Skip the math: use our free interactive calculator, live results, embeddable, no signup.
DSCR measures what lenders care about: whether the property’s cash flow, today, covers its debt service, today. A slow drift in the income stream, such as higher operating expenses or a tenant that does not renew and takes months to backfill at lower rent, can be absorbed quietly by a model focused on yield and IRR, then surface in the lender’s DSCR test at refinance, when the loan sizes smaller than the model predicted.
This is the metric that drives loan sizing in CRE. It’s also a metric with predictable failure modes, and the gap between how brokers use it and how sophisticated buyers use it is wider than most analysts realize.
DSCR comes up at every stage of a deal: LOI, financing, refinance and exit. This guide covers what a DSCR loan is, how the ratio is calculated, which NOI lenders actually test, and where the metric breaks.
What DSCR actually measures
Debt service coverage ratio is net operating income divided by annual debt service.
DSCR = NOI / Annual Debt Service
A 1.00x DSCR means the property exactly covers its mortgage payments from operations. A 1.25x DSCR means NOI is 25% above what’s needed to service the loan, a 25% cushion before coverage breaks. A 0.95x DSCR means the borrower has to fund the gap out of pocket every month.
The metric exists because lenders aren’t equity investors. They don’t care about IRR or yield-on-cost or appreciation. They care about whether the property will pay them on time, every month, until the loan matures. DSCR tests that directly: today’s NOI, today’s debt service, with a buffer.
Equity investors compute DSCR for a different reason. It tells them how levered they actually are once cash flow is netted against debt service. A property with a 1.05x DSCR is operating on knife-edge coverage. A small rent loss, a small expense increase, a tenant default: any of those moves coverage below 1.00x, and the equity has to step in. A property with 1.40x coverage absorbs those same shocks without a problem.
What is a DSCR loan?
A DSCR loan is a mortgage sized and qualified on the property’s debt service coverage ratio, NOI divided by annual debt service, rather than on the borrower’s personal income. Two different things share the name:
- Commercial real estate. It is not a separate product. Every commercial mortgage is effectively a DSCR loan: the underwriter computes NOI, sizes debt service against a minimum coverage multiple and sets the loan amount. The sponsor’s balance sheet still matters for guaranties, recourse carve-outs and net worth and liquidity covenants, but the property drives the loan size.
- Residential investor market. A “DSCR loan” is a specific non-QM product sold by specialty lenders for single-family rentals, two-to-four unit properties and small multifamily. It qualifies on the rent the property earns (from a lease or a market rent appraisal), the borrower’s credit and reserves, and the coverage ratio, without W-2s, pay stubs or tax returns. Terms, rates and minimum coverage vary by lender and program, and lower coverage usually comes with a lower LTV or a higher rate, so get quotes from lenders rather than relying on a general range.
Both use the same formula, DSCR = NOI / annual debt service, against a minimum the lender sets in its term sheet. This guide focuses on commercial underwriting, but the mechanics apply to both.
How debt service is measured
For a fixed-rate loan, annual debt service is twelve months of principal plus interest at the proposed rate and amortization. For an interest-only loan it is interest alone, which flatters the ratio, so many lenders also test against a fully amortizing payment or pair DSCR with debt yield. For a floating-rate loan, lenders stress the rate, typically at the strike of a required interest-rate cap plus a cushion. Many lenders size on a constant based on a 30-year amortization even when the actual loan is shorter, because the test is whether the property could carry a typical takeout loan.
Why lenders use DSCR
Three reasons:
It’s measured, not modeled. Cap rates depend on what investors are willing to pay. LTV depends on the appraisal. DSCR depends on income and debt service: two numbers the lender can verify against operating statements and the term sheet.
It scales across asset classes. A 1.25x DSCR works as a coverage test whether the loan is on a Class A apartment building in Austin, a Class B office in Cincinnati, or an industrial portfolio across three states. Cap rates and per-SF metrics don’t translate across asset classes; DSCR does.
It measures what lenders care about. Coverage is a core credit test because it measures the thing lenders care about: whether operations can pay the debt. The metric isn’t perfect, but it’s good enough that every lender runs it.
This is why DSCR shows up in every commercial mortgage origination, every refinance, every CMBS deal review, every covenant package, and every regulatory exam. It’s the universal language of CRE credit.
A worked example
Pull a 100,000 SF Class A industrial building. Trailing-12 NOI of $1.2M on the operating statement. The seller is asking $20M. You’re underwriting it.
The lender will compute their own NOI. Here’s how the inputs typically move:
- Vacancy adjustment: Building is currently 100% leased. Lender applies market vacancy of 5%. NOI drops $75K.
- Management fee: Seller self-manages. Lender imputes 3% market fee. NOI drops $45K.
- Replacement reserves: Seller’s NOI excludes them. Lender adds $0.20/SF, or $20K.
- Tax reassessment: The property is in a state where a sale resets the assessed value, and the lease recovers none of the increase. Taxes were $180K on the seller’s basis; the lender projects $260K at the $20M purchase price. NOI drops $80K. (On a true NNN lease most of the increase would be reimbursed by the tenant and the hit would be the non-recovered share only.)
Result: lender’s NOI is $1.2M minus $75K minus $45K minus $20K minus $80K, or $980K. About 18% below the seller’s number.
At a 6.50% / 30-year loan, the loan constant is 7.58%. If the lender wants a 1.25x DSCR:
- Maximum debt service = $980K / 1.25 = $784K
- Maximum loan = $784K / 7.58% = $10.34M
That’s a 52% LTV, even though the lender quoted 65%. DSCR is the binding constraint, not LTV.
If you’d run the same math on the seller’s NOI ($1.2M instead of $980K), you’d have come up with a $12.7M loan, about 63% LTV, and a financing assumption that turned out to be 23% too high. That’s the gap that breaks deals at the financing contingency.
DSCR versus LTV, debt yield, and yield-on-cost
DSCR is one of four standard credit and return metrics, and each measures something different:
Loan-to-value (LTV). Loan amount divided by appraised value. Measures leverage as a percentage of asset value. Sensitive to cap rate compression and appraisal optimism. LTV caps are set by each lender and program, and are usually lower for transitional assets than for stabilized ones.
Debt service coverage ratio (DSCR). NOI divided by annual debt service. Measures cash flow cushion above debt service. Sensitive to the interest rate: a higher rate increases debt service and reduces DSCR. Minimums are set by the lender in the term sheet.
Debt yield. NOI divided by loan amount. Measures the lender’s unleveraged return on principal if they took the property back. Insensitive to interest rates and cap rates. There is no industry-wide floor; minimums are set in the term sheet.
Yield-on-cost. Stabilized NOI divided by total project cost (purchase price plus capital). Measures unleveraged return for the equity investor. Doesn’t show up in the debt sizing equation but is the equity’s complement to the lender’s metrics.
A modern commercial lender sizes the loan to the most restrictive of LTV, DSCR, and debt yield. Each can be the binding constraint on different deals. In a cap-rate-compressed market, debt yield is usually binding. In a high-rate environment, DSCR is usually binding. In a low-rate environment with rich appraisals, LTV is usually binding.
This is why “70% LTV” rarely means a 70% loan. The advertised LTV is the ceiling. DSCR or debt yield is usually the actual ceiling.
Where DSCR breaks down
DSCR is a useful metric with predictable limits. Four failure modes:
1. Interest-only periods flatter coverage
A loan with a five-year interest-only period has lower debt service during the IO term than during the amortizing period. DSCR during IO might be 1.50x; DSCR during amortization might be 1.20x. The lender knows this. Most underwrite to a “constant” payment based on 30-year amortization regardless of the actual structure, or they pair DSCR with debt yield to neutralize the IO benefit. Equity investors should do the same: model both DSCR-IO and DSCR-amortizing.
2. Capitalized interest hides debt service
On construction and transitional bridge loans, interest is often capitalized into the loan balance during the lease-up period. The borrower doesn’t actually pay debt service from operations; it’s added to the loan principal. DSCR during this period is meaningless because debt service from operations isn’t happening. The relevant metric is the stabilized DSCR at takeout, not the going-in DSCR.
3. Lease-up assumptions create theoretical coverage
A value-add deal at acquisition might have NOI of $600K against debt service of $850K, a 0.71x DSCR. But the bridge lender underwrote against a stabilized NOI of $1.4M, producing a 1.65x stabilized DSCR. The deal works on paper. The risk is that lease-up takes longer than projected, costs more than projected, or comes in at lower rents than projected, and the stabilized NOI never materializes. DSCR can’t capture lease-up execution risk; it just reports the math on whatever NOI you plug in.
4. Rollover concentration
A 1.30x DSCR on a building with 47% of GLA expiring in 18 months is a different deal from a 1.30x DSCR on a building with weighted average lease term of nine years. The point-in-time coverage looks identical. The risk-adjusted coverage isn’t. Sophisticated lenders apply rollover-adjusted NOI: haircutting NOI for expiring leases, then computing DSCR on the haircut number. The result can be 0.20-0.30x below the as-reported DSCR.
Stress testing is the actual disclosure
A buyer who only computes one DSCR number is asking the wrong question. The IC-grade analysis shows DSCR at:
- Base case (forward NOI, current rate, current debt service)
- +50 bps rate shock
- +100 bps rate shock
- -10% NOI shock
- Combined shock (rate up, NOI down)
- DSCR at refi (forward to maturity, projected NOI, takeout rate assumption)
- Rollover-adjusted DSCR (NOI haircut for expiring leases)
If coverage holds above 1.10x in the combined-shock scenario, the deal is fundable. If it doesn’t, you have a rate-sensitivity problem the committee will spot.
What NOI do lenders use to calculate DSCR?
Lenders almost never use the seller’s NOI. They build their own, and the adjustments repeat on nearly every deal:
- Vacancy and credit loss. A fully leased building still gets a market vacancy and credit loss factor, set by the lender for the asset type and market.
- Management fee. A self-managing seller shows no fee, so the lender imputes a market fee on effective gross income.
- Replacement reserves. Seller NOI usually excludes them; the lender deducts them for sizing. It is a non-cash sizing deduction, not a prediction of capex.
- Real estate taxes. Sellers show trailing taxes. Where a sale resets the assessed value, lenders reset taxes to the purchase price. Whether a sale triggers reassessment, and by how much, depends on the state, so check the local rule instead of assuming. On net leases, only the non-recovered share of an increase reduces NOI.
- Free rent burn-off. Rent still in an abatement period is not credited until the abatement ends.
- Tenant credit and rollover. Lenders may haircut weaker tenants’ rent and stress near-term expirations.
- Step rents. Some lenders use average rent over the lease term, some use Year-1 contractual rent and some mark short leases to market. Each yields a different NOI, so ask which method applies.
Stack them and the lender’s NOI lands below the offering memorandum figure, by a wider margin on value-add or transitional assets.
Why deals fail the DSCR test
The failures cluster. The seller’s NOI was inflated and normalization removes the cushion; rates moved between LOI and rate lock; a major tenant’s expiration is stressed out of NOI; or taxes reset after the sale. A fifth, more recent one is a lender sizing to a stressed exit cap rate at maturity, which makes DSCR at refinance worse than DSCR at closing. The fix is usually more equity, a lower price, a rate lock or cap, an interest-only period, pre-leasing or an extension before closing, or a different debt structure (mezzanine, preferred equity, agency), rather than a different lender.
What sophisticated buyers do instead
The institutional version of DSCR analysis treats the metric as one input in a four-part credit framework:
- DSCR sensitivity table. Coverage at base case, rate stress, NOI stress, and combined stress.
- Debt yield as a parallel test. NOI divided by loan amount, rate-independent.
- DSCR-at-refi. Coverage projected to maturity, with forward NOI and stressed rates.
- Rollover-adjusted analysis. NOI haircut for lease expirations weighted by probability of renewal and probable rent movement.
Each of these requires lease-level detail. Computing rollover-adjusted DSCR means you need the rent roll, the lease abstract for every tenant, the expiration schedule, the renewal options (and their economics), and an assumption about whether each tenant will renew. On a portfolio with 80 tenants, that’s a multi-day analyst project before you even get to the financing model.
This is where Moraine saves time. It extracts lease terms with a citation back to the source page, builds a normalized rent roll, and its report catalog includes debt and sensitivity reports for office, retail and industrial deals. A qualified underwriter should still review the rollover and rate assumptions before IC. For a quick calculation, use the DSCR calculator.
The single number on the screening model is the starting point. The sensitivity table is the deliverable.