The measurement window in the NOI definition can decide a covenant breach before the covenant level does.
T-12 sums the last twelve months of actual NOI; T-3 annualised multiplies the last three months by four. After a tenant leaves, T-3 shows the new income at once and T-12 takes a year to catch up. On an illustrative $25.0m loan that loses a quarter of its NOI in month 9, the quarter-end DSCR is 1.35x on T-12 but 1.08x on T-3 annualised, against a 1.25x covenant: one definition passes, the other is in breach, on the same rent roll.
Worked in full in How to Read a Real Estate Loan Agreement by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →
The covenant level gets negotiated. The measurement window is buried in the definition of Net Operating Income or Debt Service Coverage Ratio, and it often matters more. A shorter window reacts faster to bad news and to good news, and it magnifies every lumpy item in the month it is booked. Some agreements test the lesser of two windows, which takes the worse of both.
| Input | Value |
|---|---|
| Loan | 25,000 |
| Interest rate, interest-only | 6.0% |
| Annual debt service | 1,500 |
| DSCR covenant, tested quarterly | 1.25x |
| Monthly NOI, months 1 to 9 | 180 |
| Monthly NOI from month 10, after a tenant vacates | 135 |
All figures are in $000. The departing tenant takes 45 a month of NOI, 25.0 per cent of the total. The covenant needs 1,875 of annual NOI, 156.25 a month. Before the vacancy the property earned 2,160 a year, a 1.44x run-rate; after it, 1,620, 1.08x.
T-12 = sum of months 1 to 12 = 9 × 180 + 3 × 135 = 2,025, DSCR 2,025 ÷ 1,500 = 1.35x
T-6 annualised = (3 × 180 + 3 × 135) × 2 = 1,890, DSCR 1.26x
T-3 annualised = 3 × 135 × 4 = 1,620, DSCR 1.08x
T-1 annualised = 135 × 12 = 1,620, DSCR 1.08x
Excel, with monthly NOI in B2:B13 and debt service in D1: =SUM(B2:B13)/D1 for T-12 and =SUM(B11:B13)*4/D1 for T-3 annualised.
T-12 and T-6 pass; T-3 and T-1 breach. A "lesser of T-12 and T-3 annualised" definition gives 1,620 and 1.08x, a breach.
| Quarter end, month | T-12 NOI | T-12 DSCR | T-3 annualised NOI | T-3 DSCR |
|---|---|---|---|---|
| 12 | 2,025 | 1.35 | 1,620 | 1.08 |
| 15 | 1,890 | 1.26 | 1,620 | 1.08 |
| 18 | 1,755 | 1.17 | 1,620 | 1.08 |
| 21 | 1,620 | 1.08 | 1,620 | 1.08 |
The T-12 borrower trips the covenant two quarters later, at month 18, and the ratio drifts down one step each quarter as a good month drops out of the window. Those six months are the real value of a T-12 definition to a sponsor: time to re-let the space before the cash trap or the default bites. To the lender, the same six months are a period in which the covenant reports a ratio the property no longer earns.
| Situation | T-12 NOI | T-12 DSCR | T-3 annualised NOI | T-3 DSCR |
|---|---|---|---|---|
| Lease-up: NOI 135 a month, a new tenant adds 50 from month 10 | 1,770 | 1.18 | 2,220 | 1.48 |
| No vacancy, 120 insurance premium booked in month 11 | 2,040 | 1.36 | 1,680 | 1.12 |
| Vacancy and the same premium | 1,905 | 1.27 | 1,140 | 0.76 |
In a lease-up, the short window is the borrower's friend: the new run-rate of 185 a month shows at once, at 1.48x, while T-12 still carries nine weak months and fails at 1.18x. That is why sponsors of value-add deals ask for annualised trailing three months, and why some lenders accept it only for an initial period.
The insurance row is the trap. A property with nothing wrong with it breaches on T-3 annualised because a single annual premium of 120 was booked in one month, and annualising multiplies it by four: 480 off the test NOI, against 120 off T-12. A well-drafted definition spreads annual items evenly, 10 a month here, or annualises income only and takes expenses on a trailing twelve-month or budget basis. Read which.
The same property, the same quarter, gives 1.35x or 1.08x depending on one phrase. When negotiating, a sponsor with leasing risk wants T-12; a sponsor leasing up wants T-3 annualised for the lease-up period; both want annual expenses spread evenly. The covenant models for this book are in the free workbook for this book.
Net operating income for the last three months multiplied by four. On the illustrative property, three months at 135 ($000) annualise to 1,620, a 1.08x DSCR on 1,500 of debt service, while the trailing twelve months still show 2,025 and 1.35x because nine months predate the vacancy.
It depends on the direction of income. After a tenant loss T-12 delays the breach by two quarters in the illustrative case. During a lease-up T-3 is better: a new tenant adding 50 a month lifts T-3 to 2,220 and 1.48x while T-12 shows 1,770 and fails at 1.18x.
Annualising multiplies them. A 120 insurance premium booked in one month takes 480 off T-3 annualised NOI but only 120 off T-12, so a property with no vacancy can test at 1.12x against a 1.25x covenant. Definitions that spread annual items evenly, 10 a month here, avoid it.
This article is one calculation from How to Read a Real Estate Loan Agreement. The book takes the same case from first principles to the decision, chapter by chapter, and every figure it prints is a live formula in the free companion workbooks.
Get the book on Amazon →Free companion files
Also on Amazon UK · Amazon Germany · Amazon France · Amazon Canada
Reading guide: real estate investing, finance and fund management → · All 1682 articles →
If this book helped, or didn’t, a few lines on Amazon are worth more than they look: they are what the next reader goes on. Write a review. The workbook stays free either way.