LTV moves with cap rates. DSCR moves with interest rates. Debt yield moves with nothing but your NOI, which is why it is the number that decides whether a loan written in 2016 refinances cleanly in 2026.
Refinancing arrives with a specific kind of disappointment. You ask for a number. The lender comes back lower. Your loan to value looked fine. Your coverage looked fine. Nobody can quite explain where the gap came from.
The gap is there because a lender does not run one test. It runs three, and it lends against whichever one produces the smallest loan.
The three tests
Moves with: cap rates
Moves with: interest rates
Moves with: nothing but your NOI
The first two are familiar. Loan to value is the loan divided by the appraisal, and it protects the lender against the building being worth less later. Debt service coverage is NOI divided by the annual payment, and it protects against the property not throwing off enough cash to make that payment.
The third is debt yield, and most owners have never calculated it.
That is the entire formula. A $5,000,000 loan against a center producing $500,000 of NOI is a 10% debt yield. The lender is asking something blunt: if we end up owning this building, what cash return are we earning on the money we put out?
Why the third test exists
Look at what each number actually depends on.
LTV depends on the appraisal, and the appraisal depends on cap rates. Cap rates compress, your LTV improves, and nothing about the building changed.
DSCR depends on the interest rate and the amortization schedule. Rates fall two points and your coverage improves. Again, nothing about the building changed.
Debt yield depends on two things: your NOI and the size of the loan. Not the appraisal. Not the rate. Not the amortization. It is the only one of the three that the market cannot flatter.
That is exactly why lenders leaned on it after 2008. They had watched loans that passed both of the other tests go bad, because both of those tests had been measuring the market rather than the property.
Which one actually binds
Run all three, and your loan is the smallest of the three.
Take a 28,000 square foot unanchored neighborhood center producing $525,000 of NOI, appraised at $7,000,000. The lender quotes 70% LTV, a 1.35 coverage minimum, and a 9.5% debt yield floor, at 7% on a 25 year amortization.
- NOI
- $525,000
- Value
- $7,000,000
- Cap rate
- 7.5%
- Payoff balance
- $4.67M
Coverage is the constraint. It caps the loan at roughly $4.59 million, about $85,000 short of the balance that has to be paid off. That $85,000 comes out of the owner's pocket at the closing table, and no amount of arguing about the appraisal changes it, because the appraisal was never the binding number.
At 7% money, that is the normal shape of a retail quote. Coverage bites first and debt yield sits above its floor with room to spare.
The part that matters
Now rewind the same building ten years.
In 2016 the loan was written at $5,880,000, at 4.25%, on a 30 year schedule. Same NOI. Coverage at origination was about 1.51. Everyone in the room relaxed.
Debt yield on that loan was 8.9%.
Coverage of 1.51 was not a statement about the center. It was a statement about 4.25% money. The loan was always thinner than it looked, and the number that would have told you so was sitting right there, unmoved, the entire time.
This is the mechanism behind the refinancing pressure in retail right now. Of the hard maturities on the 2026 CMBS calendar, roughly $27.3 billion carry debt yields of 8% or less, and that band is where refinances fail rather than close. Close to 39% of those hard maturities land in the fourth quarter.
Most of those are not bad buildings. They are buildings financed in a period when coverage was generous, by people who never ran the number that does not move.
What the floors look like
Debt yield floors move with the credit market and with the asset, but the ordering is consistent. Grocery anchored product with investment grade credit and long remaining term prices to roughly a 9% floor. Unanchored strip, stabilized, sits closer to 9.5%, and it carries lower maximum leverage and a higher coverage requirement to go with it.
That gap is worth sitting with. Two centers with identical NOI borrow different amounts based on who signed the leases and how long those leases run.
Which is the same thing the monthly scorecard is measuring.
The only input you control
Of the three numbers, two are handed to you by the market. The third has one input you can actually move.
Debt yield is NOI over loan amount. You can shrink the loan, which means bringing more equity, and sometimes that is the right answer. Or you can raise NOI.
Every dollar of durable NOI does two jobs at once. At a 7.5% cap rate it adds about $13.33 of value. Against a 9.5% debt yield floor it adds about $10.53 of borrowing capacity. The rent increase you negotiated, the recovery leak you closed, the vacancy you filled four months early: all of it shows up in both places, and it shows up on the one test that the next move in rates cannot flatter away.
Twenty minutes, before you call a lender
Do this before anyone orders an appraisal.
- Take trailing twelve month NOI, adjusted for anything that will not repeat.
- Divide it by the loan you want. That is your debt yield. Under 9% on an unanchored center, expect a conversation.
- Divide it by the floor you expect instead. That is the largest loan debt yield will permit.
- Do the same for LTV and for coverage.
- The smallest of the three is your loan.
If that number is short of your payoff, you now know two things: how much cash you need, and how many months you have to find it in rent instead.
Nobody enjoys finding a shortfall. But a shortfall found in September is a leasing problem. The same shortfall found at the closing table is a liquidity problem.