What break-even actually measures
The break-even month is the point at which the payments you have saved add up to the costs you paid to get them. Divide total closing costs by the monthly reduction and you have it: $5,000 of costs against a $284 saving is 17.6 months. If you will still own the house then, the refinance pays for itself; if you will have moved, you paid $5,000 for a benefit you did not collect.
That simple ratio is the industry's standard test, and it is right about the question it answers. But it quietly assumes two things worth naming. It assumes the whole saving is a saving — yet part of a lower payment can be principal you are no longer retiring, which is not money you kept. And it assumes the two loans run to the same finish line, which they do not when a 27-year balance is refinanced onto a fresh 30-year note.
So this calculator reports both tests. The break-even month answers “when does cash flow turn positive”. The net benefit answers the harder question — “if I sell in H years, am I richer or poorer for having refinanced” — by comparing everything you paid plus everything you still owe under each loan. When the new term is longer than the old one, the two tests can point in opposite directions, and the table below the results shows exactly where the crossover sits.
The three calculations behind the answer
The current payment. Your existing payment is rebuilt from the balance, the note rate and the months remaining, using M = L·r / (1 − (1 + r)−n). Rebuilding it rather than asking you to type it keeps the comparison internally consistent, and it is the same formula behind the mortgage payment calculator.
The new payment. Same formula, new rate, new term, and a loan amount that is your balance plus any cash taken out plus the closing costs if you finance them. Financing the costs is not free: it raises the loan, which raises the payment, which lengthens the break-even slightly compared with paying cash.
The comparison over a horizon. For each year, the calculator runs both loans forward and adds up what you paid plus what you still owe. The difference between those two totals is the net benefit, less any closing cost you paid in cash and plus any cash you took out. Because the balance is included, a lower payment achieved by stretching the term is correctly recorded as slower principal repayment rather than as pure gain.
Nothing is discounted for the time value of money. Discounting would make an early dollar worth more than a late one and would shorten every break-even slightly; leaving it out keeps the figures checkable against your own statements.
Worked example: $300,000 at 7.25% with 27 years left, refinanced to 6% for 30
Take the default case, with $5,000 of closing costs added to the new loan and no cash out.
- Current payment. r = 7.25 ÷ 1,200 = 0.00604167 and n = 324. (1.00604167)324 = 7.03996, so (1 + r)−324 = 0.1420452 and 1 − that = 0.8579548. M = 300,000 × 0.00604167 ÷ 0.8579548 = 1,812.50 ÷ 0.8579548 = $2,112.58.
- New loan amount. 300,000 + 5,000 of financed costs = $305,000.
- New payment. The 6% thirty-year factor is 5.99551 per $1,000, so 305 × 5.99551 = $1,828.63.
- Monthly saving. 2,112.58 − 1,828.63 = $283.95.
- Break-even. 5,000 ÷ 283.95 = 17.6 months, a year and a half.
- Lifetime interest. The old loan would cost 324 × 2,112.58 − 300,000 = $384,477 in interest. The new one costs 360 × 1,828.63 − 305,000 = $353,306. The change is −$31,170 — the refinance is cheaper over the full life even though the term stretches by three years.
- Net benefit after seven years. Keeping the old loan, you pay 84 × 2,112.58 = $177,457 and still owe $267,288, a total of $444,745. Refinancing, you pay 84 × 1,828.63 = $153,605 and still owe $273,400, a total of $427,005. The refinance leaves you $17,740 ahead.
Note that the seven-year net benefit is far larger than the 17.6-month break-even might suggest, because the saving keeps accruing long after the costs are recovered. Note too that the new balance after seven years is higher than the old one would have been — $273,400 against $267,288 — because you borrowed the $5,000 of costs and restarted a thirty-year amortisation. The saving is real; it is just smaller than the raw payment difference times 84.
How to read the two answers when they disagree
Compare the break-even against your honest holding period first. Households systematically overestimate how long they will stay, and a refinance that breaks even in 40 months is a poor bet for someone whose job history says three years. If the break-even is longer than your horizon, the reference table's net-benefit column will already be negative at that year, and the two measures agree.
They disagree in one specific regime, and it is common: a modest rate cut combined with a term extension. Look at the reference table below. Refinancing that $300,000 balance from 7.25% down to 6.5% still lowers the payment by $184.77 and still breaks even in 27 months — but total interest rises by $4,534, because the new loan runs 360 payments where the old one had 324. Below roughly 6.4% in that table the rate cut wins; above it, the term extension wins, and you are buying cash flow with lifetime cost. Neither answer is wrong; they measure different things, and you have to decide which one you are optimising.
The clean way out of that trade-off is to match the term. If you have 27 years left, refinancing into a 25- or 20-year loan captures the rate cut without restarting the clock. If your lender only offers round terms, take the 30-year note and pay it on your old schedule — the extra payment calculator shows what that costs and saves.
Finally, treat the closing-cost figure with suspicion until you have a Loan Estimate. A no-cost refinance is not free; the lender covers the costs and recovers them through a higher rate, which shows up here as a smaller monthly saving and, usually, a worse position at long horizons.
Refinancing a $300,000 balance with 27 years left at 7.25%, into a new 30-year loan
| New rate | New payment | Monthly saving | Break-even (months) | Change in total interest |
|---|---|---|---|---|
| 5.00% | $1,637.31 | $475.28 | 10.5 | −$100,046 |
| 5.50% | $1,731.76 | $380.83 | 13.1 | −$66,044 |
| 5.75% | $1,779.90 | $332.68 | 15.0 | −$48,714 |
| 6.00% | $1,828.63 | $283.95 | 17.6 | −$31,170 |
| 6.25% | $1,877.94 | $234.64 | 21.3 | −$13,419 |
| 6.50% | $1,927.81 | $184.77 | 27.1 | +$4,534 |
| 6.75% | $1,978.22 | $134.36 | 37.2 | +$22,684 |
| 7.00% | $2,029.17 | $83.41 | 59.9 | +$41,026 |
Every row lowers the monthly payment, because the term resets from 324 payments to 360. From 6.50% down the change in total interest turns positive: the payment falls and the lifetime cost rises together. That sign change is the reason the old rule of thumb about a one-point rate drop is unreliable.
Assumptions and traps in a refinance comparison
- The break-even ignores principal. Part of a lower payment can be principal you are no longer paying down. The net-benefit table corrects for this by including the balance still owed; the simple ratio does not.
- “No-cost” means the rate absorbed the cost. Lender credits are paid for through a higher note rate. Model the offer at the rate actually quoted and set closing costs to zero, then compare the net benefit against the paying-cash version.
- Points are closing costs. One point is 1% of the loan paid upfront to lower the rate. Include it in the closing-cost field and it will lengthen the break-even, which is the correct treatment.
- Escrow is not a cost and not a saving. At closing you fund a new escrow account and your old one is refunded, typically within 20 days. The two roughly cancel, and neither belongs in this comparison.
- Cash-out changes the question. Taking cash raises the loan and the payment, so the payment break-even may vanish entirely. Judge a cash-out refinance against what the cash costs from other sources, not against the old payment.
- Recasting may be the cheaper answer. If you have a lump sum and your rate is already competitive, a recast re-amortises the reduced balance at your existing rate for a few hundred dollars — no appraisal, no title, no new note.
- The APR on the two loans is not comparable across different terms. APR spreads costs over the loan's life, so a 15-year and a 30-year quote with identical fees show different APRs. Compare on payment, total interest and net benefit instead.
When a refinance is the wrong tool
Three alternatives solve problems people bring to refinancing, usually better. A recast lowers the payment after a lump-sum prepayment without changing the rate or paying closing costs. Extra principal shortens the term at zero cost — see the extra payment payoff calculator — and is the right answer when your rate is already good and you simply want out sooner. And a home equity line draws cash without touching a low first-mortgage rate, which matters enormously to anyone holding a note from the low-rate years.
Refinancing is the right tool in four situations: the rate available is materially below yours and you will stay long enough to recover the costs; you want to leave an adjustable rate before it adjusts; you want to remove mortgage insurance that will not otherwise cancel, which happens on FHA loans with less than 10% down; or you need to remove a borrower from the note after a divorce or a partnership ending.
Whatever the reason, run the numbers on the Loan Estimate rather than the advertisement. Regulation Z requires that document within three business days of application, on a standard form, with the closing costs itemised in comparable sections — which is precisely what makes two offers comparable. And re-check your affordability if anything else has changed since you bought; the home affordability calculator works from the same ratios your new lender will apply.