Is Single-Premium PMI Cheaper Than Monthly PMI? One $350,000 Loan, Followed to Cancellation

Short answer: on this example loan, single-premium PMI becomes the cheaper choice at about month 42 — three and a half years — and stays cheaper after that. The surprise is how little the investment-return assumption matters: across returns from 0% to 10% a year, the crossover only moves between month 37 and month 44.

Rather than argue the two structures in the abstract, this post follows one borrower through one loan, from the closing table to the date the servicer has to cancel PMI without being asked, and prices both choices at each stop along the way.

New single-family homes under construction in a U.S. suburban development, a generic illustration not tied to this article's specific property

Photo: New homes under construction in Winchester, California (generic illustration, not the property described in this article). Credit: Z3lvs, Wikimedia Commons, CC0 Public Domain

Key takeaways

  • Cash-outlay crossover: month 37. Crossover after charging the upfront premium an opportunity cost at 7% a year: month 42.
  • The return assumption barely moves the answer. At 0% it is month 37, at 5% month 40, at 7% month 42, at 10% month 44.
  • Monthly PMI on this loan ends on its own at month 119 (78% of original value). The borrower can ask at month 104 (80%) — but only in writing, current on payments, with no junior lien and evidence the home has not lost value.
  • Financing the premium raises the loan to $354,200, which is 91.08% LTV — outside the 85.01–90% bracket the 1.20% rate was quoted in, so that rate would not survive the change.
  • Over the full 30 years the financed premium costs about $10,386 in payments, against $13,536 of monthly premiums if the loan runs to month 119 untouched.

Figures checked: November 7, 2026.

Closing day: the loan, the rates, and where each number comes from

One loan carries the whole post: a $350,000 mortgage at 90% loan-to-value — a $388,889 home with $38,889 down — on a 30-year fixed note at 7.32%. That is the Freddie Mac Primary Mortgage Market Survey average for the week of October 22, 2026, re-confirmed against the Federal Reserve's MORTGAGE30US series on November 7: 7.32% that week, 7.03% four weeks earlier on September 24, and 6.19% in the same week of 2025. The 15-year average that week was 6.68%. Rates have since eased for three consecutive weeks to 7.18% on November 5, which does not change the structure of the comparison but does mean a borrower pricing this today should rerun the payment line.

Principal and interest on that note come to $2,404.26 a month, recomputed for this post. Two mortgage-insurance figures sit on top of it, both of them one insurer's published example rather than a market rate: a non-refundable single premium of 1.20% of the loan amount (a 760-plus credit score, the 85.01–90% LTV bracket, 25% coverage, loan terms over 20 years) and a monthly premium of about 0.39% of the loan amount a year for a 90% LTV, 30-year fixed loan.

Those two rates were attributed to MGIC's published rate card when this post first ran. On November 7, 2026 the insurer's public rate page returned a 404, so the two percentages could not be re-read at their source and are carried here as the example's stated rates, not as a verified current quote. Anyone making this decision should put their own lender's two written quotes into the arithmetic below; the structure of the answer survives a different pair of rates, the crossover month does not.

On this loan that produces the two numbers everything else is built from: $350,000 × 1.20% = $4,200 due at closing, or $350,000 × 0.39% ÷ 12 = $113.75 a month.

Month 0 to month 37: the crossover a calculator will show

At the closing table the single-premium borrower is $4,200 poorer and has no PMI line in the payment. The monthly-premium borrower keeps the $4,200 and writes $113.75 extra every month. Dividing one by the other gives the number nearly every online tool reports: $4,200 ÷ $113.75 = 36.9 months. Past that point, on cash out of pocket alone, the single premium has cost less.

That is also the number that matters to a borrower who already knows they are selling or refinancing early. At three years the monthly borrower has paid $113.75 × 36 = $4,095 — $105 less than the single premium — and has had the $4,200 available the entire time. A standard non-refundable single premium returns nothing if the loan ends the month after closing, which is the whole asymmetry of the choice: one path is reversible and the other is not.

Month 42: the crossover once the cash is given a job

The $4,200 a monthly-PMI borrower keeps is not idle. Charging both paths the same assumed return — 7% a year, an illustration rather than a forecast — the single premium's foregone growth is compared against the growing pile of premiums the monthly borrower has handed over. The two future values cross at month 42.

Time in the loanFuture value of the $4,200 given upFuture value of monthly premiums paidCheaper path
1 year$4,494$1,408Monthly, by $3,086
3.5 years (month 42)$5,322$5,375Crossover — single premium ahead by about $53
5 years$5,891$8,099Single premium, by $2,208
7 years$6,744$12,187Single premium, by $5,443
Month 119 (automatic cancellation)$8,216$19,235Single premium, by $11,019

Every figure above assumes a 7% annual return on uninvested cash, no extra principal, and no home-price appreciation. The first version of this post said the 7% assumption "is doing real work." Recomputed across a range of returns, it does not — which is a more useful finding than the crossover itself, because it means the decision does not hinge on a number nobody can know.

Assumed annual return on the cashCrossover monthShift from the cash-only answer
0% (cash kept, not invested)Month 37—
3%Month 39+2 months
5%Month 40+3 months
7%Month 42+5 months
10%Month 44+7 months

The whole 0%-to-10% range fits inside seven months. A borrower who is sure about a three-year horizon and a borrower who is sure about a seven-year horizon both get the same answer no matter which return they assume; only someone landing between roughly month 37 and month 44 would see the assumption change the verdict, and nobody can forecast a holding period that precisely anyway.

Month 104 and month 119: the two dates monthly PMI ends by itself

Monthly PMI on this loan has a built-in expiry, and it has two dates rather than one. Both are set against "original value," which the Consumer Financial Protection Bureau defines as the contract sales price or the appraised value at purchase, whichever is lower — $388,889 here.

TriggerBalanceMonth on this loanWhat the borrower must do
Borrower-requested cancellation at 80% of original value$311,111Month 104Request in writing, be current on payments, certify no junior lien, show the home has not lost value
Automatic termination at 78% of original value$303,333Month 119Nothing — but the borrower must be current, or termination waits until payments are caught up
Midpoint of the amortization scheduleAnyMonth 181Backstop only; on this loan the 78% date arrives first

The month figures are this post's own amortization arithmetic on the $350,000 note at 7.32% with no extra principal. The 15-month gap between month 104 and month 119 is worth $1,706 in premiums at $113.75 a month, and it is the one part of the monthly-PMI path a borrower can actually win by acting — which is also why the four request conditions matter. A second mortgage taken out in year five, or an appraisal that comes back below $388,889, can turn month 104 back into month 119. A single premium has no equivalent lever, because there is nothing left to cancel.

The branch where the premium is financed instead of paid

A borrower who likes the single-premium economics but does not want to hand over $4,200 at closing can usually roll it into the loan. Doing that on this example means borrowing $354,200 instead of $350,000, which raises principal and interest from $2,404.26 to $2,433.11 — an extra $28.85 a month, a quarter of the $113.75 monthly premium it replaces.

Two things about that branch were missing from the first version of this post, and both are arithmetic rather than opinion.

First, the loan-to-value ratio. $354,200 against an original value of $388,889 is 91.08%, which sits outside the 85.01–90% bracket the 1.20% rate was drawn from. A premium quoted in one LTV bracket does not survive being financed into a loan that lands in the next one up, so the $4,200 figure — and therefore the $28.85 — is the floor of that branch rather than its price. The question to put to a loan officer is specific: which loan amount and which LTV bracket is the single premium quoted against, before or after financing.

Second, the extra payment never stops. Monthly PMI on this loan ends at month 119 whether anyone acts or not; a financed premium is part of the note for all 360 months. Over the full term the financed branch costs $28.85 × 360 = $10,386, of which $6,186 is interest on the premium. Monthly PMI run untouched to automatic cancellation costs $113.75 × 119 = $13,536. Financing still wins by about $3,150 in nominal dollars on this example — but the two amounts are paid on completely different schedules, and a borrower who refinances in year four never reaches either total.

What one case cannot settle

Four things push against the table above, and none of them is a number this post can supply.

  • Liquidity at closing. $4,200 spent on a premium is $4,200 not available for a repair, a move, or an appliance. The monthly path keeps it reachable, and that option has a value no spreadsheet row captures.
  • An unknown holding period. The crossover sits between month 37 and month 44, which is close to the range in which a job change, a growing household, or a rate drop worth refinancing for is most likely to appear. Few buyers know in advance which side of three and a half years they will land on.
  • Refundable single premiums exist. Some insurers sell a version that returns a prorated amount if the loan ends early, charging more upfront for less downside. That product was not priced here, and its break-even is necessarily later than the non-refundable one.
  • The two rates are one example. Credit score, coverage level, insurer and LTV bracket all move both percentages, and as noted above neither could be re-read at its source on November 7, 2026.

Where confidence, not arithmetic, decides it

If this had to be decided today on the arithmetic alone, the split would be about confidence in the holding period rather than about the return assumption. A borrower who can say with some conviction that the loan will still be in place past four years, and who can pay $4,200 without touching an emergency fund, is very likely to spend less in total on the single premium — the five-year, seven-year and month-119 rows only widen. A borrower who is unsure, or who would be cash-thin after closing, keeps more optionality on the monthly path: the cash stays reachable, month 104 can be pulled earlier with extra principal, and nothing is sunk. The financed branch is worth asking about by name, with the LTV-bracket question attached, because its headline $28.85 is the one figure in this post most likely to change when a real quote arrives.

FAQ

Is single-premium PMI cheaper than monthly PMI?

On this $350,000 example it becomes cheaper at about month 42 — month 37 on cash out of pocket alone — and the gap widens after that. Before roughly three and a half years, monthly PMI costs less and keeps the cash reachable.

How is the PMI break-even actually calculated?

The simple version divides the upfront premium by the monthly premium: $4,200 ÷ $113.75 = 36.9 months. The fuller version compares the future value of the $4,200 given up against the future value of the premiums paid month by month, which moves the crossover to month 42 at an assumed 7% return.

Does the assumed investment return change the answer?

Very little. Across assumed returns from 0% to 10% a year the crossover only moves from month 37 to month 44, so the decision turns on the expected holding period rather than on the return assumption.

When does monthly PMI cancel on a 90% LTV loan?

On this example, automatic termination at 78% of original value falls at month 119, and a borrower request at 80% is possible from month 104. The request must be in writing, the borrower must be current, there must be no junior lien, and the home must not have lost value.

Can I finance the PMI single premium into my mortgage?

Usually yes. On this loan it raises the balance to $354,200 and the payment by $28.85 a month. It also raises LTV to 91.08%, outside the bracket the 1.20% rate came from, so ask which loan amount and bracket the premium is quoted against.

Is a single PMI premium refunded if I sell or refinance early?

A standard non-refundable single premium returns nothing, which is why the early-exit case favours monthly premiums. Refundable versions exist at a higher upfront cost and were not priced in this example.

Related reading

Tip: Break-even math like this is easier to trust when you have run it yourself against your own closing numbers rather than a lender-supplied estimate. A dedicated financial calculator and a simple budget planner make it easier to track what the premium choice costs you month to month. (These are Amazon Associate links — we may earn a small commission on qualifying purchases.)

Correction November 7, 2026: the claim that the 7% return assumption "is doing real work" is replaced with a sensitivity table showing the crossover moves only from month 37 to month 44 across 0% to 10% returns; the borrower-request date at 80% of original value is added as month 104 along with the four conditions the CFPB requires, and the "current on payments" condition on automatic termination; the financed-premium branch now states that $354,200 is 91.08% LTV, outside the 85.01–90% bracket the 1.20% rate was quoted in, and that the extra payment runs for all 360 months; and the sources section, which previously carried no links, now links each figure to its origin. The two mortgage-insurance rates could not be re-read at the insurer's public rate page, which returned a 404 on this date, and are labelled accordingly. (Sources: Freddie Mac PMMS and the Federal Reserve MORTGAGE30US series for the rate, CFPB guidance on PMI cancellation, and this post's own amortization and future-value arithmetic.)

Sources

Editorial note and disclaimer: this post uses one illustrative $350,000 loan and one insurer's stated example rates. It is not a quote, an offer, or a promise of approval, savings, or any specific insurance cost, and it is not personalised financial advice. Mortgage insurance pricing varies by lender, insurer, credit score, loan-to-value ratio and coverage level; investment returns are never guaranteed at any assumed rate. Figures reflect United States federal rules in force on November 7, 2026. Readers making an actual PMI decision should obtain written quotes from their lender for both payment structures and, for anything involving taxes or significant sums, consult a licensed mortgage professional or financial adviser who can review their full situation.

Get new guides by email. Free, unsubscribe anytime.


Start Here: every guide by topic · About · Privacy Policy

Comments

Post a Comment

Popular posts from this blog

Is a HELOC Cheaper Than a Fixed Home Equity Loan for a $40,000 Renovation?

Image
A homeowner planning a $40,000 renovation usually checks one number first: which loan has the lower rate today. Right now that is the home equity line of credit, not the fixed-rate home equity loan. But a HELOC's rate is variable for the full repayment period, while a fixed home equity loan locks in today's rate for the life of the loan. Financed over 10 years, that difference can flip the real cost in either direction. So the honest way to read the comparison is not "which rate is lower" but "what is the one number that separates them actually made of." That number is $3,256.70, and the rest of this post takes it apart. Photo: Celestin bonaventure, Wikimedia Commons , CC BY-SA 4.0 — generic illustrative renovation photo, not tied to the specific loan figures below. Key takeaways As of November 4, 2026, the national average HELOC rate is 7.49%, versus 8.77% for a 10-year fixed home equity loan, per Bankrate's lender survey. On a $40,000, 10-year...
Read more

Car Loan or Mortgage First? What a 1.25-Point Rate Gap Actually Costs Over 5 Years

Image
A household has a $350,000 mortgage balance at today's going rate and a $28,000 car loan, and $300 a month they could put toward either one. The instinct for a lot of people is to attack the car loan first — it's the smaller balance, it feels winnable, and being "debt-free" on the car sooner has an obvious appeal. The math says something different, and the gap between the intuitive answer and the correct one is worth walking through with real numbers, because the size of that gap depends entirely on the interest rates involved — and in this specific case, it's smaller than you'd expect. A car dealership lot (file photo, illustration only — not the household or vehicle described in this article). Photo: Harrison Keely, Wikimedia Commons, original file page , CC BY 4.0 The two loans, at today's actual rates Freddie Mac's Primary Mortgage Market Survey put the national average 30-year fixed rate at 7.40% as of October 15, 2026. Assume our househol...
Read more

80/10/10 Piggyback Loan or Monthly PMI? The Real 5-Year Cost on a $400,000 Home With 10% Down

Image
On a $400,000 home with 10% down, splitting the financing into an 80/10/10 piggyback loan comes out about $29,000 ahead of taking one 90%-loan-to-value mortgage and paying monthly private mortgage insurance (PMI) over the first five years, in the worked example below. The gap keeps growing every year PMI stays on the single loan — which, by amortization alone, is more than eight and a half years. A new model home opens for buyers. Photo by Don Pinder, via Florida Keys–Public Libraries, Wikimedia Commons , licensed under CC BY 2.0 . Illustration of a new-home purchase; not the property discussed in this article. Key takeaways On a $400,000 home with 10% down, an 80/10/10 piggyback loan's combined payment is about $2,659/month versus about $2,916/month for one 90% loan with monthly PMI, in this worked example. Over 5 years, the piggyback path costs roughly $29,080 less once both the cash paid and the remaining loan balances are counted together. The PMI route doesn't jus...
Read more

Should You Assume a Seller's 3.25% FHA Loan, or Take a New One at 7.18%? The Real 5-Year Cost

Image
Correction November 11, 2026: The rate gap between the assumed 3.25% FHA loan and the 7.18% new-loan rate was given as roughly 394 basis points; 7.18% minus 3.25% is 393 basis points. The paragraph on approval risk also stated FHA’s underwriting thresholds as a flat “580 or higher” score and “43% or lower” debt-to-income; HUD publishes the manual-underwriting limits as a grid (31%/43% with no compensating factors, 37%/47% with one, 40%/50% with two), and that passage has been rewritten to match. Source: HUD FHA Resource Center, maximum qualifying ratios for manually underwritten loans, updated October 8, 2026. A seller's FHA mortgage rarely comes up in a home search, but it should. Assuming an existing FHA loan means taking over the seller's balance, rate, and repayment schedule instead of applying for a new loan. In the worked scenario below, a buyer who assumes a seller's $280,000 FHA loan at 3.25% comes out roughly $49,000 ahead over five years, and about $290 a month a...
Read more

15-Year Mortgage or 30-Year Plus Investing the Difference? The Real 10-Year Net-Worth Comparison on a $350,000 Loan

Image
Photo: Aerial view of Levittown, Pennsylvania, circa 1959. Public domain (published in the U.S. before 1978 without a copyright notice). Via Wikimedia Commons. Levittown was built on the FHA/VA-backed long-amortization mortgages that made the 30-year loan standard; illustrative, not a depiction of the homes discussed below. Key takeaways On a $350,000 loan at Freddie Mac's November 5, 2026 average rates (30-year: 7.18%, 15-year: 6.57%), the 15-year payment is $691 a month higher: $3,062 versus $2,371. After 10 years, the 15-year loan has paid off $193,749 of principal versus just $48,397 on the 30-year loan -- a $145,352 equity gap before counting any investment returns. Investing that $691 monthly difference instead would need to earn roughly 10.45% a year, every year, for 10 years just to match the 15-year loan's guaranteed equity position -- above most long-run historical stock market averages. Over a shorter 5-year window, the needed return rises to about 15% a year...
Read more

Balance Transfer or Personal Loan to Pay Off $8,000 in Credit Card Debt: Which Costs Less?

Image
To pay off $8,000 in credit card debt, a 0% balance transfer card usually costs less than a personal loan, but only if the monthly payment stays high. At a steady $376.21 a month, a transfer with a 4% fee and a 15-month promotional period costs about $542 in fees and interest, a 24-month personal loan at 11.90% costs about $1,029, and leaving the balance on a card charging 22.36% costs about $2,283. Drop the payment to $200 a month and the same transfer costs about $2,415, more than double the loan. Credit cards in a wallet (file photo). Photo: Chris Potter, Wikimedia Commons , CC BY 2.0 Key takeaways On an $8,000 balance paid at $376.21 a month, the 0% transfer (4% fee, 15 months) costs about $542, the 24-month loan at 11.90% about $1,029, and the original 22.36% card about $2,283. The transfer's advantage disappears once the payment slips to roughly $280 a month; at $200 a month it costs about $2,415 and takes 53 months. A loan origination fee changes the gap: a hypotheti...
Read more

What's a Good Homeowners Insurance Deductible? The Real 10-Year Cost of $500 vs. $2,500

Image
Correction November 8, 2026: the one-claim comparison had the $500 deductible subtracted twice. The net 10-year position for the $2,500 deductible with one claim is +$2,620 (not +$3,120), and the gap between the two deductibles in that case is $3,120 (not $3,620). The underlying inputs — $512 a year in premium savings, $5,120 over 10 years, and a $2,000 difference in out-of-pocket cost per claim — are unchanged, as is the two-claim result and the 2.56-claim break-even. Source: arithmetic on the Insurance.com rate data already cited below. Raising a homeowners insurance deductible from $500 to $2,500 saves the average U.S. household $512 a year, according to Insurance.com's state-by-state rate data updated October 28, 2026. Over 10 years that is $5,120 in premium savings — but the real question is whether that savings still wins once you factor in how often a typical homeowner actually files a claim, and the answer depends on a calculation most people never run. Illustration o...
Read more

Is a No-Closing-Cost Refinance Really Free? The Real Break-Even on a $350,000 Loan

Image
A no-closing-cost refinance is not free, and it is also not the worst of the options on the table. Those costs — title, appraisal, recording, underwriting — still exist and still get paid; what changes is whether they leave your bank account at closing, get buried in a higher rate, or get added to the loan balance. Those are three different prices for the same thing, and on a $350,000 refinance they do not rank the way the marketing or the usual advice suggests. What follows puts all three side by side at this week's rates, finds the month each one overtakes the others, and then shows how little of that answer depends on the rate level and how much depends on two numbers a lender will quote you in a single phone call. "A Little Mortgage," an 1899 etching — illustration only, not tied to any real transaction. Source: Library of Congress, Popular Graphic Arts collection, Wikimedia Commons , public domain Key takeaways Freddie Mac's survey put the 30-year fixed a...
Read more