The BRRRR Exit: Underwriting the DSCR Refinance Before You Buy the Rehab

The BRRRR model is usually taught as four independent steps and one hopeful assumption. Buy, rehab, rent, refinance, repeat — with the refinance treated as a formality that happens once the property is stabilized. In practice the refinance is the only step with hard third-party constraints, and it is the step investors model last. That inversion is why deals that looked like they returned 90% of capital return 55%.

The takeout is a DSCR loan, and a DSCR loan has exactly two ceilings: a loan-to-value cap and a debt service coverage floor. Whichever binds first determines the check. Both can be estimated within a few thousand dollars before the purchase contract is signed, which means the correct time to underwrite the refinance is during due diligence on the acquisition — not during month five, when the money is already spent.

The two ceilings, and which one usually binds

Cash-out DSCR refinances on a single-family rental price at 70% to 75% LTV at most lenders, with 75% requiring a 700-plus score and a clean 12-month rental history. A few programs stretch to 80% on rate-and-term, but cash-out is where the industry draws its line, and 75% is the realistic planning number.

The DSCR floor is separate. Most programs require 1.00 minimum with pricing tiers that improve at 1.10, 1.20 and 1.25. On a cash-out refinance the same 1.20 threshold that gets par pricing on a purchase becomes the practical constraint on loan size, because every dollar of additional proceeds raises the payment and lowers the ratio.

Work an actual case. A $95,000 purchase in a $140,000-ARV neighborhood, $38,000 of rehab, $8,000 of holding and closing costs. Total in: $141,000. The appraisal lands at $142,000 — a good outcome. At 75% LTV the maximum loan is $106,500.

Now check coverage. At 7.375% on a 30-year fixed, $106,500 amortizes to roughly $735 a month. Taxes at a 2.1% effective rate add $248. Landlord insurance at $1,300 annually adds $108. PITIA is about $1,091. The property rents at $1,325. DSCR is 1.21 — it clears, barely.

Change one input. If the effective tax rate is 2.6% instead of 2.1%, taxes are $308 and PITIA is $1,151, putting DSCR at 1.15. The loan is now constrained by coverage, not LTV, and the lender sizes down to roughly $101,000 to restore 1.20. That is $5,500 of capital that stays in the deal because of a county tax rate the investor could have looked up in ten minutes. The DSCR ratio and BRRRR calculators exist specifically to run this before the offer, and the underlying arithmetic — what goes in PITIA, what rent figure the lender is allowed to use — is spelled out in how DSCR loans work.

The appraisal is the whole model

Everything downstream of the ARV estimate is arithmetic. The ARV estimate itself is a guess, and it is the guess that determines whether the deal works.

Appraisers on a refinance use closed sales within roughly a mile and typically the last six months, adjusted for size, condition and bed/bath count. They do not credit rehab spend. A $38,000 renovation that took the property from C5 to C3 condition is worth whatever the comparable C3 sales say it is worth, which in a neighborhood with a hard ceiling might be $18,000. Investors who over-improve relative to the block are converting cash into a condition rating no comp supports.

Three habits prevent the miss. Pull three to five closed comps in the actual subdivision before the offer, not neighborhood-wide averages from an AVM. Scope the rehab to the finish level of those specific comps rather than to a personal standard. And expect the appraisal to come in at or slightly below the honest comp range, because appraisers on cash-out refinances are institutionally conservative — they know the loan is being sized off their number.

The same appraisal also carries the 1007 rent schedule, and a low market rent opinion is as damaging as a low value. Most programs use the lesser of in-place rent and appraised market rent. A property leased at $1,325 that comes back with a $1,225 market rent opinion loses about 0.09 of DSCR on a figure the borrower never negotiated. The mechanics of rebutting that, and what evidence actually moves an appraiser, are covered in the 1007 rent schedule guide.

Seasoning: the constraint that decides the pace

Cash-out DSCR refinances generally require three to six months of title seasoning, measured from the deed date, and the appraisal must be based on current appraised value rather than purchase price. Six months is the conservative planning assumption; a meaningful set of lenders will do 90 days, and a smaller set will do zero-seasoning cash-out at reduced LTV — typically 65% to 70% — which is worth precisely the difference between the proceeds forgone and the hard money interest saved.

Delayed financing is the separate mechanism, and it is frequently confused with cash-out. On a property purchased with cash, some programs allow a rate-and-term refinance up to 75% of the lesser of appraised value or total documented acquisition-plus-rehab cost, with no seasoning requirement, provided the investor can produce the settlement statement and paid rehab invoices. That path recovers documented basis, not created equity. On the $141,000-in, $142,000-ARV deal above the two produce nearly identical proceeds; on a $110,000-in, $150,000-ARV deal the cash-out path is worth roughly $30,000 more and is worth waiting six months for.

Reserve requirements compound this. Most programs want three to six months of PITIA in verified reserves at closing, and reserves are checked after the cash-out proceeds are excluded from the calculation on many programs. An investor who has modeled the deal down to the last dollar of proceeds discovers a $4,000 reserve shortfall at the clear-to-close stage. The specifics vary more than any other DSCR condition, and the ranges are laid out in the reserves and seasoning guide.

Where the strategy still works, arithmetically

BRRRR requires a gap between all-in cost and ARV of at least 25%, because the refinance caps at 75%. Markets where that gap exists in 2026 are the ones with meaningful C4-to-C3 condition spreads and rent-to-price ratios above roughly 0.85%.

Cleveland remains the clearest example: purchase basis in the $70,000 to $110,000 range on stock that supports $1,200 to $1,500 rents, with the caveat that Cuyahoga County effective tax rates above 2.3% eat coverage and many lenders impose loan minimums of $75,000 to $100,000 that exclude the cheapest end of the inventory entirely. The market-specific numbers are in the Cleveland metro analysis, and the county-by-county tax spread that decides which suburbs work is in the Ohio state overview.

Indianapolis works differently — higher entry basis, materially lower tax rates around 1.05% for owner-unoccupied residential under Indiana's cap structure, and stronger rent growth, which means coverage rarely binds before LTV does. That makes it a more forgiving BRRRR market even though the spread on paper looks tighter. The Indianapolis breakdown covers the tax cap detail, which is the single most consequential difference between the two markets.

Sequencing the lender relationship

The most common structural error is treating the refinance lender as a vendor to be sourced in month five. Rehab funding usually comes from hard money at 10% to 12% with 2 points, and that clock is running while the takeout is being shopped.

The better sequence is to get the takeout lender's guidelines in writing before the acquisition closes: minimum loan amount, seasoning requirement, maximum cash-out LTV at the borrower's actual credit tier, DSCR floor, reserve requirement, whether the lender uses in-place or appraised rent, and whether a lease signed by a tenant who moved in three weeks ago counts as rental history. Those seven answers determine the proceeds. Nothing about them requires a property address, which means they can be collected during due diligence at zero cost.

Program boxes differ enough on seasoning and cash-out LTV that the choice of lender is worth more than a quarter point of rate on a BRRRR file. Comparing those boxes side by side across the lender directory before the offer, and shortlisting two rather than one, is what prevents the month-five scramble — and the second lender matters, because the first one's appraisal review can come back with a value the file cannot survive.

What a realistic outcome looks like

On the working example: $141,000 in, $142,000 appraised, $106,500 loan at 75% LTV with DSCR clearing at 1.21, minus roughly $3,800 in refinance closing costs and the hard money payoff. Capital recovered is about $102,700 against $141,000 deployed — 73%, not the 100% the model promises. The remaining $38,300 is trapped equity that produces $234 a month of cash flow after PITIA and a 10% management and maintenance reserve.

That is a good deal. It is not a full capital recycle, and the difference between the two is a 75% LTV ceiling that was never going to move. Investors who underwrite to 73% recovery and get it repeat indefinitely. Investors who underwrite to 100% recovery run out of cash on deal three, which is the actual failure mode the strategy hides.