Stress-Test DSCR With Vacancy Scenarios
A worked explanation of stress-test dscr with vacancy scenarios, including the key limitation and a checkable example.
How to Stress-Test DSCR With Vacancy Scenarios
A borrower should model debt coverage by computing the Debt Service Coverage Ratio (DSCR) for each plausible NOI state, not by averaging them. A single average number hides the risk that a specific vacancy level would break the loan. The correct method is to stress-test DSCR with vacancy scenarios: list each NOI outcome, assign it a probability, calculate DSCR per scenario, then compare every DSCR against the lender's minimum threshold. The lender will use the worst-case DSCR, not the average.
A Worked Example
Work a concrete example using the base NOI of $100,000 and annual debt service of $80,000 from the page-specific fact sheet. That debt service comes from a loan with a $6,666.67 monthly payment. Three plausible vacancy states exist:
- Scenario A (90% occupancy): NOI = $90,000. DSCR = $90,000 / $80,000 = 1.125x. Probability of this state = 0.6.
- Scenario B (80% occupancy): NOI = $80,000. DSCR = $80,000 / $80,000 = 1.00x. Probability = 0.3.
- Scenario C (70% occupancy): NOI = $70,000. DSCR = $70,000 / $80,000 = 0.875x. Probability = 0.1.
The borrower runs three separate DSCR calculations, one per scenario. The lender will look at the lowest DSCR, 0.875x, and likely reject the loan because it falls below every agency minimum: Fannie Mae and Freddie Mac require 1.25x, and CMBS lenders demand 1.25x on stressed NOI. The borrower should also stress test at a 10% vacancy on the base NOI, which gives $90,000 NOI and a 1.125x DSCR, still short of 1.25x.
Calculation Steps
To calculate the probability-weighted NOI, use the formula from the fact sheet: Expected NOI = Σ (NOI_i × P_i). For the three scenarios, that is ($90,000 × 0.6) + ($80,000 × 0.3) + ($70,000 × 0.1) = $54,000 + $24,000 + $7,000 = $85,000. The probability-weighted DSCR is $85,000 / $80,000 = 1.0625x. That number is what an expected value calculator would produce, but it is the wrong number to give a lender. The 1.0625x DSCR makes the deal look barely viable when every individual scenario except the most optimistic is below 1.25x.
Build a table to compare the scenarios and the probability-weighted result against the lender's minimum.
| Scenario | NOI | Probability | DSCR | Passes 1.25x? |
|---|---|---|---|---|
| A (90% occ.) | $90,000 | 0.6 | 1.125x | No |
| B (80% occ.) | $80,000 | 0.3 | 1.00x | No |
| C (70% occ.) | $70,000 | 0.1 | 0.875x | No |
| Probability-weighted | $85,000 | 1.0 | 1.0625x | No |
Why the Method Works
Probability-weighted NOI is a borrower's planning tool, not a substitute for lender underwriting. The lender uses the worst-case or a stressed NOI, such as 75% occupancy, regardless of the borrower's probability estimates. In this example, the base DSCR is exactly 1.25x ($100,000 / $80,000), which provides zero cushion for vacancy or expense creep. A single month of 80% occupancy would push the DSCR to 1.00x, meaning the property generates exactly enough NOI to cover debt service and nothing else. The borrower would have no funds for repairs, leasing commissions, or capital expenditures.
One specific failure case: the borrower uses the probability-weighted DSCR of 1.0625x to justify the loan to a lender. The lender runs the underwriting NOI, which includes a 5-10% vacancy stress, and calculates a DSCR of 1.125x at best. The loan is denied because the lender's minimum is 1.25x on underwritten NOI. The borrower wasted application fees, appraisal costs, and 60 days of rate lock risk. The probabilities are subjective estimates; the lender may assign different probabilities or ignore them altogether and use a deterministic stress scenario.
The borrower is better served by reducing the loan amount or increasing the equity injection so that even the worst plausible vacancy produces a DSCR of at least 1.25x. For the scenarios above, the annual debt service must fall to $56,000 to make the 70% occupancy scenario pass at 1.25x. That means a monthly payment of $4,666.67, which would lower the loan amount by roughly 30% from the $80,000 debt service case. That is the trade-off a borrower must accept when NOI can land in several plausible states.
Common Questions
How many vacancy scenarios should a borrower model?
Model at least three: base occupancy, a moderate stress (e.g., 80% occupancy), and a severe stress (e.g., 70% occupancy). The lender will likely use a single stressed scenario between 75% and 90% occupancy, so your moderate and severe cases should bracket that range.
Does the probability-weighted DSCR ever help the borrower?
It helps the borrower understand the expected NOI over a long holding period, but never for a single underwriting decision. The lender cares about the minimum DSCR, not the average, because a single bad year can cause a default before the good years arrive.
What if the borrower's probability estimates are accurate?
Accuracy does not matter. Lenders use deterministic stress tests based on their own underwriting standards, not borrower-provided probabilities. The Fannie Mae and Freddie Mac minimum of 1.25x applies to the underwritten NOI, which already includes a built-in vacancy and expense stress.
Can the borrower use the probability-weighted DSCR to negotiate a lower rate?
No. The rate is set by the loan's risk profile, which includes the property's DSCR, LTV, and debt yield, not by the borrower's probability-weighted average. The lender will quote a spread over the 10-year Treasury based on the worst-case DSCR, not the average.
What is the most common mistake borrowers make when stress-testing DSCR?
Using gross rents instead of net operating income after vacancy and expenses. A borrower who uses gross rents might calculate a DSCR of 1.40x, but the real DSCR after the 10% vacancy and 35% expense ratio is 0.95x. The "DSCR miscalculation" failure mode is the most frequent cause of loan rejection.