"NMHC's QAP lets me pick 20/50, 40/60, or income averaging like anywhere else — but whose income limits am I actually underwriting to out here, what utility allowance method does NMHC expect, and is that 1.15x DCR a real requirement or just scoring?"
Rent restriction formula and the utility-allowance gap
The QAP states the rent formula precisely: units "must be rent-restricted to thirty percent (30%) of the imputed income limitations for each unit, based upon HUD area median incomes and size of units," with rents imputed by bedroom count — "a unit which does not have a separate bedroom" counts as one individual, and "a unit with one (1) or more separate bedrooms" counts as 1.5 individuals per bedroom (Section V.4.A). "Gross rent must include any allowance for utilities" (Section V.4.B). If a previously-verified tenant's income rises above 140% of the applicable limit, the unit stays qualified so long as the next comparable-or-smaller vacant unit goes to a qualified low-income tenant and the original unit's rent stays restricted (Sections V.3.F.ii and V.4).
What the QAP does not state anywhere is which utility-allowance methodology NMHC expects a developer to use to size that allowance — it requires gross rent to include a utility allowance and requires "utility allowance on file" as an audit item (Section V.6), but never specifies whether that means the HUD Utility Schedule Model, a local Public Housing Authority utility allowance schedule, or actual rate estimates from the Commonwealth Utilities Corporation (CUC), the sole electric, water, and wastewater provider on Saipan, Tinian, and Rota. A developer should confirm NMHC's expected methodology directly rather than assume any one of these applies by default.
Whose AMI: HUD calculates CNMI's income limits on a different basis than the 50 states
Rents and set-asides both key off HUD's published area median incomes, and CNMI's are not built the same way a mainland state's are. HUD's own FY2026 income-limits methodology states that it uses "2024 American Community Survey (ACS) and Puerto Rico Community Survey (PRCS) median family income data as the basis of FY 2026 income limits for all areas except the U.S. Virgin Islands and the Pacific Islands (Guam, American Samoa, and the Northern Mariana Islands)" — for those Pacific Islands, HUD instead bases median family incomes "on 2020 Decennial Census data... which is the most current information available," reporting 2019 median family incomes, and then trends those incomes forward using the change in the national median family income measured by the ACS. In practice, this means CNMI's income limits move on a slower, differently-sourced cadence than a mainland state's ACS-driven limits — a real underwriting variable, not a technicality, when projecting rent growth over a hold period.
Underwriting benchmarks: DCR and reserves are scored, not a stated Minimum Threshold — and the scoring itself has an unresolved gap
NMHC's QAP does reference a debt-service-coverage figure and a reserve figure, but neither appears among the QAP's four stated Minimum Thresholds (Market Study, Site Control, Capital Needs Assessment, Developer Fee — Section III.1). Both live instead inside Criterion 15, "Overall project feasibility," a scored item: "Documentation of development costs" (2 points), "Documentation of operating costs" (2 points), "Debt Service Coverage Ratio of >1.15x" (2 points), "Operating reserves equal to 3 months of monthly operating expenses" (2 points), and "Financial Commitments in place" (7 points). A project that underwrites below 1.15x DCR or carries less than three months of reserves is not disqualified by the QAP text — it simply forgoes those points.
| Component | Points |
|---|---|
| Documentation of development costs | 2 |
| Documentation of operating costs | 2 |
| Debt Service Coverage Ratio of >1.15x | 2 |
| Operating reserves equal to 3 months of monthly operating expenses | 2 |
| Financial Commitments in place | 7 |
Itemized total: 15 points. Flag: the QAP's own summary scoring table (Section III.2) lists Criterion 15's range as "0 - 10," which does not reconcile with this itemized breakdown's total of 15 points, and the QAP text never resolves the discrepancy. This research treats the fully itemized breakdown (which the QAP labels "Criterion 15 (0 – 15 Points)" in its own header) as the operative figure, since it is the more specific of the two, but a developer or scorer should confirm directly with NMHC which number governs before relying on either for a competitive bid.
A related, smaller gap sits in Criterion 2 (set-aside generosity scoring): the QAP awards 1 point for a 20/50-or-40/60-style commitment, 2 points for a deeper 40/50-or-60/60 commitment, 3 points for a 60/50-or-80/60 commitment, and then jumps straight to 10 points for a project that sets aside 100% of units at ≤60% AMI — with no stated value for a project that commits, for example, 90% of units at ≤60% AMI. NMHC's evaluation of a project that falls in that gap is not addressed in the QAP text.
Fees that hit the pro forma directly
Beyond the $2,500 application fee, several NMHC fees flow straight into a project's sources-and-uses and should be budgeted, not treated as incidental: a good faith deposit of "five percent (5%) of the first year's federal tax credit reservation... payable at the time the executed binding agreement between the tax credit recipient and its investor is finalized," of which NMHC retains 75% as an administrative fee once IRS Form 8609 is issued (the remaining 25% is refunded, unless the project is cancelled within two years of approval, in which case NMHC keeps that portion too); a compliance monitoring fee of "up to $200 per unit for all units within each project... charged annually," adjustable by NMHC each January 1; a Qualified Contract processing fee of $150 per unit; a reallocation processing fee of $500; and, on any transfer or sale of credits to another party, a fee to NMHC equal to 50% of the proceeds received.
Where this goes wrong
- Treating NMHC's >1.15x DCR and 3-month reserve figures as binding underwriting minimums — the QAP places both inside "Overall project feasibility," a scored criterion, not among its four stated Minimum Thresholds.
- Trusting the summary scoring table's "0-10" range for Criterion 15 without checking the QAP's own itemized breakdown for that same criterion, which totals 15 points and is separately labeled "(0 – 15 Points)" — the QAP never reconciles the two numbers.
- Assuming CNMI's HUD income limits are built the same way a mainland state's are — HUD bases Pacific Island limits (CNMI, Guam, American Samoa) on 2020 Decennial Census data trended forward with national ACS growth, not on current-year, area-specific ACS surveys.
- Sizing a utility allowance without first confirming NMHC's expected methodology — the QAP requires gross rent to include a utility allowance and an on-file utility allowance for audits, but never states whether it expects the HUD Utility Schedule Model, a PHA schedule, or CUC's own rate estimates.
- Assuming the Criterion 2 set-aside point table is a smooth, interpolatable scale — it jumps from 3 points to 10 points with no stated value for a set-aside commitment that falls between those two tiers.
- Treating income averaging as a self-executing federal election in the CNMI — NMHC's own Appendix 2 requires affirmative NMHC consent, syndicator/investor sign-off, and market-study support before it will accept the election.
- HUD
- LIHTC
- State QAPs
- IRS § 42
- Housing Finance Agencies
