Debt Service Coverage Ratio (DSCR): The Ratio That Sizes the Loan

A building can lose twenty million dollars of borrowing capacity without a single thing changing inside it.

Almost every explanation of the debt service coverage ratio in circulation was written by a lender, for a borrower being assessed. It describes a test to pass. But from the sponsor's side of the table, the ratio determines how much money shows up at closing, and it keeps deciding for years afterward, at every extension test and every quarterly reporting date.

What is the debt service coverage ratio?

The debt service coverage ratio measures whether a property's income covers its loan payments, and by how much. It is expressed as a multiple: a 1.30 DSCR means the building throws off $1.30 of net operating income for every dollar it owes the lender in debt service that year.

Both halves have to be measured correctly. Net operating income stops above the debt line: after operating expenses, before debt service, capital projects, and income taxes.

The denominator is the part that gets misread. Debt service means more than the interest bill. In an amortizing loan, the full annual payment is interest plus principal, because a lender underwriting a mortgage is not asking whether the building can service the interest indefinitely. It is asking whether the building can pay the loan down on schedule.

A 1.0 ratio is the breakeven line, where operating income exactly equals the annual payment, leaving nothing over. A bank will not write that loan against a stabilized building, one that is leased up and running at normal occupancy, on permanent debt: it sits a rounding error away from a missed payment.

And coverage above 1.0 is not profit, which is the second common misread. It is the cushion between the building's performance and the lender's downside. Most of it goes to reserves, capital projects, and absorbing the ordinary swings of an operating asset.

Transitional deals are the exception. But a building in lease-up or mid-renovation can produce positive net operating income and still fail to cover its debt service. Lenders fund those anyway, either by setting the required coverage far lower than a stabilized deal would require or by sizing an interest reserve into the loan and drawing on it until the income arrives. The reserve is the lender betting on the business plan.

The DSCR formula in practice

Take a stabilized asset bought for $50 million at a 5.25 percent going-in cap rate, financed with a $33 million loan at 4.50 percent fixed, amortizing over 30 years with no interest-only period.

Base case

Everything that follows moves one variable in that base case and leaves the rest alone.

What moves the ratio

Annual debt service in real estate is twelve months of scheduled principal and interest. Four things move it.

The interest rate

Rate is the variable sponsors watch. Hold the loan at $33 million, and the same building goes from comfortably financeable to unfinanceable when the rate moves three points. The building never changes.

$33M loan, $2,625,000 NOI, 30-year amortization

The right-hand column is the one that matters when refinancing. The building and its operating income have not changed, but three points of rate takes $9.5 million off what a lender will fund against it.

The amortization period

The loan term and the amortization schedule are different numbers, and the payment is calculated off the schedule. A five-year loan amortizing over 30 years is common; so is a ten-year loan on a 20-year schedule. And shortening the schedule increases the annual payment, which reduces coverage on identical income.

$33M loan at 4.50%, NOI held at $2,625,000

That bottom row is also the interest coverage ratio: the same building measured against interest alone rather than the full payment. The gap between 1.77 and 1.31 is amortization. So an interest-only period is worth negotiating for, and coverage steps down when one expires.

Because the loan matures long before a 30-year schedule finishes, the unpaid balance comes due as a balloon payment at maturity.

Net operating income

Rate and amortization are set at closing. After that, the variable that moves is income, and it moves down. A covenant tested quarterly is a test on this table, not on the other two.

$33M loan at 4.50%, 30-year amortization. Debt service holds at $2,006,474 throughout, because the loan does not care how the building is doing.

The useful number is the distance to the covenant, not the ratio itself. A 1.31 against a 1.20 minimum means net operating income can fall 8.3 percent before the loan breaks. Against a 1.30 minimum, the same deal has almost no room. So two lenders can look at the same building, quote the same rate, and hand the sponsor very different operating margins.

The day-count convention

Interest calculated as actual/360 rather than 30/360 charges for the actual number of days in each month against a 360-day year. A 6.75 percent all-in rate becomes an effective 6.84 percent. Nine basis points looks cosmetic. It lands in the denominator.

On floating-rate debt, the denominator moves on its own. A loan priced over SOFR, the Secured Overnight Financing Rate, resets each interest period. A rate floor protects the lender's yield if the index falls; a required interest rate cap, which the borrower buys, limits how far debt service can rise if the index rises. So sponsors underwrite against a projected forward curve rather than the spot rate, and Chatham Financial is one of several providers that publish one.

What is a good debt service coverage ratio?

Conventional minimums cluster between 1.20 and 1.30. A 1.20 or a 1.25 is what comes up most often. Most banks would describe a 1.30 as safe; 1.20 is also common but leaves less margin for error.

Bridge and transitional loans go lower, and a 1.10 minimum appears in real term sheets for value-add deals where income is expected to grow into the debt.

The range runs upward too. Some long-term-hold developers underwrite to 1.5 or 2.0, accepting lower leverage in exchange for surviving a downturn without a conversation with the lender. That is a strategy choice, not a lending requirement.

Treat all of these as conventional defaults rather than published thresholds. Minimums are negotiated and vary by asset class, lender type, where a property sits in its business plan, and how badly the lender wants the deal, which is why two experienced people will give different answers about what “healthy” looks like. They also drift with the credit cycle. Of course, the federal banking agencies have periodically issued joint statements flagging “an easing of CRE underwriting standards” and urging institutions to maintain discipline. There is no industry-wide number to look up.

How lenders size a loan off it

Coverage is not a pass-fail gate applied to a loan amount someone else decided. Thesis Driven founder Brad Hargreaves calls it “often the main limiting factor to loan size, not LTV.” It is one of several tests that together determine the loan amount, and the one that binds changes to the deal. On stabilized assets, coverage is often it. In development and value-add deals, where income has not yet arrived, cost and debt yield tend to bind first.

A loan's maximum potential size is determined by several tests, and banks are willing to fund the most restrictive (smallest) result. One real bridge structure capped the total commitment at the lesser of a stated dollar figure, 65 percent of as-stabilized value, and 75 percent of project cost. The closing proceeds then ran through a second stack: loan-to-cost, loan-to-value, a minimum debt yield, and the coverage stipulation.

Consider a stabilized asset producing $8.5 million of net operating income, refinancing into a 30-year amortizing loan. At 3.75 percent, it supports $127.5 million, with annual debt service of $7.09 million and coverage at 1.20. But reprice the same building at 5.25 percent, and the payment climbs to $8.45 million. Coverage falls to 1.01, a margin of roughly $50,000 a year on a nine-figure loan.

No bank makes that loan. Holding coverage at 1.20, the same building supports roughly $107 million, about 16 percent less, on identical income.

A refinancing sized to repay limited partner capital and accrued preferred return no longer repays it. The pref keeps accruing on the shortfall, and the promote, which is the sponsor's share of profits above that hurdle, never comes into reach.

The debt service coverage ratio covenant

Most sponsors think of default as a missed payment. But a coverage covenant obliges the borrower to maintain the property at a stated coverage level for the life of the loan, tested periodically against numbers the lender recalculates, which means a loan being paid on time can lose good standing without anyone missing a wire.

A hit to net operating income that lowers the ratio below that level puts the borrower in technical default, even if all payments are current and the lender is fully paid.

The ratio the lender tests is not the one in the sponsor's model. Coverage in a term sheet is based on underwritten net operating income, which is the lender's own version of the number after applying a vacancy factor, not crediting leases that are expiring soon or not yet commenced, and normalizing for one-time items.

Jay Dunn, a former Morgan Stanley real estate banker, has “negotiated loan agreements where the underwriting adjustments are multiple pages long.” Each adjustment lowers the numerator.

The steps that follow a breach are incremental and determined by the loan documents. A lender generally has to give notice and allow a cure period. But remedies escalate from a cash sweep that traps distributions at the property, to penalty interest, to acceleration of the full balance. Where payments are current, and only the covenant has broken, the sweep is the remedy a sponsor is most likely to meet. It can cut off distributions to limited partners in a quarter where the building is performing.

The covenant is negotiable. A lender wants the required coverage set high; a sponsor wants it low, because a tight covenant constrains spending on the asset for years. The test also recurs: exercising an extension option re-triggers a fresh appraisal, a debt yield test, and a coverage test, so a loan that cleared in year one has to clear again in year three.

DSCR versus debt yield

Debt yield is net operating income divided by the loan amount: the unlevered return the lender would earn if it took the building back. Where debt yield equals the cap rate, the loan equals the property value and the deal is fully levered, so lenders want the yield well clear of it. Bridge lenders commonly look for 9 to 10 percent or better against stabilized income. A going-in figure as low as 5 percent appears in value-add deals with below-market rents, where the lender is underwriting income that does not yet exist.

So lenders watch both, and for a structural reason.

Coverage moves with the rate, the amortization schedule, and the day-count convention. Debt yield moves with none of them; it sees only income relative to the loan balance. A borrower can improve coverage with a longer amortization or an interest-only period, and cannot improve debt yield at all.

A cap rate indicates what a building is worth to the market. Coverage tells a lender how much of that is fundable. Sponsors who track the first and assume the second follows are the ones who find out at maturity.

Keep building your real estate finance stack

→ Workshops on underwriting, capital raising, and development: thesisdriven.com/workshops. Live sessions with Thesis Driven's principals and guest operators on the mechanics this guide touches, from building a pro forma to structuring the capital stack that sits on top of it.

Related reading:

• Making the Leap to Long-Term Holding
• AI Comes for CRE Lending
• Capital to the Rescue