DSCR Investor Loan Rates and Requirements 2026: How Much They Really Cost

What DSCR Investor Loan Rates Actually Are in September 2026

Four active lenders published rate data across August and September 2026, and they disagree — usefully. Griffin Funding quoted fixed DSCR loan rates starting at 6.375% for 30-year fixed, 40-year fixed, and 5-year ARM products, with 1-year adjustable products starting at 5.375%, in a rate sheet dated September 1. HomeAbroad set its baseline par rate at 6.125% for domestic investors at 740 FICO and 70% LTV with zero points, last updated August 5. Defy Mortgage, which publishes a full FICO-by-LTV pricing grid rather than a starting-at figure, quoted a band of 6.375% to 7.875%. Investment Property Loan Exchange, which brokers across a wider lender network including weaker credit profiles, reported a 6.5% to 8.75% band for residential investment property.

Those ranges are not in conflict. They describe different slices of the same distribution. Griffin and HomeAbroad publish what a strong file gets; Defy publishes the whole grid; the broker network publishes what everybody gets. The useful takeaway is the floor and the ceiling: roughly 6.1% at the top of the credit box, roughly 8.75% at the bottom.

Note the foreign national figure. At 7.00% par against a 6.125% domestic par, non-resident investors pay roughly 88 basis points more for identical collateral — a pure documentation-risk premium, not a property-risk one.

Note also what happened at the top of the table since summer. In July, the conventional benchmark sat at 6.55% and the sharpest DSCR par rate sat just below it. The benchmark has since climbed in three straight weekly steps — 6.66% on August 27, 6.71% on September 3, and 6.76% on September 10 — while HomeAbroad has held its DSCR par at 6.125% since the spring. That means the strongest DSCR pricing is now a full 64 basis points below the owner-occupied conventional survey rate, up from 55 basis points a month earlier — before any investor-occupancy adjustment is layered onto the conventional side. The premium investors are told to expect has not merely narrowed. At the top of the credit box, it keeps inverting further.

Why the Spread Exists: Treasury Yields, Not the Fed

DSCR loans are not priced off the federal funds rate. They are priced off long-term Treasury yields, because that is where the securitization bid comes from. The 10-year Treasury note has moved sharply since mid-August. It closed at 4.63% on August 13, per the Federal Reserve’s H.15 constant maturity series, then climbed nearly in a straight line: 4.79% by September 2, and 4.95% by September 10 — a fresh high for the cycle and roughly 32 basis points above the reading investors were quoting just four weeks earlier. Lenders layer a spread on top of that yield to cover credit risk, servicing, prepayment uncertainty, and margin.

Work the arithmetic forward. A 4.95% 10-year plus the 200 to 225 basis point spread that active DSCR desks quote for a standard 30-year fixed at 75%–80% LTV produces a baseline around 6.95% to 7.20% before any borrower-specific adjustment. That gap between the mechanical baseline and where lenders are actually pricing has widened further: HomeAbroad’s par rate of 6.125% now prices 83 to 108 basis points below that baseline, meaning the lender is compressing margin more aggressively than at any point this cycle to hold volume in a competitive non-QM origination market. Investors who understand why mortgage rates track the 10-year Treasury stop timing their locks around Fed meetings and start watching auction results instead.

The last month made that point better than any explanation could. The Federal Open Market Committee met on July 29 and held the target range at 3.50% to 3.75%. Long yields rose anyway, and kept climbing through August and into September — from 4.63% to 4.95% in under a month, with no rate action from the Fed in between. Nothing the Fed did moved DSCR pricing during that stretch; the term premium on the long end did all the work. The Committee’s next meeting falls September 15–16, and positioning ahead of it is genuinely split this time: futures markets moved from pricing meaningful odds of a cut in late July to assigning real odds to a hike by early September, after Fed Chair Kevin Warsh’s Jackson Hole remarks were read by markets as putting tightening back on the table rather than closing the door on it. An investor who spent the summer watching FOMC statements and ignoring the 10-year auction calendar had the wrong screen open — and the screen matters even more heading into a meeting where the outcome is no longer a foregone conclusion.

Compare that spread to the conventional side. The Freddie Mac survey rate of 6.76% describes an owner-occupied loan with 20% down and excellent credit. An investment property on that same conventional track gets hit with Fannie Mae loan-level price adjustments — cumulative, risk-based fees assessed on occupancy, credit score, and LTV, which lenders typically convert into rate. Investor-occupancy adjustments commonly add 0.50% to 1.50% to the delivered rate, though several active DSCR desks now quote the practical range as narrower, closer to 0.50% to 0.75%. So the honest comparison is not 6.125% DSCR versus 6.76% conventional. It is 6.125% DSCR versus a conventional investor rate that has already been marked up off that 6.76% base.

Rate movement inside a single month matters more than investors assume, and the plateau that showed up in mid-August didn’t hold. The Freddie Mac 30-year fixed-rate mortgage sat at 6.66% on July 30 and again on August 27, then moved higher in back-to-back weekly prints: 6.71% on September 3 and 6.76% on September 10 — a 10-basis-point move in two weeks after roughly a month of near-flat readings. A plateau at a higher level still resets every quote written against it, and a resumed climb resets it again; DSCR pricing tracks that level with a lag, which makes rate lock timing and extension costs a real line item rather than paperwork.

One reference point worth updating if you have been anchored on it since spring: the conventional benchmark remains above where it stood a year ago, and the gap has widened, not narrowed. At 6.76% on September 10, the 30-year fixed sits 41 basis points above the 6.35% it averaged in the same week of 2025 — up from a 9-basis-point gap just four weeks earlier. For most of the first half of 2026 the year-over-year comparison ran the other way, and a good deal of investor commentary still assumes it does. It does not, and the gap is now the widest it has been all year.

Requirements Lenders Actually Enforce

Every DSCR lender advertises “no tax returns.” Almost none advertise the reserve requirement, which is where marginal files die. Total Quality Lending’s published program matrix requires two months of PITIA in reserves on standard loans, six months on loans above $1.5 million, and twelve months on loans above $2.5 million. Other lenders run tighter — 1st Nationwide Mortgage describes three to six months of PITIA per financed property, scaling upward as the borrower’s portfolio grows.

Credit floors cluster in a narrow band across lenders, but what those floors cost has become the more useful number. Expect a minimum FICO between 620 and 680 depending on program, with 740-plus required for the headline pricing. New Silver puts the common threshold at 660, rising to 700 or above for an 80% LTV; Defy Mortgage writes down to 640 and Griffin Funding to 620. Defy’s published grid puts hard numbers on the cost of a weak score: the spread between its 640 and 680 cells exceeds 200 basis points in some scenarios. That tiering mirrors conventional lending closely enough that the credit score impact on mortgage rates by tier translates almost directly, only with wider steps.

Coverage-ratio minimums have loosened materially over the past year, and this is the single biggest change to the qualification picture in 2026. The old rule of thumb — 1.00 or no deal — no longer describes the market. Defy sets its floor at 0.75. Griffin Funding has removed the minimum entirely, funding sub-1.0 files against reserves and running a no-ratio program that does not use cash flow to qualify at all — in July 2026 alone, Griffin closed DSCR loans with coverage ratios ranging from 0.70 to 2.23, including two below 1.0. That does not make weak coverage cheap. It makes it writeable.

One structural detail catches investors off guard: the appraisal does double duty. On a conventional loan it confirms value. On a DSCR loan it also establishes fair market rent, typically via a Form 1007 rent schedule — and that appraiser-determined rent, not your signed lease, is often what the underwriter plugs into the coverage calculation. A lease above market rent does not always help you.

The Prepayment Penalty Trade: What You’re Actually Selling

Conventional mortgages carry no prepayment penalty. DSCR loans almost always do, and this is the single most misunderstood pricing lever in the product. Because DSCR loans fall outside consumer mortgage protections that apply to owner-occupied lending, lenders can and do impose penalties on early payoff — Total Quality Lending’s guidelines allow prepayment periods up to five years.

Here is the trade. Accepting a longer prepayment penalty term buys you a lower rate. Griffin Funding explicitly lists prepayment penalty term, from zero to five years, as one of the five variables that set its pricing, alongside credit score, coverage ratio, down payment, and buydown points — and quantifies the trade at 0.25% to 0.75% of rate between a five-year penalty and no penalty at all. Defy Mortgage bakes a five-year step-down structure into its published grid outright, which means the 6.375% headline is a five-year-penalty rate, not a clean one. An investor planning a fifteen-year hold gives up almost nothing by accepting that. An investor planning to refinance in eighteen months gives up a great deal.

State law overrides lender preference in several jurisdictions. Prepayment penalties are not permitted in Alaska, Kansas, Michigan, Minnesota, New Mexico, or Rhode Island. They are barred on loans vested to individuals — rather than to an entity — in Illinois and New Jersey. Pennsylvania prohibits them below a statutory base figure that its Department of Banking adjusts annually for inflation; the last figure we verified was $319,777. Ohio caps penalties on one-to-two unit properties at 1% of the loan balance during the first five years.

Investors in the six no-penalty states should expect slightly higher quoted rates as a result, because the lender cannot recover margin through the penalty. That is not a broker taking advantage of you; it is the pricing model working as designed. The same logic governs mortgage points and rate buydown math — you are always trading one form of cost for another, and the right answer depends on your holding period.

DSCR vs Conventional Investor Loan: Which Is Better for a Rental Purchase?

Take a concrete deal. A $400,000 single-family rental, 25% down, $300,000 loan amount, 30-year fixed, borrower with a 740 FICO. Rent is $2,900 per month. Taxes, insurance, and association dues run $700 per month.

On the DSCR track at the HomeAbroad par rate of 6.125%, principal and interest come to roughly $1,822 per month. Add the $700 in escrow and PITIA is about $2,522. Coverage ratio: $2,900 ÷ $2,522 = 1.15. That clears the 1.00 floor but sits below the 1.25 threshold that unlocks best pricing, so a real quote would likely land slightly above par — call it 6.375%, the published floor at both Griffin and Defy, or about $1,872 in principal and interest.

On the conventional track, the Freddie Mac survey rate of 6.76% is the starting point, not the ending point. Layer investor-occupancy and LTV loan-level price adjustments, and a 0.50% to 1.50% rate add is typical. At the midpoint — roughly 1.00% — the conventional investor rate lands near 7.76%, producing about $2,152 in principal and interest. That is roughly $280 more per month than the DSCR quote, up from about $261 a month earlier, as the conventional benchmark has climbed while DSCR pricing held flat.

The conventional loan does bring advantages: no prepayment penalty, generally lower origination fees, and a rate that is not marked up for documentation flexibility you may not need. But it also requires two years of tax returns, counts the new mortgage against your personal debt-to-income ratio, and caps you at ten financed properties. If write-offs and depreciation have suppressed your reported income — the standard situation for any investor doing tax planning correctly — the conventional loan may not be available at any rate. Comparing offers properly means running APR, rate, and total borrowing cost side by side rather than fixating on the note rate.

What Most Investors Get Wrong

Four errors show up repeatedly, and each one is expensive.

Mistake 1: Shopping the rate before structuring the deal

Investors call four lenders and compare quotes on identical assumptions, not realizing that a 5% larger down payment or a different prepayment term changes the quote materially. The consequence is comparing a 75% LTV quote from one lender against an 80% LTV quote from another and concluding the first lender is cheaper. The correct action: fix your LTV, coverage ratio, and prepayment term first, then request quotes on that identical structure. The controllable factors for securing the lowest mortgage rate apply here with more force than on conventional loans, because non-QM pricing grids have wider adjustment ranges.

Mistake 2: Treating a 1.00 coverage ratio as adequate

A coverage ratio of exactly 1.00 means rent covers PITIA with nothing left. It does not cover vacancy, maintenance, capital expenditure, or management. The consequence is a property that qualifies on paper and bleeds cash in practice, since one month of vacancy per year alone consumes roughly 8% of gross rent. This mistake has become easier to make, not harder: with lenders now writing at 0.75 coverage and below, the underwriting floor has stopped functioning as a sanity check. The correct action: underwrite your own deal to 1.20 or better before you care what the lender’s minimum is.

Mistake 3: Accepting the default prepayment term without asking

Lenders quote a default structure, commonly a stepped-down five-year penalty. Accepting it on a property you intend to refinance in year two can cost several percent of the loan balance at payoff. The correct action: state your exit timeline before the quote, and ask what the rate becomes at each prepayment term. Matching the penalty period to the actual hold plan costs nothing but the conversation.

Mistake 4: Ignoring origination fees while optimizing the rate

Non-QM origination fees run higher than conventional, and a 6.25% rate with three points can cost more over a five-year hold than a 6.75% rate with one point. The consequence is chasing a headline rate into a worse total cost. The correct action: calculate total cost over your actual holding period, since origination fee impact on true mortgage cost compounds differently than most investors assume.

Fixed or Adjustable: Running the Break-Even

Griffin Funding’s adjustable DSCR rates start at 5.375% in September 2026, against a fixed floor of 6.375% — a full percentage point of initial savings, unchanged from August. On a $300,000 loan, that difference is $192 per month in principal and interest, or $2,300 per year.

Whether that trade pays depends entirely on holding period and refinance intent. An investor executing a BRRRR strategy — buying, renovating, renting, refinancing within eighteen to thirty months — captures the full adjustable-rate discount and exits before the first adjustment. An investor buying a stabilized rental for a twenty-year hold is taking meaningful rate risk to save $2,300 annually in the early years.

Market context has only gotten harder to read since midsummer, and it has moved again since. The Federal Open Market Committee held at 3.50% to 3.75% on July 29, 2026, but did so on a 9–3 vote, with Presidents Hammack, Kashkari, and Logan all dissenting in favor of a quarter-point hike — the first time since 2016 that three policymakers dissented in the same direction. Under Chair Kevin Warsh the Committee has also stopped publishing forward guidance, which removes the signal investors previously used to position. Through early August, weak employment data and a July CPI print of 0.1% monthly and 3.4% annually, with core at 2.5%, had pulled September hike odds down to roughly one in three. That reversed at the Jackson Hole symposium in late August: markets read Warsh’s remarks there as putting a rate increase back on the table for the September 15–16 meeting, and futures pricing swung toward meaningful odds of a hike rather than a hold. The August CPI print, due September 11 — days before the decision — is the last major data point the Committee sees beforehand; its result could not be confirmed against a primary BLS source at the time of writing, so it is not included here. What is confirmed is the price action: the 10-year Treasury climbed roughly 32 basis points from mid-August to September 10, and lender commentary now frames the September decision explicitly as a live risk to lock around, not a formality.

Read that honestly rather than optimistically. The direction of the next move is up or nowhere, not down, and the Committee contains a bloc that has already voted to tighten. That skew argued against floating-rate exposure for long-horizon holders even when hike odds cooled through early August — and it argues more strongly now that odds have swung back toward a live possibility of a hike at the September meeting. The asymmetry hasn’t just persisted; for adjustable-rate borrowers without a short, defined exit, it has gotten louder, not quieter. The ARM versus fixed break-even analysis framework transfers cleanly to DSCR products, and the fixed versus adjustable rate cost comparison is worth running with your actual numbers rather than defaulting to fixed out of habit.

Is a DSCR Loan Worth It for You?

Conditional logic, not a blanket recommendation.

A DSCR loan is worth the premium if: your tax returns show materially less income than you actually earn; you already hold four or more financed properties and are approaching conventional limits; you vest title in an LLC and want to keep it that way; you need to close in under thirty days, since DSCR files close in as few as six days at active lenders with a 34-day average; or your personal debt-to-income ratio disqualifies you from conventional financing regardless of the property’s performance.

A DSCR loan is probably the wrong tool if: you are buying your first rental with strong W-2 income and clean returns; the property’s coverage ratio sits below 1.00 and you are stretching to make the deal work; you plan to sell or refinance within twelve months and cannot negotiate away the prepayment penalty; or the loan amount falls below the $75,000 to $100,000 minimum most lenders enforce.

Scale changes the calculus more than anything else. DSCR loans do not count against personal debt-to-income, which is the mechanical reason portfolio investors use them. An investor with eight rentals cannot get a ninth conventional loan easily no matter how strong the deals are. The same investor can keep adding DSCR loans as long as each property covers itself and reserves hold up. Investors comparing against other high-balance options should also check how jumbo and conforming loan rates differ, since DSCR programs extending to $4.5 million overlap directly with jumbo territory.

How We Researched This Article

Rate figures in this article come from three categories of source, weighted by reliability. The conventional mortgage benchmark comes from the Freddie Mac Primary Mortgage Market Survey, which collects rates from thousands of loan applications submitted through Loan Product Advisor and publishes weekly on Thursdays. We pulled the three most recent weekly readings — 6.66% on August 27, 6.71% on September 3, and 6.76% on September 10, 2026 — to establish trend rather than relying on a single print. The year-ago comparison figure of 6.35% comes from the September 10 release.

Treasury yield data comes from the U.S. Department of the Treasury daily par yield curve series, cross-checked against the Federal Reserve’s H.15 release as published by FRED at the Federal Reserve Bank of St. Louis. Constant maturity Treasury values are interpolated from bid-side quotations collected near 3:30 p.m. each trading day, which means the 10-year figure cited here is a close-of-day reading, not an intraday level. The 4.95% figure is the most recent close confirmed in the H.15 release at the time of writing, reached September 10, 2026, up from 4.63% on August 13 — a roughly 32 basis point move in under a month.

Monetary policy detail — the target range, the July 29 9–3 vote, and the identity and direction of the three dissents — comes from the FOMC statement issued that day and the accompanying press conference transcript, both published at federalreserve.gov. Fed Chair Kevin Warsh’s remarks at the Jackson Hole symposium in late August, and subsequent market pricing ahead of the September 15–16 FOMC meeting, are drawn from contemporaneous financial press coverage and futures-market pricing; hike-odds figures cited here are directional sentiment from those markets, not certainties, and can move quickly. Inflation figures come from the Bureau of Labor Statistics Consumer Price Index release for July 2026, published August 12. The August 2026 CPI release, scheduled for September 11, 2026, could not be confirmed against a primary BLS source at the time of writing; the July figures above are retained and this gap is noted rather than estimating a result. Rate-path probabilities are market-implied from federal funds futures and move daily; they are reported here as approximate and were current in mid-September 2026.

DSCR pricing has no federal survey equivalent, and this is the central limitation of any DSCR rate analysis including this one. No agency collects and publishes non-QM origination rates. We therefore drew from published rate pages and program guideline documents of active originators — Griffin Funding, HomeAbroad, Defy Mortgage, Investment Property Loan Exchange, and Total Quality Lending — and reported the range across them rather than computing a false average. These are trade sources with a commercial interest in the rates they publish, so they are used to establish observed ranges and program mechanics, not as authoritative market averages. Where lender figures conflicted, we reported both endpoints and explained the difference in borrower profile that produces it.

Three source limitations are worth stating plainly. Griffin Funding’s product page, updated September 1, 2026, republishes the same starting rates as its August version — 5.375% ARM floor, 6.375% fixed floor — alongside updated year-to-date origination figures (508 DSCR loans for $145 million through July 31, 2026); it still does not publish an upper bound for its fixed range, and an older Griffin comparison page shows a 6.125%–7.50% band that its current rate page does not support, so we used the current page. Defy Mortgage’s published headline band of 6.375%–7.875% is internally inconsistent with cells in its own grid, which show 6.125% at 740+ FICO and 75% LTV and 8.375% at 680 FICO and 80% LTV; we used the headline band and the 640/75% cell as published, and could not confirm a fresher figure this cycle. Investment Property Loan Exchange’s 6.5%–8.75% residential band is a full-year 2026 figure last updated in February 2026, not a monthly print, and should be read as a wide annual range rather than current pricing. Separately, Griffin Funding published July 2026 origination volume but not an average borrower FICO or loan-to-value for the month; that figure could not be sourced for July 2026, and we cite the lender’s disclosed 2025 average of 739 with its year labeled rather than carrying forward the prior period’s number.

Conventional investor pricing mechanics come from the Fannie Mae Loan-Level Price Adjustment Matrix, effective for whole loans purchased on or after May 1, 2023 under the framework directed by the Federal Housing Finance Agency.

All payment figures in the comparison sections are modeled, not measured. They use standard 30-year amortization on the stated loan amounts and rates, exclude escrow unless explicitly noted, and assume no points. Actual quotes will differ based on lender margin, loan size, property type, reserves, and prepayment structure. The roughly $280 monthly difference in the DSCR versus conventional comparison depends on the loan-level price adjustment midpoint assumption of 1.00% applied to the September 10 Freddie Mac benchmark; at the 0.50% low end the gap is about $177, and at the 1.50% high end it widens to about $384. State prepayment restrictions are drawn from Total Quality Lending’s published program overlays and reflect that lender’s compliance posture — verify against current law in your state, since overlays vary between originators. The Pennsylvania threshold in particular is a statutory base figure adjusted annually by the Department of Banking and published in the Pennsylvania Bulletin; the 2026 adjustment could not be confirmed from a primary source, so the last verified figure is retained and the gap noted here.

Research last conducted September 13, 2026. All figures were verified against named primary sources before publication, except where explicitly noted above as unconfirmed for the current period.