Guide

Debt service coverage ratio (DSCR) explained

Harbor Properties owns twelve suburban office parks financed with non-recourse mortgages. In Q1 the asset manager reported interest coverage of 3.1× at the portfolio level — well above the 2.0× internal warning line. Six weeks later the special servicer issued a notice: the debt service coverage ratio (DSCR) on the largest loan had fallen to 1.12×, below the 1.25× covenant minimum. The problem was not occupancy collapse; it was the debt service stack. Floating-rate interest had risen $4.2 million year over year, and a 2024 refinancing moved $18 million of bullet maturity into a five-year amortization schedule that added $6.8 million of annual principal. Net operating income still covered coupons comfortably, but total scheduled payments — interest plus principal — consumed nearly all of it. Harbor sold two non-core buildings, used proceeds to pay down the highest-coupon tranche, and renegotiated a rate cap through 2028. Within three quarters portfolio DSCR recovered to 1.48× and stabilized above the 1.30× internal floor. The DSCR is the lender's question in one number: does cash flow from the asset or business cover the full payment obligation, not just the coupon?

This guide covers DSCR formulas (NOI, EBITDA, and CFADS variants), how DSCR differs from fixed charge coverage and interest coverage, covenant floors by asset class, the Harbor Properties refactor, a technique decision table, pitfalls, and an investor checklist.

What DSCR measures

The debt service coverage ratio compares cash available to pay debt to total debt service — typically interest plus scheduled principal amortization on term debt and sometimes mandatory lease payments or sinking-fund contributions. Unlike interest coverage, which stops at the coupon, DSCR asks whether the borrower can meet the contractual payment schedule without refinancing, drawing reserves, or selling assets.

DSCR appears everywhere lenders need amortization-aware underwriting:

Commercial real estate — NOI divided by annual mortgage payment (P&I).
Project finance and infrastructure — CFADS (cash flow available for debt service) over senior debt service.
Corporate term loans — EBITDA or EBIT over interest plus scheduled principal (sometimes called “total debt service” in credit agreements).
Small-business and SBA lending — business cash flow over P&I on the proposed loan.

A DSCR below 1.0× means cash flow does not cover scheduled debt payments without external support. Lenders rarely underwrite at 1.0×; they require a cushion — often 1.20×–1.35× for stabilized real estate and higher for cyclical corporates.

Formula and worked example

The canonical real-estate form:

DSCR = Net operating income (NOI) ÷ Total debt service

Where total debt service = interest + scheduled principal (P&I). Corporate credit agreements may substitute EBITDA or EBIT in the numerator:

Corporate DSCR = EBITDA ÷ (Interest + Scheduled principal)

Project finance often uses:

DSCR = CFADS ÷ Senior debt service

CFADS starts from operating cash flow, subtracts maintenance capex and taxes, and excludes discretionary distributions — cash truly available to creditors before equity takes anything.

Harbor Properties — Harbor Park III (simplified, $ millions):

Effective gross income: $42
Operating expenses (excl. debt): $18
NOI: $24
Interest expense: $11.2
Scheduled principal: $6.8
Total debt service: $18.0

DSCR = $24 ÷ $18.0 = 1.33× on a stabilized basis. After the rate step-up and amortization shift in the stressed quarter:

NOI (after two tenant move-outs): $20.2
Total debt service: $18.0

Stressed DSCR = $20.2 ÷ $18.0 = 1.12× — covenant breach at 1.25× minimum. Interest coverage on the same quarter: EBIT proxy ≈ $20.2, interest $11.2 → 1.8× on interest alone (or higher if using NOI/interest = 1.8×). The gap between comfortable interest coverage and thin DSCR is the principal and rate story.

Post-refactor (asset sales pay down $45M principal, rate cap limits further increases, re-leasing fills one vacancy):

NOI: $22.8
Total debt service: $15.4

DSCR = $22.8 ÷ $15.4 ≈ 1.48×.

DSCR vs interest coverage vs FCCR vs net debt/EBITDA

Credit metrics stack from narrow to broad obligation coverage:

Metric	Numerator (typical)	Denominator	Principal included?
Interest coverage	EBIT or EBITDA	Interest only	No
Fixed charge coverage (FCCR)	EBIT + fixed charges	Interest, leases, preferreds (varies)	Sometimes partial
DSCR	NOI, EBITDA, or CFADS	Interest + scheduled principal	Yes
Cash flow to debt	CFO or FCF	Total debt outstanding	Stock, not flow
Net debt/EBITDA	EBITDA (implicit)	Net debt	Leverage years, not coverage

Harbor Properties illustrates the trap: strong interest coverage masked a DSCR breach because amortization and floating rates expanded the denominator. Conversely, a company with interest-only debt may show weak net debt/EBITDA but acceptable DSCR until maturity wall hits. Use DSCR for payment capacity, net debt/EBITDA for balance-sheet leverage, and cash flow to debt for deleveraging speed from the cash flow statement.

NOI, EBITDA and CFADS: which numerator?

The numerator must match what lenders treat as available cash:

NOI (real estate) — rental revenue minus operating expenses before debt service, capex reserves, and depreciation. Standard for stabilized property loans.
EBITDA (corporate) — operating earnings before interest, taxes, D&A. Watch adjusted EBITDA add-backs in non-GAAP footnotes; covenants often cap synergies and one-time items.
CFADS (project finance) — after maintenance capex and taxes; closest to creditor cash. See free cash flow for the broader equity residual.

Maintenance capex treatment splits models. Some real-estate DSCR calculations subtract replacement reserves from NOI before dividing by debt service (sometimes called “DSCR after capex” or “DSCR_nc”). Corporate agreements may deduct maintenance capex from EBITDA in the numerator. Always align with the credit agreement — two analysts can compute 1.35× and 1.18× on the same asset with different reserve assumptions.

Covenant floors and lender thresholds

Illustrative minimum DSCR bands at origination (actual terms vary):

Context	Typical minimum DSCR	Notes
Stabilized multifamily (agency)	1.20× – 1.25×	Often tested on underwritten NOI, not trailing
Office / retail (bank)	1.25× – 1.35×	Higher for single-tenant or short WALT
Project finance (senior)	1.30× – 1.50× average life	Sculpted debt may allow lower early-year DSCR
Corporate term loan	1.50× – 2.00× (EBITDA-based)	Tested quarterly; cure periods common
SBA 7(a) business acquisition	1.15× – 1.25× (global)	Includes owner compensation adjustments

Breach mechanics matter as much as the ratio. Cash trap, blocked distributions, mandatory amortization acceleration, and equity cure rights appear in loan agreements. A borrower at 1.26× against a 1.25× floor has effectively no cushion — model a 5% NOI decline before calling the loan “safe.”

Harbor Properties refactor

Harbor's remediation followed a standard special-servicer playbook:

Asset sale — two secondary-market buildings at 6.8% cap raised $62M; $45M applied to highest-spread loan, cutting principal service $2.1M annually.
Rate risk — purchased interest-rate cap through 2028, limiting further floating-rate expansion in the denominator.
Occupancy — re-leased 38k sq ft at modestly higher rent; NOI recovery $1.4M run-rate.
Covenant renegotiation — temporary 1.15× floor for two quarters in exchange for cash-sweep and prohibited distributions.

Portfolio-weighted DSCR moved from 1.12× trough to 1.48× without equity injection. The lesson for equity investors: interest coverage alone would not have flagged the breach early enough; DSCR and maturity/amortization schedules did.

Technique decision table

Your question	Best metric	Why not DSCR alone?
Can EBIT pay coupons?	Interest coverage	DSCR double-counts principal concern when debt is bullet or interest-only
Can cash pay P&I on this building?	NOI-based DSCR	Interest coverage ignores principal; net debt/EBITDA ignores payment timing
Lease-heavy retailer solvency?	FCCR	DSCR may miss operating lease rent if not in “debt service” definition
Years to pay down debt from earnings?	Net debt/EBITDA	DSCR is flow coverage, not stock leverage
Deleveraging from operating cash?	Cash flow to debt	DSCR uses scheduled service, not total debt balance

Common pitfalls

Using interest-only in the denominator — that is interest coverage, not DSCR.
Ignoring floating-rate paths — stress the denominator at forward curves, not trailing coupons.
Underwriting NOI without vacancy/credit loss — pro forma DSCR on 100% occupancy misstates risk.
Mixing numerators — EBITDA DSCR compared to a peer quoted on NOI DSCR.
Excluding reserves — lender definitions often subtract replacement reserves; headline NOI DSCR overstates coverage.
Forgetting maturity walls — interest-only periods inflate DSCR until bullet repayment refi risk arrives.
Consolidated vs asset-level — a parent can look fine while a non-recourse SPE breaches.

Investor checklist

Identify debt service definition in the credit agreement or offering memo.
Compute trailing and forward DSCR at base and stressed rate scenarios.
Separate interest coverage and DSCR — flag gaps wider than 0.5×.
Map amortization schedule: when does principal step up?
For real estate, build NOI from rent roll with vacancy and concessions.
Subtract maintenance capex or reserves if covenant requires it.
Check consolidated vs property-level tests for REITs and SPE structures.
Cross-check net debt/EBITDA and cash flow to debt for leverage context.
Read breach remedies: cash trap, cure periods, equity contribution rights.
Stress NOI or EBITDA −10% and recompute distance to covenant floor.

Key takeaways

DSCR = cash available ÷ total debt service — interest plus scheduled principal, not coupon alone.
Interest coverage can look fine while DSCR breaches — Harbor Properties at 3.1× interest vs 1.12× DSCR.
Numerator choice matters — NOI for property, EBITDA or CFADS for corporate and project finance.
Covenant floors typically sit at 1.20×–1.50× depending on asset class and structure.
Pair DSCR with net debt/EBITDA and maturity profile — coverage and leverage tell different stories.