Fixed vs Adjustable Mortgage 2026: Which Costs Less?

Fixed-Rate Mortgages: Predictability and Stability

A fixed-rate mortgage has an interest rate that remains constant for the entire life of the loan. Whether you choose a 15-year or 30-year fixed loan, your monthly principal and interest payment will never change. That certainty is the product's entire pitch: you know exactly what you will pay in year one and in year twenty-nine, regardless of what the economy does in between.

This stability makes budgeting easy and protects you from rising interest rates. It is the most popular choice for buyers who plan to stay in their homes for the long term, and in October 2026 the price of that certainty is visible in the numbers. Freddie Mac's October 8, 2026 survey put the average 30-year fixed at 7.40% and the 15-year fixed at 6.73%, both up from a year earlier. Every borrower who chooses fixed at these levels is paying a premium for predictability, which is exactly why the ARM comparison below deserves a fair hearing.

Adjustable-Rate Mortgages (ARMs): Lower Introductory Rates

An adjustable-rate mortgage (ARM) starts with a lower initial interest rate than a comparable fixed-rate mortgage. This initial rate is locked for a set period (usually 5, 7, or 10 years). During that fixed window the loan behaves exactly like a fixed-rate mortgage: same payment, same amortization math, no surprises.

After the initial period ends, the interest rate adjusts periodically based on a financial index plus a fixed margin. If market rates rise, your interest rate and monthly payment will increase. If rates drop, your payment may decrease. In early October 2026, 5/1 ARM start rates were running near 6.60%, roughly 0.8 points below the 30-year fixed average, and ARMs accounted for about 10.3% of mortgage applications, their highest share in a year. Borrowers are clearly doing the math again.

Understanding ARM Rate Caps

To protect borrowers, ARMs include interest rate caps that limit how much your rate can increase. These caps are usually structured as three numbers (e.g., 2/2/5). The first number is the initial adjustment cap (the maximum rate increase at the first adjustment date). The second number is the periodic adjustment cap (the maximum increase at each subsequent adjustment). The third number is the lifetime cap (the absolute maximum rate increase over the life of the loan).

Walk through a 2/2/5 cap on a 5/1 ARM starting at 6.60%. At the first adjustment, the rate can rise to at most 8.60%, no matter what the index says. At each later annual adjustment, it can rise at most another 2 points from wherever it stands. And over the whole life of the loan, it can never exceed 11.60% (start rate plus 5). Caps apply to decreases symmetrically in most contracts, so a falling index can only pull your rate down by the capped amount per period.

With six-month adjustment products, the middle number is usually 1, as in 2/1/5: up to 2 points at the first adjustment, then up to 1 point every six months, with a 5-point lifetime ceiling. The lifetime cap is the number to stress-test against, because it defines your true worst case. A 2/2/5 structure sounds protective until you translate it: on a 400,000 dollar loan, the lifetime-cap payment can run more than 1,000 dollars a month above the fixed-rate alternative. Caps limit the damage; they do not prevent it.

5/1, 7/1, 10/6: How to Read the ARM Label

The two numbers in an ARM name tell you the two things that matter most. The first number is how many years the initial rate is fixed. The second number is how often the rate adjusts after that, and this is where you need to read carefully because the convention changed. In the classic format, 5/1 means five years fixed, then adjustments once per year. 7/1 means seven years fixed, then annual adjustments.

Newer products use months for the second number. A 5/6 ARM is fixed for five years, then adjusts every six months. A 10/6 is fixed for ten years, then adjusts every six months. The six-month adjustment interval is now the common structure for newly originated ARMs, replacing the old annual-adjustment 5/1 as the default quote. Always confirm which interval your quote uses, because a loan that can adjust every six months reaches its caps twice as fast as one that adjusts annually.

Less common variants exist too: 3/1 and 3/6 for borrowers with a three-year horizon, and 7/6 or 10/6 for borrowers who want a longer fixed runway without paying full 30-year fixed pricing. The naming never tells you the caps or the margin, so treat the label as the headline and the loan estimate as the fine print.

The Index, the Margin, and How Your Adjusted Rate Is Actually Set

When the fixed period ends, your new rate is not chosen by the lender's mood. It is computed from a formula: index plus margin, rounded per the loan's terms, then constrained by the caps. The index is a published benchmark rate that moves with the market. Since LIBOR ended in June 2023, newly originated US ARMs use SOFR, the Secured Overnight Financing Rate, in most cases.

The margin is the lender's fixed markup, set at origination and never changing. A common margin is around 2.75%. So if SOFR is 4.30% at your adjustment date, your fully indexed rate is 4.30 + 2.75 = 7.05%. That is the rate the formula wants to give you, before caps are applied. If your start rate was 6.60% and your first-adjustment cap is 2 points, the rate can rise to at most 8.60% at the first adjustment even if the formula says 9%.

Two details borrowers miss: the floor and the lookback. Most ARMs have a floor, often equal to the margin, so your rate cannot fall below it no matter how far the index drops. And the index value used is typically the one published 30 to 45 days before the adjustment date, not the value on the day itself. Neither detail changes the big picture, but both belong in your spreadsheet before you sign.

Rates at a Glance: October 2026

Figures updated October 2026. Freddie Mac's Primary Mortgage Market Survey for the week ending October 8, 2026 reported the 30-year fixed-rate mortgage averaging 7.40%, up from 7.28% the prior week and 6.30% a year earlier. The 15-year fixed averaged 6.73%, up from 6.60% the prior week. These are averages for conforming loans to borrowers with strong credit and 20% down; individual quotes vary with credit score, loan size, and location.

For ARMs, early-October 2026 pricing put 5/1 ARM start rates near 6.60%, roughly 0.8 points below the 30-year fixed average, with 7/1 products priced a little higher than 5/1 and a little lower than fixed. The spread is the whole economic argument for the ARM: you are paid about three-quarters of a point to accept adjustment risk after year five. The 10-year Treasury, the benchmark fixed rates loosely follow, was hovering around 5.25% in early October, which explains why both fixed and ARM pricing sit where they do.

Two warnings about using these figures. First, the spread between fixed and ARM pricing moves with the yield curve; when the curve is flat, the ARM discount shrinks and the trade looks worse. Second, the rate you are quoted is a start rate, not a promise. Every number in the worked scenario below uses these October 2026 figures so you can see the actual dollars involved.

• How Fed rate decisions move mortgage pricing

Worked Scenario: 400,000 Dollars at 7.40% Fixed vs 6.60% 5/1 ARM

Take a 400,000 dollar loan, the kind of figure a buyer in many US markets is financing in 2026. The 30-year fixed at 7.40% costs 2,770 dollars a month in principal and interest. The 5/1 ARM at a 6.60% start rate costs 2,555 dollars a month for the first five years. Same loan, same term, 215 dollars less every month.

Over the five-year fixed window, the ARM borrower pays 12,893 dollars less in total. That is real money: it covers moving costs, furnishing, or a healthy head start on an emergency fund. After 60 payments, the ARM balance stands at about 374,872 dollars, and the borrower has also paid down slightly more principal than the fixed-rate borrower would have, because more of each early payment went to principal at the lower rate.

So far the ARM is winning comfortably. The question is what happens at month 61, when the rate is free to move. The next two sections run the two paths that matter: the bad-but-plausible adjustment and the genuine worst case.

For reference, the fixed loan's lifetime cost is fully knowable: 2,770 dollars x 360 payments = 997,200 dollars total, of which about 597,200 dollars is interest. The ARM's lifetime cost cannot be known in advance. That asymmetry is the entire trade in one sentence: the fixed loan sells you a fact, the ARM sells you a distribution.

ARM vs Fixed: Side by Side on the Same Loan

It helps to see the two products on identical terms. A 30-year fixed at 7.40% on 400,000 dollars: payment 2,770 dollars from day one to payoff, total interest about 597,200 dollars over the full term, zero adjustment risk, best for horizons beyond ten years. A 5/1 ARM at 6.60% with 2/2/5 caps on the same loan: payment 2,555 dollars for five years, then anywhere from 2,324 dollars (if rates fall to 5.60% at the first adjustment) to 3,044 dollars (if the first cap binds at 8.60%) to 3,803 dollars (if rates climb to the 11.60% lifetime cap), best for horizons under seven years.

The fixed loan's cost is a fact; the ARM's cost is a distribution. Borrowers who need a fact, because their budget has no slack or their horizon is the full thirty years, should buy the fixed loan. Borrowers who can live with a distribution, and whose plans end before the distribution gets wide, can reasonably buy the ARM. Most bad ARM stories are not product failures; they are horizon mismatches, where a borrower bought a five-year product for a fifteen-year plan.

The Break-Even Question: When Does the ARM Stop Winning?

Suppose that at the first adjustment, the index-plus-margin formula calls for a rate well above the 2-point initial cap, so the loan adjusts to the maximum allowed: 8.60%. The payment is recalculated on the remaining balance of 374,872 dollars over the remaining 300 months, and it comes out to 3,044 dollars a month. That is 274 dollars more than the fixed-rate payment of 2,770 dollars.

Now the ARM is losing 274 dollars a month, and it has 12,893 dollars of accumulated savings to burn through. At that pace the savings last about 47 months. The break-even point lands around month 107, during the ninth year. Sell or refinance before then under this rate path, and the ARM still wins overall. Stay past it, and the fixed rate was the better choice.

Now run the opposite path. Suppose rates fall and the ARM adjusts down to 5.60% at month 61. The recalculated payment drops to about 2,324 dollars a month, which is 446 dollars less than the fixed payment, and the ARM keeps winning for as long as rates stay down. This is the honest shape of the trade: the ARM wins if your horizon is short or rates cooperate, and it loses if you stay long into rising rates. There is no scenario where the ARM wins forever without rate luck, because the fixed rate never moves.

The Worst Case: What the Lifetime Cap Really Costs

The lifetime cap on a 2/2/5 ARM starting at 6.60% is 11.60%. If rates climb relentlessly and each adjustment hits its cap, the payment on the remaining balance eventually reaches about 3,803 dollars a month. That is 1,033 dollars a month more than the 2,770 dollar fixed payment, every month, for years.

This is the number to put in front of anyone considering an ARM on a tight budget. The caps did their job: without them, the payment could have gone higher still. But 'capped' and 'affordable' are different words. Before choosing an ARM, run your household budget at the lifetime-cap payment, not the start-rate payment. If the budget breaks at the cap, the ARM is not a discount; it is a gamble you cannot afford to lose.

One more structural point: the worst case assumes you stay in the loan. In practice, borrowers who face a painful adjustment refinance or sell. That exit is only available if your home has not lost value and your credit and income still qualify you for a new loan. The 2008 vintage of ARM borrowers learned this the hard way: the adjustment you cannot refinance away from is the one that hurts.

When an ARM Wins

An ARM is the rational choice in a handful of well-defined situations. You plan to sell or refinance within the fixed period: the classic case is a starter home you expect to outgrow in five to seven years, or a borrower who expects a credit-score improvement that will unlock a refinance. You expect meaningfully higher income later: a medical resident, a junior partner on a partnership track, or anyone whose earnings will step up can use the lower early payments and absorb a later adjustment. You have the discipline to save the difference: banking the 215 dollars a month from the worked scenario builds a cushion that covers the adjustment if rates rise.

Rate structure also matters. On jumbo loans, where the fixed-vs-ARM spread is often wider than on conforming loans, the savings during the fixed window are proportionally larger. And if you genuinely believe rates will fall, perhaps because you expect the Federal Reserve to ease, an ARM lets you capture the decline automatically without paying refinance closing costs. Just never confuse a forecast with a plan: hope is not a hedge.

When an ARM Punishes You

An ARM is the wrong choice when your time horizon is long and uncertain. If this is your 'forever home' or you cannot say when you will move, you are buying adjustment risk without a matching benefit. It is also wrong when your budget only works at the start rate: if a 2-point adjustment would break your monthly cash flow, the product does not fit, full stop.

Be careful with two traps. The first is the interest-only ARM, sometimes marketed alongside standard ARMs: during the interest-only period you build zero equity from payments, so a flat housing market leaves you with nothing to show for years of payments. The second is assuming you will refinance later. Refinancing requires equity, income, and credit at the time you need it, and all three can deteriorate at exactly the moment rates rise. If your ARM only works with a future refinance, you have not bought a mortgage; you have bought an option on your future self.

How Lenders Qualify ARM Borrowers (Not on the Teaser Rate)

You do not get to qualify for an ARM at its start rate. Under ability-to-repay rules, lenders typically qualify you at the higher of the fully indexed rate (index plus margin) or the note rate plus 2%. On a 5/1 ARM starting at 6.60% with a 2.75% margin and SOFR at 4.30%, the fully indexed rate is 7.05%, but the note rate plus 2% is 8.60%, so you qualify at 8.60%.

This is consumer protection working as intended: it stops lenders from approving borrowers who can only afford the teaser. It also means the ARM's lower payment does not help you qualify for a bigger loan the way borrowers sometimes assume. If your debt-to-income ratio is tight, run the qualification math at the qualifying rate before you fall in love with the start rate.

Smarter Ways to Use an ARM

If your horizon is six or seven years, price the 7/1 ARM against the 5/1. The start rate is slightly higher, but you buy two extra years of certainty, which is cheap insurance if your moving plans slip. Match the fixed period to your horizon with a one-to-two-year buffer; the most common ARM regret is a fixed window that expires twelve months before the borrower is ready to move.

Compare the ARM against a temporary rate buydown on a fixed loan, not just against the plain fixed rate. A 2-1 buydown gives you lower payments in years one and two with a fixed rate thereafter, which suits a two-year horizon better than a five-year ARM commitment. And set a calendar reminder for month 54 of a 5/1 ARM: that is when you start shopping the refinance or listing the home, not month 61 when the adjustment letter arrives.

Sources

The figures and policy details in this article come from the following primary sources, checked at the time of writing.

• CFPB: Fixed vs adjustable-rate mortgages
• CFPB: ARM rate caps explained
• Federal Reserve: LIBOR transition (SR 23-2)
• Freddie Mac Primary Mortgage Market Survey