may or may not be available at a better rate. Second, can you carry the capped payment without changing how you live? If the answer is no, the ARM is not saving you money, it is lending you optimism. Third, what is the actual cost of being wrong? A borrower five years from retirement and a borrower who expects to relocate for work in three years are looking at the same loan with completely different downside. An ARM is a reasonable instrument for a short, known horizon. It is a poor substitute for a house you cannot afford at the fixed rate.
Why the two rate numbers you keep seeing do not match
You may have noticed that the figures above do not line up with the number in the news. Freddie Mac's Primary Mortgage Market Survey for the week of October 15, 2026 put the 30-year fixed at 7.40% and the 15-year fixed at 6.71%. The MBA survey, for the week ending October 9, had the conforming 30-year at 7.50% and the 15-year at 6.87%. Both are accurate. They measure different things over different weeks: Freddie Mac surveys lenders on what they are offering, while the MBA measures rates on applications that were actually submitted, and MBA reports points alongside the rate. Neither one is the rate you will be quoted, because yours depends on your credit, your down payment, your loan size, your property type, and how many points you are willing to pay. Treat both as a weather report, not a price tag.
What to do with this in New England
If an ARM is on the table, ask for both quotes in writing, on the same day, with points disclosed, and compare the total cost over the years you actually expect to hold the loan rather than over 30. Ask your loan officer for the index, the margin, the first adjustment cap, the periodic cap, the lifetime cap, and the adjustment frequency, and ask them to show you the fully indexed payment at the lifetime cap. A lender who will not put that on paper has told you something useful. And be clear with yourself about what the 10.5% figure represents. It is not a signal that ARMs are suddenly a better product. It is a measure of how many buyers have run out of room at 7.50%, which is a reason to look hard at the loan, and an even better reason to look hard at the price you are agreeing to pay.
Frequently asked questions
Is an adjustable-rate mortgage a bad idea in 2026?
Not inherently, and it is not the 2006 product either. Today's ARMs are fully underwritten and carry caps on how far and how fast the rate can move. Whether it is a bad idea for you comes down to your time horizon and whether you can carry the payment if it adjusts to its ceiling. If you need the ARM to afford the house at all, that is the warning sign, not the loan type.
How much lower is an ARM rate than a 30-year fixed right now?
In the MBA survey for the week ending October 9, 2026, the average contract rate on a 5/1 ARM was 6.54% against 7.50% on a 30-year fixed conforming loan, a gap of 0.96 percentage points. Keep points in view when you compare: that ARM rate carried 1.13 points versus 0.83 on the fixed, so some of the gap was purchased up front.
What does the 5 in a 5/1 ARM mean?
The first number is how many years the initial rate is fixed, and the second is how often the rate can adjust after that. A 5/1 is fixed for five years, then adjusts annually. The rate after the fixed period is an index plus a margin stated in your note, limited by the caps in your note, so read those two documents rather than relying on the product name.
Can I refinance out of an ARM before it adjusts?
Sometimes, and it is a plan rather than a guarantee. Refinancing requires that rates, your credit, your income, and your home's appraised value all cooperate at the moment you need them to, and none of that is promised five years out. Build your decision around being able to carry the adjusted payment, and treat a future refinance as upside rather than as the exit.
Why are refinance applications down so much?
Because very few homeowners have a rate worth replacing. The MBA's Refinance Index was 56% below the same week a year earlier for the week ending October 9, 2026, which is what happens when the prevailing rate sits well above the rate most existing borrowers already hold. Refinance activity tends to return only when rates fall meaningfully below the rates already on the books.
Sources: Mortgage Bankers Association Weekly Applications Survey for the week ending October 9, 2026, and the Freddie Mac Primary Mortgage Market Survey for the week of October 15, 2026. Payment figures are illustrative calculations.
