When the NOI in a loan file is higher than the NOI in the trailing-12 (T-12) statement it cites, the debt service coverage ratio (DSCR) changes by the gap divided by annual debt service. In one finished run, a loan package reported a trailing NOI of USD 1,312,400, the T-12 it cited totaled USD 1,029,089, and the USD 283,311 difference turned a 1.27x DSCR into 0.99x. The gap was not in income. It was in operating expenses. This page works through that run in three numbers and lists what a credit committee should ask next.
Written by the Titleman team with AI assistance (Claude). Reviewed by the Titleman review desk. The worked example is one finished Deal Review run on a Titleman QA account, read on 15 October 2026. It reviewed a sample 120-unit multifamily package with no client data and no address. Every figure below is from that run or is arithmetic on its figures. This is a method for a credit file, not an appraisal, a loan quote or lending advice.
The question
A loan file usually carries an NOI from the sponsor's offering memorandum or model. The operating statement behind it carries another. When the two differ, the question for a lender or credit team is: how big is the gap, and what does it do to DSCR? The answer decides whether the loan, as sized, still clears the coverage test.
The run reviewed a package of three documents: a memorandum, a T-12 with rent roll, and the buyer's underwriting model. The loan in the package was USD 14,040,000 at 6.25%, 65% of the guidance price, with annual debt service of USD 1,037,360.
Three numbers from the run
| Number | In the loan package | On the T-12 and rent roll | Gap |
|---|---|---|---|
| Trailing NOI | USD 1,312,400 | USD 1,029,089 | USD 283,311 (21.6% below the package figure) |
| DSCR on a USD 1,037,360 debt service | 1.27x | 0.99x | 0.27x |
| Occupancy | 96.0% stated | 91.7% (110 of 120 units, 30 September 2026 rent roll) | 4.3 points |
All three come from the run's findings. The NOI gap and the DSCR are one problem seen twice: the DSCR falls 0.27x because USD 283,311 of NOI is missing. The occupancy gap is a second, separate place where the file does not tie to the rent roll, and it is covered in its own section below.
Where the USD 283,311 comes from
Add each document's operating expenses back to its NOI and both land on the same total income, USD 1,990,955. So income agrees, and the whole gap sits in expenses.
| Line | Memorandum | T-12 | Difference |
|---|---|---|---|
| Total income (NOI plus expenses) | USD 1,990,955 | USD 1,990,955 | USD 0 |
| Operating expenses | USD 678,555 | USD 961,866 | USD 283,311 |
| NOI | USD 1,312,400 | USD 1,029,089 | USD 283,311 |
| Expense ratio | 34.1% | 48.3% | 14.2 points |
The run states this directly: the USD 283,311 is exactly the gap between the T-12's USD 961,866 of operating expenses and the memorandum's USD 678,555. The T-12 spends 42% more than the memorandum shows. The run does not say why. That is the first question for the sponsor, and the list below starts with it.
What the gap does to DSCR
DSCR is NOI divided by annual debt service, so the same loan gives two answers.
| Basis | NOI | DSCR | Debt yield |
|---|---|---|---|
| Loan package | USD 1,312,400 | 1.27x | 9.35% |
| T-12 | USD 1,029,089 | 0.99x | 7.33% |
Four readings of the table matter to a credit file:
- The package figure clears 1.25x by USD 15,700. Coverage of 1.25x on this debt service needs NOI of USD 1,296,700. The package NOI sits just above that line, so a small correction would have tipped it.
- The T-12 misses 1.25x by USD 267,611. To reach 1.25x on the old balance, T-12 NOI would have to rise 26.0%. To reach the 1.35x that the run's rate sheet of 1 September 2026 asked for at 65% leverage, it would have to rise 36.1%, to USD 1,400,436.
- At 1.0x the loan has no margin. T-12 NOI covers debt service 0.99x, which is USD 8,271 short of break-even before any reserves or capital spending.
- The loan would size smaller. At 6.25% on a 30-year amortization, the T-12 NOI supports about USD 11,142,000 at 1.25x. That is USD 2,898,000, or 20.6%, below the USD 14,040,000 in the package. At 1.35x it supports about USD 10,317,000, which is USD 3,723,000 below.
A useful rule from this run: on a loan with USD 1,037,360 of annual debt service, each USD 100,000 of NOI is worth about 0.10x of DSCR. Capitalized at the memorandum's own 6.08% rate, the USD 283,311 gap is also about USD 4.66 million of value, a figure the appraisal side of the file would care about.
The second place the file does not tie: occupancy and rent
The memorandum says the property is 96.0% occupied, with an average in-place rent of USD 1,515 against a market average of USD 1,495. The 30 September 2026 rent roll shows 110 of 120 units occupied (91.7%) and an average in-place rent of USD 1,458.23, which is below market, as the memorandum's own page 3 says. The T-12 vacancy line is 8.0% of gross potential rent for the year, so the claim of occupancy above 95% for eight quarters is not borne out by the period provided.
The T-12 income is not fully clean either. September's T-12 rent after loss to lease and vacancy was USD 163,183, which is USD 2,778 a month (USD 33,336 a year) above the rent roll's in-place rent of USD 160,405. A monthly non-revenue unit charge of USD 1,595 has no unit on the rent roll. So the T-12 is the better figure in this file, but it is still a figure to test, not a final answer.
The buyer's model adds a third number. It carries an in-place NOI of USD 1,290,000, labeled T-12, which is 25.4% above the actual T-12. The run found that this in-place NOI matches neither the memorandum nor the T-12.
What a credit committee would ask next
- What makes up the USD 283,311? Ask for a bridge from the memorandum's USD 678,555 of expenses to the T-12's USD 961,866, line by line. A reclassification, a missing line and a one-time cost have very different consequences.
- Does the T-12 tie to the ledger and the bank? The run lists the general ledger, the tax bill, the lease files and the loan documents as the items that would settle what is still open. Ask for them before sizing, not after.
- Which NOI was the loan sized on? If it was sized on the package figure, re-run the sizing on the T-12 and show both cases side by side. In this run, that is 1.27x against 0.99x.
- Does the rent roll support the occupancy and rent in the memo? Compare 110 of 120 units and USD 1,458.23 with the 96.0% and USD 1,515 stated, and ask what the sponsor's covenant assumes.
- Do the sponsor's own documents agree with each other? The model's in-place NOI, the memorandum's NOI and the T-12 gave three different numbers. The model also had a saved loan amount of USD 15,120,000 although its formula gives USD 14,040,000, and a typed exit balance of USD 13,049,000.
- What happens to coverage when the rate resets? A loan that covers 0.99x at 6.25% has no room for a higher rate.
- What could not be checked? In this run no address was given, so the property's recorded title, liens, court cases, permits, taxes and flood status, and its recorded sales, competitors and new supply, were not checked. A committee should know which parts of the file were tested and which were not.
What this does not tell you
- That the memorandum is wrong. A gap shows that two documents disagree. One of them has to be explained, and the explanation may be innocent, such as a one-time cost in the T-12.
- The right NOI. The T-12 is the better base case here because it is the operating statement, but the run still found items to review in it.
- A forecast. The DSCR figures are trailing and use one rate, 6.25%, from the package. They are not a projection of the loan's future coverage.
- A decision. The credit committee decides. A review gives it the numbers side by side and marks what was not checked.
For the full list of what the run flagged on this package, see what a Deal Review catches before you sign. For the ratio itself, see what DSCR is, and for the statement behind it, what a T-12 is and pro forma vs in-place NOI.
To see the memorandum NOI, the T-12 and the DSCR tied out on one of your own loan packages, with every figure marked verified, to review or red flag, request a demo.
Frequently asked questions
What should a lender do when the NOI in the loan file is higher than the T-12 NOI? Size and test the loan on the trailing-12 NOI, show the DSCR on both figures side by side, and ask the sponsor for a line-by-line bridge between the two. In the worked run above, the loan package reported a trailing NOI of USD 1,312,400 and the T-12 it cited totaled USD 1,029,089, a gap of USD 283,311. The same loan covered 1.27x on the package figure and 0.99x on the T-12.
How much does an NOI gap change DSCR? DSCR moves by the NOI gap divided by annual debt service. On a loan with USD 1,037,360 of annual debt service, each USD 100,000 of NOI is worth about 0.10x of coverage, so a USD 283,311 gap is worth about 0.27x. That is the difference between 1.27x and 0.99x.
Does a gap between the loan file NOI and the T-12 mean the numbers are wrong? Not by itself. A gap means two documents in the same file do not agree, and one of them has to be explained. The T-12 may include one-time costs, or the loan package may leave out costs the property really bears. The point of the check is to make the sponsor explain the difference before the loan is sized.
Which NOI should a credit committee use to size a loan? The one that ties to the operating statement and the bank and ledger records behind it, which is usually the trailing-12 NOI, adjusted only for items the sponsor can document. A marketed or pro forma NOI can be shown as a second case, not as the base case.
Written by the Titleman team with AI assistance (Claude). Reviewed by the Titleman review desk. Figures come from one finished Deal Review run of a sample 120-unit multifamily package on a Titleman QA account, read on 15 October 2026, with the 6.25% rate, the USD 14,040,000 loan and the 1.35x requirement taken from the package and the run's rate sheet of 1 September 2026. Debt yield, supported loan amounts, required NOI and the capitalized value are arithmetic on those figures. A review is not an appraisal, a loan quote or lending advice.