This article is for educational purposes and is not mortgage, tax, or investment advice; all rate and loan-limit figures reflect 2026 data verified against Freddie Mac and the Federal Housing Finance Agency as of August 2026, and lender pricing varies daily.
TL;DR — Quick Verdict
- One discount point costs 1% of the loan amount. On a $400,000 loan that is $4,000, and it typically buys a rate reduction of roughly 0.25 percentage points.
- Freddie Mac’s Primary Mortgage Market Survey put the 30-year fixed-rate mortgage at 6.67% as of August 13, 2026 — down from 6.69% the prior week but above the 6.58% average of a year earlier.
- Buying one point on a $400,000 loan at 6.67% down to 6.42% cuts the monthly principal-and-interest payment by about $66 and produces a simple break-even of roughly 61 months.
- Comparison result: paying points beats putting the same cash toward principal only if you hold the loan past break-even. Below 61 months, the principal paydown wins on a $400,000 note.
- Rate reduction per point is set by each lender’s rate sheet, not by a national standard — Rocket Mortgage, Chase, and a local credit union can quote three different reductions for the identical point cost.
- Recommendation: buy points only when your realistic tenure in the loan exceeds break-even by a wide margin and you have cash left after closing costs and reserves.
A discount point is the only closing cost that is genuinely optional and genuinely permanent. Every other line item on a Loan Estimate — appraisal, title, recording — is a fee you pay to get the loan closed. A point is a bet. You hand the lender 1% of the loan amount at closing in exchange for a lower interest rate that lasts as long as you keep the mortgage, and the bet pays off only if you keep it long enough.
Freddie Mac reported the 30-year fixed-rate mortgage averaging 6.67% as of August 13, 2026, just under the 6.69% posted a week earlier — the highest reading since July 2025. At that level, a borrower financing the 2026 baseline conforming loan limit of $832,750 — set by the Federal Housing Finance Agency — would pay $8,328 for a single point. That is real money with a real payback period, and lenders including Rocket Mortgage and Chase quote wildly different rate reductions for the same 1% cost.
This article shows the arithmetic: what a point costs at several loan sizes, how to calculate break-even correctly rather than with the shortcut most calculators use, when a buydown loses to simply making a larger down payment, and the specific situations where paying points is the wrong move regardless of the math.
What One Point Costs at 2026 Loan Sizes
Pricing is mechanical. One discount point equals 1% of the loan amount — not the purchase price, not the appraised value. A borrower buying a $500,000 home with 20% down finances $400,000, so one point costs $4,000, not $5,000.
The rate reduction is where the mechanics stop. Industry convention holds that one point buys roughly 0.25 percentage points of rate reduction, but that convention is a rule of thumb, not a rule. Actual reductions on 2026 rate sheets range from about 0.125 to 0.375 percentage points per point depending on loan type, credit tier, loan-to-value, and how aggressively the lender wants volume that week. Borrowers comparing lender APR and total borrowing cost will see this variation immediately in the APR column.
Rates from Freddie Mac Primary Mortgage Market Survey, 08/13/2026 (Freddie Mac PMMS). Conforming limit from the Federal Housing Finance Agency (verify at fhfa.gov). Payment figures are original calculations using the standard 360-month fixed-payment formula; principal and interest only, excluding taxes, insurance, and mortgage insurance.
Notice the scaling problem. Point cost rises linearly with loan size, and so does monthly savings — which means break-even in months is nearly identical across every row. Loan size does not make points a better or worse deal. Only tenure does.
How Break-Even Is Actually Calculated (And Why the Common Method Is Wrong)
Most online calculators divide point cost by monthly payment savings and call the result break-even. On the $400,000 example: $4,000 ÷ $66 = 61 months, or roughly five years and one month. That number is useful as a first screen, but it hides two corrections that pull in opposite directions.
First, the payment-savings method ignores opportunity cost. The $4,000 spent on a point could have sat in a Treasury money market or a high-yield savings account earning something. At a 4% annual yield, that $4,000 generates about $163 in the first year — pushing break-even out to roughly 68 months once compounding is layered in on both the forgone cash and the recovered savings.
Second, working the other direction, a lower rate means each monthly payment allocates more to principal. After five years on the $400,000 loan, the 6.42% borrower owes roughly $1,070 less than the 6.67% borrower on identical payment histories. Equity accrues faster, and that advantage is real the moment you sell or pay off.
Net of both adjustments, break-even lands near 54 months — earlier than the simple calculation, because at this rate spread the faster principal paydown outweighs the opportunity cost drag. The two corrections roughly bracket the 61-month figure, which is why the shortcut survives as a first screen. That bracketing does not hold at every rate spread. When the rate gap is small (0.125 points per point), the equity effect shrinks, opportunity cost dominates, and true break-even stretches well beyond the simple calculation.
The tenure assumption is the fragile input, not the arithmetic. A buyer who says “I’ll be here 10 years” and moves in year four recovers about $3,160 in payment savings plus roughly $850 in faster principal paydown against a $4,000 outlay — a nominal wash, and a modest real loss once the opportunity cost of the cash is counted. The point did not blow up; it simply did nothing, having tied up $4,000 for four years to no purpose. Anyone weighing this alongside 15-year versus 30-year total interest should model both decisions together, since a shorter term compresses the window in which a buydown can pay for itself.
Points vs. Larger Down Payment: Which Wins on a $400,000 Loan?
Suppose you have $8,000 of discretionary cash beyond your planned down payment and reserves. Two uses compete: buy two points to drop the rate from 6.67% to 6.17%, or put the $8,000 toward principal and finance $392,000 at 6.67%.
Original amortization calculations by the author. Base rate of 6.67% from Freddie Mac Primary Mortgage Market Survey, 08/13/2026 (Freddie Mac PMMS). Rate reduction of 0.25 percentage points per point assumed; actual lender pricing varies. Balances rounded to nearest $100.
The buydown wins on monthly cash flow by $80. The principal paydown wins on equity through year five by about $5,300, and still leads by $2,500 at year ten. Cumulative payment savings from the buydown ($4,800 by year five, $9,600 by year ten) close that gap in month 65 — just under five and a half years — after which the buydown pulls ahead permanently.
Verdict
For a borrower confident of holding the loan past five and a half years without refinancing, the two-point buydown is the stronger position — lower payment, lower lifetime interest, and cumulative savings that eventually overtake the equity advantage. For anyone with a realistic horizon under that, the principal paydown is safer because it cannot be stranded. The equity is yours the day you close; a bought-down rate evaporates the moment you refinance or sell. The refinance risk is the variable that has changed most since spring: with the Federal Reserve holding rates steady over three dissents that favored a hike, and mortgage rate forecast data no longer pointing clearly downward, the near-term refinance window that made buydowns look reckless in February is harder to count on.
What Determines Your Rate Reduction Per Point
Lender rate sheets are built as a grid. Each cell prices a rate against a cost expressed in points, and the spacing between cells is what determines how much rate you get per point paid. That spacing is not constant.
Credit tier compresses or widens the grid. A borrower at 780 FICO with 20% down sits in a pricing band where an eighth-point of rate might cost only 0.4 points; a 660-FICO borrower at 95% loan-to-value faces loan-level price adjustments that already consume much of the pricing range, so buying down costs more per basis point. The relationship between credit score and mortgage rate tiers shapes the buydown math before the point discussion even starts.
Loan product matters just as much. Government-backed loans price differently from conventional — a comparison of FHA versus conventional rate and total cost shows that FHA borrowers often see flatter buydown curves because mortgage insurance premiums, not the note rate, drive their total cost. VA loan rates compared to conventional follow yet another pricing logic, and VA specifically caps what borrowers may pay in certain fee categories.
Loan size crosses a pricing boundary at the conforming limit. A $850,000 loan in a baseline county exceeds the FHFA 2026 limit of $832,750 and is priced as a jumbo, where jumbo versus conforming rate differences can either help or hurt buydown economics depending on the lender’s portfolio appetite that quarter.
Timing is the last variable, and it has been unusually punishing this summer. Rate sheets reprice intraday when the 10-year Treasury moves, and the 10-year closed near 4.68% on August 14, within striking distance of the 4.75% level tested earlier that week — its highest in roughly nineteen months. The relationship between mortgage rates and the 10-year Treasury means a point quoted Monday morning may buy a different reduction Thursday afternoon. Lock timing interacts directly — rate lock duration and extension costs can quietly consume the savings a point was supposed to deliver.
What Most Borrowers Get Wrong About Points
Five errors show up repeatedly, and each one has a specific dollar cost.
Confusing discount points with origination fees
An origination charge compensates the lender for processing the loan and buys you nothing. A discount point buys rate. Both appear in Section A of the Loan Estimate and both are quoted as a percentage of loan amount, which makes them easy to conflate. Consequence: a borrower believes they bought a rate reduction and instead paid a fee. Correct action: read the Loan Estimate line labels literally, and compare the full origination fee impact on true mortgage cost across every quote before assuming the low-rate offer is cheaper.
Buying points with money needed for reserves
Closing costs typically run 2% to 5% of the loan amount before any points are added. A borrower who spends their last liquid $8,000 on a buydown and then faces a $6,000 HVAC replacement in month seven has converted an emergency fund into an illiquid rate reduction. Correct action: fund three to six months of full housing payment in cash before considering any point purchase.
Assuming the rate reduction is standardized
Nothing requires a lender to give 0.25 percentage points per point. Consequence: borrowers accept 0.125 per point without shopping, cutting monthly savings on a $400,000 loan from $66 to about $33 and stretching break-even from roughly 61 months to more than 120. Correct action: request each lender’s par rate — the rate at zero points — alongside their buydown grid, then compare reductions per dollar spent rather than comparing headline rates.
Paying points on a loan you plan to refinance
Points are consumed at closing and are not refundable or portable. A borrower who buys down in a high-rate environment and refinances 18 months later has forfeited the unrecovered balance. The judgment call here is harder than it was earlier in 2026: rates have climbed most of a percentage point off their February low and the Fed’s own committee is split toward tightening, so “wait and refinance” is a thinner plan than it looked in the spring. Correct action: if the only reason you want a lower rate is that current rates feel high, that reasoning still argues against locking in a sunk cost — but weigh it against the possibility that no refinance window arrives. Consider whether fixed versus adjustable rate cost comparison addresses the same problem more cheaply.
Ignoring the tax treatment question entirely
Points paid on a purchase mortgage for a primary residence may be deductible in the year paid under certain conditions, while points on a refinance generally must be amortized over the loan term. The rules turn on specifics — whether the points were customary in your area, whether you itemize, and how the funds were paid at closing. Correct action: read IRS Publication 936, “Home Mortgage Interest Deduction,” for the current-year tests and confirm with a tax professional before treating any deduction as certain. Do not build a break-even model that assumes a deduction you may not qualify for.
Who Should Buy Points — and Who Should Not
The decision reduces to four conditions that must all hold.
Condition one: your realistic tenure clears break-even with margin. If break-even is 61 months, a 7-year horizon is thin and a 12-year horizon is comfortable. Median homeowner tenure has historically run well under a decade, so treat your own estimate skeptically.
Condition two: you have a view on whether a refinance is coming, and you are honest about it. Freddie Mac’s August 2026 reading of 6.67% is nine basis points above the 6.58% average of a year earlier and 69 basis points above the 5.98% low set on February 26, 2026 — a swing worth about $180 a month on a $400,000 loan. The February borrower who skipped points made the right call. Whether today’s borrower should is a different question, because the direction of travel has changed. The Federal Open Market Committee held its target range at 3.50%–3.75% on July 29 by a 9–3 vote, with all three dissenters preferring a quarter-point increase, and the Fed under Chair Kevin Warsh has deliberately pared back forward guidance rather than signaling a path. Futures markets have priced roughly a one-in-three chance of a hike at the September meeting. A buydown is dangerous when a refinance is likely and merely expensive when it is not; the case for points is stronger now than at any point this year, which is precisely when the other three conditions deserve the most scrutiny.
Condition three: the cash is genuinely surplus. Points come after down payment, after closing costs, after reserves, after moving expenses.
Condition four: your lender’s grid actually offers a competitive reduction. If the best quote is 0.125 percentage points per point, the answer is usually no regardless of the other three conditions.
Two profiles should almost never buy points. Borrowers early in a career with likely relocation face tenure risk that dominates every other variable. And investors financing rental property through DSCR investor loan programs generally do better deploying capital into acquisition or reserves than into a rate reduction on a property they may sell or refinance on a business timetable.
Retirees financing a final residence occupy the strongest position. A 15-year expected tenure, no refinance intent, and cash from a prior home sale make the buydown arithmetic work cleanly — provided the reserve condition holds. Where the cash comes from matters too: some lender types quote materially better grids than others, and online lender versus bank versus credit union pricing is worth checking before committing five figures to a single lender’s rate sheet.
Frequently Asked Questions
Can I buy a fraction of a point?
Yes. Lenders routinely price in eighths and quarters, so a borrower can buy 0.5 points for 0.5% of the loan amount. On a $400,000 loan that is $2,000, typically buying about 0.125 percentage points of rate reduction — roughly $33 a month at current pricing. Fractional purchases are common when a borrower has a specific payment target rather than a specific cash budget, and the break-even math scales proportionally.
Can the seller pay my discount points?
Yes, through a seller concession, subject to caps set by the loan program and loan-to-value ratio. Fannie Mae and Freddie Mac limit interested-party contributions on conventional loans, with the allowable percentage varying by occupancy and down payment. Seller-paid points are usually a better deal for the buyer than an equivalent price reduction, because the payment savings persist for the full loan term while a price cut only reduces the financed principal modestly.
Do points lower my APR?
Points raise the APR relative to the note rate because APR incorporates prepaid finance charges. A 6.42% note rate with one point on a $400,000 loan produces an APR above 6.42% — often near or above the 6.67% zero-point alternative in the early years. This is precisely why APR is a poor tool for comparing point-heavy quotes against zero-point quotes when your holding period is short.
Are temporary buydowns like a 2-1 the same thing?
No. A 2-1 buydown funds an escrow account that subsidizes the payment for the first two years, then the rate reverts to the note rate permanently. Discount points reduce the note rate for the full 360 months. Temporary buydowns are often seller-funded builder incentives; they help with early cash flow but produce no lifetime interest savings, and any unused escrow is typically credited if you refinance early.
How We Researched This Article
Rate data comes from Freddie Mac’s Primary Mortgage Market Survey, the weekly national benchmark released Thursdays at noon Eastern. We used the survey week ending August 13, 2026, which reported the 30-year fixed-rate mortgage at 6.67% and the 15-year fixed-rate mortgage at 5.96%. Freddie Mac collects PMMS data from loan applications submitted through Loan Product Advisor, covering conventional, conforming, fully amortizing purchase loans for borrowers with 20% down and excellent credit. Full methodology and current readings are available from Freddie Mac’s PMMS page.
Loan limit figures come from the Federal Housing Finance Agency’s 2026 conforming loan limit announcement, which set the baseline one-unit limit at $832,750 and the high-cost ceiling at $1,249,125. Monetary policy references are to the Federal Open Market Committee statement of July 29, 2026, published by the Board of Governors of the Federal Reserve System. Tax treatment guidance references IRS Publication 936, available from the Internal Revenue Service; we describe the framework rather than asserting deductibility, because eligibility depends on individual filing circumstances we cannot evaluate.
All payment and amortization figures in this article are modeled, not measured. We calculated them using the standard fixed-payment formula over a 360-month term, principal and interest only, excluding property taxes, homeowners insurance, mortgage insurance, and HOA dues. Balances are rounded to the nearest $100 and payments to the nearest dollar. The rate reduction of 0.25 percentage points per point used throughout is an industry convention applied for illustration; it is not a verified national average.
One important limitation: Freddie Mac discontinued reporting average fees and points when it revised PMMS methodology, because Loan Product Advisor does not capture that data. No current national average for points paid per loan is published by a primary source. Figure unavailable at publication — Freddie Mac did not return points-specific data for this period. Readers should treat any published “average points paid” figure with caution and instead compare their own lender quotes directly. Lender pricing grids are proprietary and change intraday; the ranges cited reflect commonly observed conventional pricing rather than a surveyed distribution.
Research last conducted August 2026. All figures were verified against named primary sources before publication.