How to Calculate Your Refinance Break-Even With Rates Over 7%: The 33-Month Example on a $368,000 Mortgage and the Term-Reset Trap
Say you have a $368,000 balance at 7.75%, with 28 years left on the loan. This week a lender quotes you 7.05% to refinance. That's a 0.70-point drop, and your gut says "obviously do it."
Then you look at the closing costs and the calendar. The Fed raised rates on Wednesday. Mortgage rates sit over 7%. You start wondering whether to lock now or wait for a dip. Meanwhile your existing payment keeps ticking.
This guide is the arithmetic for that moment. I'll show the formula, run it on the $368,000 example, and show where it breaks. Every loan figure below is a worked example I built for illustration, including the 7.75% existing rate, the 7.05% quote, and the closing costs. The market context comes from the articles cited. Your numbers will differ based on your specific situation, and that's the point of the exercise.
What This Week's Data Says (and Doesn't Say)
Here is what the sources actually tell us.
NerdWallet's "Mortgage Rates Today, Thursday, September 17: Fed Hikes, Rates Over 7%" reports that mortgage rates had already absorbed the Fed's widely anticipated increase. The next day, "Mortgage Rates Today, Friday, September 18: No Change," described a breather as bond markets digested the week's Fed news.
The Bureau of Labor Statistics' Major Economic Indicators page shows the backdrop:
- CPI: +0.4% in August 2026
- Unemployment: 4.1%
- Payroll employment: +162,000 (preliminary)
- Average hourly earnings: +$0.10 (preliminary)
A 0.4% monthly CPI reading compounds to about 4.9% annualized (1.004¹² ≈ 1.049) if it repeated for a year. One month isn't a trend, but it explains why nobody can promise you rates are about to fall.
The honest takeaway is that "no change" on Friday is not a forecast. It's a pause, and it's the only rate data you have. You can refinance on today's quote or model what waiting costs. Neither is free.
The 3-Step Refinance Break-Even Formula
Step 1: Compute your current payment and the new payment.
Monthly principal-and-interest payment = B × r ÷ (1 − (1 + r)⁻ⁿ)
Here B is the balance, r is the annual rate divided by 12, and n is the number of months. Use principal and interest only. Leave out taxes and insurance, because they don't change.
Step 2: Subtract to get monthly savings.
Current payment − new payment = monthly savings.
Step 3: Divide your total closing costs by monthly savings.
Closing costs ÷ monthly savings = months to break even.
Now run it on the example.
- Current loan: $368,000 at 7.75%, 336 months left → $2,685/month
- New loan: $368,000 at 7.05%, 360 months → $2,461/month
- Monthly savings: $225
- Closing costs (assumed): $7,400, about 2% of the balance
- Break-even: $7,400 ÷ $225 = about 33 months
Someone with a 2.7-year plan to stay is roughly at zero. Someone planning to stay 7 years is comfortably ahead. That difference is why a generic "refinance if you can drop the rate by a point" rule fails. It has no idea how long you're staying.
Net Savings by How Long You Stay
| Months in the home after refi | Gross savings ($225/mo) | Minus $7,400 costs | Net position |
|---|---|---|---|
| 24 | $5,394 | −$7,400 | −$2,006 |
| 36 | $8,091 | −$7,400 | +$691 |
| 60 | $13,486 | −$7,400 | +$6,086 |
| 84 | $18,880 | −$7,400 | +$11,480 |
This is the kind of analysis Kavivero runs for you, so you don't have to build the spreadsheet yourself.
The Term-Reset Trap: Where the Savings Quietly Shrink
The table above hides something. Your old loan had 336 months left. The new one has 360. You've added 24 payments.
Here is the full-life comparison for the example:
- Keep the old loan: $2,685 × 336 = about $902,300 in remaining payments
- Refinance: $2,461 × 360 = about $885,800, plus $7,400 in costs = about $893,200
The full-life savings are about $9,000, not the $75,000 you'd get by multiplying $225 by 336 months. The extra 24 payments of $2,461, about $59,000, absorb most of it.
None of this makes the refinance wrong. If the $225/month matters to your budget today, that's a real benefit. If you're likely to sell in 5 to 7 years, the reset barely matters. It matters a lot if you plan to hold the loan to the end and hate the idea of an extra two years of payments.
One fix is to keep paying the old $2,685 on the new loan, which shortens the payoff. Another is to ask for a 25-year or 20-year term. Both change the math, and your quote will tell you which are on the table. I dug into the same issue in the hidden cost of resetting your loan clock at rates over 7%.
How Much One Rate Move Changes the Answer
The rate you can lock in the next few days may not equal the rate in the last quote. Here is the same $368,000 example with the new rate shifted by a quarter point (closing costs held at $7,400):
| New rate | New payment | Monthly savings | Break-even |
|---|---|---|---|
| 6.80% | $2,399 | $286 | 26 months |
| 7.05% | $2,461 | $225 | 33 months |
| 7.30% | $2,523 | $163 | 46 months |
A quarter-point move shifts the break-even by 7 months in one direction and 13 in the other. It isn't linear, and rules of thumb can't capture that. It also means a rate lock can matter as much as the rate itself.
Cost of Waiting: Is a Better Rate Worth the Delay?
Suppose you wait six months hoping for 6.80%. You give up six months of $225 savings, or $1,349. If the rate does hit 6.80%, you save an extra $61.53/month ($286.29 vs $224.76). That's about 22 months to earn back the cost of waiting.
If rates don't fall, or they rise to 7.30%, you've lost the $1,349 and face a 46-month break-even. Waiting is a bet on direction. With CPI at +0.4% for August and the Fed having just raised, I wouldn't call it a free option. I wouldn't call it a bad one either. It depends on how long you'll stay and whether you can absorb being wrong.
You can also refinance now and again later if rates fall. But a second refinance means a second set of closing costs, and the break-even clock restarts. For the fuller framework, see the 5-question checklist for a refinance after the September Fed hike.
Rate-and-Term vs Cash-Out on the Same House
Now add a second variable. You want $50,000 for a roof and some debt cleanup. Same house, assumed value of $530,000 (for illustration).
| Rate-and-term | Cash-out | |
|---|---|---|
| New balance | $368,000 | $418,000 |
| Rate (assumed) | 7.05% | 7.30% |
| Closing costs (about 2%) | $7,400 | $8,360 |
| Monthly payment | $2,461 | $2,866 |
| LTV | 69.4% | 78.9% |
I assumed a 0.25-point rate premium on cash-out. That's common in pricing, though your quote may differ. The payment difference is $405/month, and it breaks down like this:
- About $343 is the payment on the extra $50,000.
- About $62 is the cost of repricing your existing $368,000 at the higher rate.
That second piece is easy to miss. The premium applies to your whole balance, not just the new money.
Over 30 years, the cash-out route costs about $146,800 more in payments and fees than rate-and-term to put $50,000 in your hands. That figure is nominal and not discounted, and it assumes you hold the loan the full term. If the cash retires debt costing far more than 7.30%, or funds a repair that protects the home's value, the trade may still make sense. If it funds something that doesn't pay you back, you're paying about 2.9× the amount you borrowed.
Watch the LTV line. Many lenders cap cash-out around 80% loan-to-value. If your home's appraisal comes in at $510,000 instead of $530,000, that $418,000 balance is a 82% LTV, and the cash-out may not be approvable at that size. Home price movement can decide whether the option exists at all.
For a fuller side-by-side, see rate-and-term vs cash-out at 7.04%, including the true cost gap. Also compare a HELOC or home equity loan quote against the cash-out numbers, since a second lien leaves your first mortgage untouched.
When "Free Money" Isn't Free: Lender Credits and Closing Costs
NerdWallet's "Locked Out: Should You Take 'Free Money' to Buy a Home?" looks at homebuying assistance programs that lower upfront costs and says to weigh the trade-offs first. Refinance has a cousin of this: the no-closing-cost refinance, where the lender covers your fees in exchange for a higher rate.
Here's the math on an illustrative offer. The lender covers the $7,400 and quotes 7.55% instead of 7.05%.
- Payment at 7.55%: $2,586
- Payment at 7.05%: $2,461
- Difference: about $125/month more to skip the upfront $7,400
$7,400 ÷ $125 = about 59 months. If you'll sell or refinance again before then, the credit route wins. If you stay longer, you'll have paid more than the fees you avoided. Compare net positions:
| Horizon | Pay costs at 7.05% | Lender credit at 7.55% |
|---|---|---|
| 24 months | −$2,006 | +$2,393 |
| 60 months | +$6,086 | +$5,982 |
| 84 months | +$11,480 | +$8,375 |
The 7.55% pricing is my assumption. Real lender-credit pricing varies, so get both quotes side by side. As with any "free money" offer, the question is what it costs and when it stops being a good deal.
You can model this for your specific situation at Kavivero, including the extra term and the credit trade-off.
Decision Grid: Which Factor Tips Your Answer?
| If your situation looks like this... | It tends to point toward... | Why |
|---|---|---|
| Staying 7+ years, old rate well above quote | Rate-and-term, paying costs | Lowest rate, longest payoff of the upfront cost |
| Moving or refinancing within 2–4 years | Lender credit or no refinance | The break-even may never arrive |
| Need cash, LTV near 80% | Compare cash-out with HELOC | The 80% cap and full-balance repricing both bite |
| Rate quote within 0.25 point of current | Probably wait or skip | Break-even stretches past 46 months quickly |
| Unsure about rate direction | Model both 6.80% and 7.30% | Sensitivity shows the range of outcomes |
Why Your Numbers Will Differ
Everything above rests on assumptions that are mine, not yours:
- Your remaining term. A loan that's 10 years in has a different reset penalty than one 2 years in.
- Your actual closing costs. I used about 2%. Some quotes come lower, and some include points.
- Your time horizon. The single largest driver. Being honest about it moves the break-even more than any rate change.
- Your equity. Appraised value decides cash-out eligibility and whether mortgage insurance is in play.
- Your alternatives. A HELOC, paying down debt from savings, or doing nothing.
The math is simple, but there's a lot of it, and each assumption changes the answer. If you want to see how the same steps play out at another balance, my 3-step break-even walkthrough for a $364,000 mortgage covers that.
Run Your Own Numbers Before the Next Rate Move
A Friday "no change" is a breather, not a signal. You can't predict the next move, and with August CPI at +0.4% and the Fed having just hiked, nobody can promise you a better rate. What you can do is decide what rate and closing-cost combination makes sense for your own balance, term, and time horizon, and act only if the math clears. If it doesn't, waiting is a perfectly good answer.
Put your balance, rate, remaining term, and quote into Kavivero. You'll see your break-even, the term-reset cost, and the rate-and-term vs cash-out comparison for your situation. That way you're deciding on math, not on headlines.
Sources
- Mortgage Rates Today, Friday, September 18: No Change — NerdWallet
- Major Economic Indicators Latest Numbers — Bureau of Labor Statistics
- Mortgage Rates Today, Thursday, September 17: Fed Hikes, Rates Over 7% — NerdWallet
- How I Earned 1 Million Points With My Family Cruise Booking — NerdWallet
- Locked Out: Should You Take ‘Free Money’ to Buy a Home? — NerdWallet