"What does this actually cost, and does NH Housing's own dollar cap let it pencil?"
Four passes, and the locked number is a dollar cap, not a limit table
Cost estimation in a New Hampshire LIHTC deal follows the same four-pass structure every LIHTC state uses, but what gets locked, and against what, differs from both California and Texas. There is no county-by-county basis-limit table here — the number a New Hampshire application is tested against is a single, statewide dollar figure computed from the project's own unit count and bedroom mix.
| Pass | Who produces it / NH Housing document | Precision | Timing |
|---|---|---|---|
| Napkin / screening estimate | Developer's own model, in Excel | ±30–40% | Minutes to hours |
| Concept estimate | Architect's schematic design (SD) set, priced informally by a GC or estimator | ±15–20% | Weeks after schematic design |
| Preliminary and final application budget | Developer plus LIHTC consultant, entered in NH Housing's financing application and tested against the TDC Weighted Average | Locked for the funding round | Preliminary application / Letter of Intent (minimum requirements posted by May 1 each year), then the final application |
| GMP or final cost certification | GC or CM pricing under HFA 111 procurement rules; certified by a CPA at 8609 issuance | ±3–5%, then audited | 10% test within 12 months of the Carryover Allocation Agreement; final cost certification at close-out |
New Hampshire's readiness threshold makes the application-stage number more disciplined than a pure napkin figure. To be considered complete, an application must already include a detailed scope of work with cost estimates current within six months, plus schematic design plans and specifications — so the number NH Housing tests is closer to a concept estimate than a back-of-envelope one, even at the earliest competitive round.
Who is actually in the room mirrors California's list minus one role: developer project manager, architect, GC or CM preconstruction estimator, and LIHTC consultant. A labor compliance consultant only enters the picture if a federal program in the capital stack triggers Davis-Bacon — New Hampshire itself has no state prevailing-wage apparatus to staff for.
One structural question the QAP text does not resolve: whether the TDC Weighted Average ceiling is re-tested at placed-in-service, the way CTCAC re-tests California's high-cost limit against a revised, placed-in-service-year figure. The closeout package visible in the QAP (Appendix C) calls for a CPA-certified final cost certification checked for consistency with the application's Sources and Uses — that reads as a consistency check, not an explicit second pass against the $397,000/$425,000 ceiling, but the QAP does not say so outright. Confirm the actual mechanics with NH Housing's Construction Services Manager before assuming California's pattern carries over.
The TDC Weighted Average is the spine of the phase
Where CTCAC tests eligible basis against a threshold basis limit table keyed to county and bedroom count, NH Housing tests total development cost against a single blended formula, evaluated against three flat dollar ceilings that apply the same way whether the project sits in Manchester or Colebrook.
| Ceiling | Applies to | Amount |
|---|---|---|
| TDC Weighted Average, standard | All projects not qualifying for the high-cost category | $397,000 |
| TDC Weighted Average, high-cost | Projects meeting at least one of seven enumerated high-cost criteria | $425,000 |
| Total development cost per unit, unweighted | Every project, regardless of high-cost status | $464,000 |
| Investment limit (NH Housing capital subsidy + LIHTC equity) | Per unit, all projects (evaluated case-by-case below 30% AMI) | $315,000 |
TDC Weighted Average = ([2 × TDC/unit] + [TDC/bedroom]) ÷ 3. For mixed-use projects, only the residential portion, including common areas, is used. Applications that miss the applicable ceiling are rejected absent an approved HFA 117 waiver.
There is no regional variation in the base formula. Unlike CTCAC's per-county, per-bedroom threshold basis limit table, New Hampshire runs one statewide number for every county from Coos to Rockingham. The QAP's only nod to geography is the high-cost county carve-out below — a binary yes-or-no by county, not a continuous regional coefficient like Terner Center's California figures.
| Criterion | Documentation |
|---|---|
| Adaptive reuse project | Thorough supporting documentation submitted with the application |
| Uses Historic Rehabilitation Tax Credits | Same |
| Site meets the EPA's brownfield definition | Same |
| Required to comply with the Build America, Buy America Act (BABA) | Same — see the labor package section below |
| Designed and built to achieve Passive House certification | Same — also earns 2 scoring points under HFA 109.07 |
| Faces an inadequate bidder pool, impeding a competitive bid process | Same |
| Located in Grafton, Carroll, or Coos County | Same |
Meeting one of the seven raises the ceiling from $397,000 to $425,000 — a $28,000-per-unit swing on a formula NH Housing applies unit by unit. The application must submit thorough evidence, and NH Housing decides eligibility on review, not the developer at budgeting time.
The BABA criterion is the one most likely to appear or disappear without the developer choosing it. Applicants may not request a specific NH Housing capital subsidy source, and the QAP instructs applicants to assume the most restrictive one — typically HOME, which carries Build America, Buy America Act compliance. If BABA turns out not to be required, the project drops out of the high-cost category (back to the $397,000 ceiling) unless it qualifies on another ground; NH Housing has discretion to grant an exception to that reclassification.
The investment limit is evaluated once, at application, using projected equity pricing and NH Housing subsidy — a later equity-pricing swing does not by itself trigger a re-evaluation. That is real, stated protection against LIHTC equity market volatility between application and closing.
Cost reasonableness is tested twice: a hard cap, then a moving target
New Hampshire runs two separate cost tests out of two separate parts of the QAP, and they behave differently. One is a fixed threshold that either passes or fails an application outright. The other is a competitive, relative penalty that a developer cannot fully price until the round's other applications are in.
| Test | Where it lives | Mechanism | Consequence |
|---|---|---|---|
| Hard ceiling | HFA 109.06.G Cost Reasonableness / HFA 109.04.C | TDC Weighted Average (or unweighted per-unit cost) exceeds the applicable $397,000 / $425,000 / $464,000 figure | Application rejected, absent an approved HFA 117 waiver |
| Relative penalty | HFA 109.07 Scoring, Item 11 (Project Cost) | Project's TDC Weighted Average compared to the average of all TDC Weighted Averages submitted in the current round | Negative scoring points, on a sliding scale |
| Over the round average by... | Negative points |
|---|---|
| 5% (meets no high-cost criteria) | -2 |
| 10% (meets no high-cost criteria) | -4 |
| 25% (meets no high-cost criteria) | -6 |
| 10% (meets one high-cost criterion) | -2 |
| 10% (meets two or more high-cost criteria) | 0 |
Preservation/recapitalization projects are excluded from this scoring category and from the round-average calculation.
The second test is the harder one to underwrite to, because it is relative. A developer can sit comfortably under the $397,000 hard ceiling and still take a scoring hit if every other application in that round happens to come in lean — the benchmark is this round's own peer group, not a fixed figure published in advance. That is a genuinely different risk shape than California's high-cost test, where 30% over a published, county-specific limit is knowable the moment the year's limit memo is out.
A waiver against the hard ceiling runs through the HFA 117 Waiver Rules process — a real but discretionary path, not a formula. Treat it as an exception mechanism, not a planning assumption.
The labor package is almost entirely a federal question, because New Hampshire has none of its own
New Hampshire has no state prevailing-wage law. RSA Chapter 280, "Minimum Wages of Employees in Public Works," New Hampshire's own little-Davis-Bacon statute, was repealed in its entirety by Laws 1985, 117:1, effective July 13, 1985, and nothing has replaced it. Compare that to California's Labor Code §1720 machinery — the broad public-funds trigger, the six subcategories that catch donated land and below-market loans, the strict judicial construction of the narrow exemptions. In New Hampshire, that entire layer of cost and compliance risk simply does not exist at the state level.
| Program | Trigger | Citation |
|---|---|---|
| HOME | 12 or more HOME-assisted units in the construction contract | 24 CFR §92.354 |
| CDBG | Rehabilitation of residential property with 8 or more units | 24 CFR §570.603 |
| Project-based Section 8 | New construction or substantial rehabilitation at 9 or more assisted units | HUD labor-standards guidance — confirm against HUD Handbook 1344.1 for a specific deal |
| Public Housing (1937 Act) | No unit threshold | HUD labor-standards guidance — same caveat |
| NAHASDA | No unit threshold; $2,000 contract threshold | HUD labor-standards guidance — same caveat |
The HOME and CDBG thresholds are read directly from the current Code of Federal Regulations. The Section 8, Public Housing and NAHASDA rows are standard HUD program thresholds carried from general HUD guidance; verify the exact figure against HUD Handbook 1344.1 before relying on it for a specific deal.
BABA is a second, separate federal layer, and it is a materials rule, not a wage rule. The Build America, Buy America Act requires domestic sourcing of iron, steel, manufactured products and construction materials on projects funded with covered federal financial assistance — it does not set a wage floor the way Davis-Bacon does. It reaches a New Hampshire deal the same way Davis-Bacon does: through whichever HOME, HTF or other federal capital-subsidy dollars NH Housing puts into the stack, which is exactly why the QAP folds BABA compliance into its high-cost criteria list rather than treating it as a labor-standards item.
HUD Section 3 — economic-opportunity and hiring-preference requirements for residents of HUD-assisted housing and low-income area residents — shows up in NH Housing's own Underwriting Standards as a required competency area for the owner's representative, alongside Davis-Bacon and BABA. It is a contracting and hiring-preference obligation for HUD-funded projects, not a wage-rate rule, but it is a real compliance and administrative cost that a purely materials-and-wages view of "the labor package" will miss.
The practical upshot: a New Hampshire deal financed with LIHTC equity, a conventional first mortgage, and NH Housing's own state-funded Affordable Housing Fund — no HOME, HTF, CDBG or project-based Section 8 in the stack — can plausibly clear this entire phase with zero Davis-Bacon exposure. That is close to structurally impossible on a Southern California 9% deal, where Terner Center's own data puts prevailing-wage prevalence at 52.6% of 2020–2023 California LIHTC awards. Whether a specific New Hampshire deal clears it depends entirely on which capital-subsidy dollars end up in the stack, which is exactly why the QAP tells applicants to assume the worst case until financing is locked.
The structural caps that bind the budget, and the order to run this in
| Cost of the Work | Combined OH&P + general conditions cap |
|---|---|
| $3,000,000 or more | 14% |
| $1,300,000 to $3,000,000 | 16% |
| $1,300,000 or less | 20% |
Excludes performance/payment bonds and building permits. Applies under General Contracting, Construction Management, or Design-Build delivery alike.
| Unit tier | General occupancy | Age-restricted |
|---|---|---|
| First 10 units | $23,760/unit | $21,600/unit |
| Next 30 units | $19,440/unit | $17,280/unit |
| Next 60 units | $16,200/unit | $14,580/unit |
| Above 100 units | $7,020/unit | $7,020/unit |
Capped overall at 15% of total development cost. Acquisition-only deals (rehab/new construction under 50% of TDC, not counting land) use a flat $6,300/unit instead. Tax-exempt bond 4% deals may take up to 15% of TDC, but any amount above what the per-unit schedule would otherwise allow must be loaned back to the project as a developer fee loan.
Where there is an identity of interest between the builder and the developer, a second cap stacks on top of the schedule above: the combined developer fee plus builder overhead, profit and general conditions may not exceed 18% of total development cost on projects with construction costs above $2.5 million, or 20% at or below $2.5 million. Separately, HFA 111.12 does not permit an identity-of-interest Construction Manager or General Contractor to participate by default — it requires a written request and case-by-case approval from NH Housing's Executive Director, a self-contained gate inside HFA 111 itself, not the HFA 117 Waiver Rules process used for the QAP's own cost ceilings.
Two different contingency numbers get conflated in practice. The developer's own construction contingency is 5% of the construction amount for new construction and 10% for rehabilitation, under the Underwriting Standards. The Construction Manager's own GMP contingency, a separate figure inside the CM's fee structure, is capped at 3.5% and may not be used to supplement general conditions or self-performed work. They live in two different documents and answer two different questions.
Bids over budget trigger a defined process, not a quiet adjustment. A GMP or bid within 10% of the application budget can be negotiated with the low bidder subject to design-team and NH Housing approval. Anything over 10% forces a meeting among the developer, architect, low bidder and NH Housing, and NH Housing can compel a redesign and re-bid at its discretion — a real schedule and cost consequence for a budget that was priced thin at application.
Modular construction gets a defined procurement rule and nothing else. Projects using modular construction must obtain three competing bids from modular contractors, the same competitive-bidding discipline applied to any other subcontractor. Neither the QAP nor the Underwriting Standards publish a modular cost-savings benchmark — the same gap California's own research corpus flags, and for the same reason: no controlling state-specific study appears to exist yet.
Two more real soft-cost lines are easy to underbudget: the LIHTC Allocation Fee runs 7% of the total LIHTC allocation amount (1% due with the application, 6% due with the final allocation package, before IRS Form 8609 is issued), and the LIHTC Upfront Monitoring Fee is a flat per-unit charge — $675 per unit for 4% LIHTC projects, $850 per unit for 9% LIHTC projects — due with the final allocation package, and does not vary by the LURA's affordability-period length.
| Step | Action | Why |
|---|---|---|
| 1 | Fix the unit mix and get at least a current (within 6 months), SD-level cost estimate | Both are readiness-threshold requirements before an application is even considered complete |
| 2 | Run the TDC Weighted Average formula and honestly test each of the seven high-cost criteria before assuming any of them apply | A $28,000-per-unit ceiling swing turns on documentation NH Housing reviews after the fact, not on the developer's own judgment at budgeting time |
| 3 | Identify the likely capital stack and price Davis-Bacon/BABA exposure at the most restrictive assumption (HOME) until financing is locked | The QAP itself instructs this; pricing zero labor-standards exposure because the subsidy source isn't picked yet inverts the QAP's own guidance |
| 4 | Size the developer fee off the correct per-unit schedule (general occupancy vs. age-restricted) and check the identity-of-interest 18%/20% cap separately | These are two different caps that bind independently of each other |
| 5 | Underwrite to the round-average scoring penalty, not just the hard ceiling | Being under $397,000 does not protect against losing points to a round where the competition costs less |
Where this goes wrong
- Treating the $397,000 TDC Weighted Average as a target to approach rather than a hard ceiling. Going even modestly over it without an approved HFA 117 waiver means automatic rejection under HFA 109.06.G, not a negotiation with staff.
- Underwriting only to the fixed dollar ceiling and ignoring the separate scoring penalty. A project safely under $397,000 per the weighted formula can still lose 2 to 6 points if the round's other applicants come in leaner — the benchmark is this round's own average, not a number published in advance.
- Assuming a high-cost criterion applies without documenting it. The QAP requires thorough evidence for each of the seven criteria, reviewed and approved by NH Housing after submission — budgeting to the $425,000 ceiling because a site 'probably' counts as a brownfield or bidder pool is 'probably' inadequate is not the same as having it approved.
- Pricing zero Davis-Bacon or BABA exposure because the NH Housing capital subsidy source hasn't been selected yet. The QAP explicitly instructs applicants to assume the most restrictive source (typically HOME) until financing commitment — the opposite assumption inverts the QAP's own guidance and can leave a real cost surprise sitting in the stack.
- Importing a California-style prevailing-wage cost premium into a New Hampshire pro forma. New Hampshire repealed its own public-works wage law (RSA 280) in 1985; absent a federal Davis-Bacon trigger from HOME, CDBG, project-based Section 8, Public Housing, or NAHASDA funding, there is no state-level wage floor to price at all.
- Signing a construction contract with an identity-of-interest general contractor or construction manager without NH Housing's prior approval. HFA 111.12 requires a written request and case-by-case approval by NH Housing's Executive Director before an affiliated GC or CM may participate at all; discovering that after running a full procurement process with the developer's own affiliated GC wastes the cycle.
- Conflating the Construction Manager's own 3.5% GMP contingency cap with the developer's overall construction contingency of 5% (new construction) or 10% (rehabilitation). They are two different numbers in two different documents, and using one where the other belongs misstates the available cushion.
- Treating a GMP that comes in 10%+ over the application budget as routine value engineering. HFA 111.11(b) requires the developer, architect, low bidder, and NH Housing to meet, and NH Housing can compel a redesign and re-bid — a real schedule and cost consequence, not a formality.
- Sizing the developer fee off the wrong per-unit tier. General-occupancy and age-restricted projects run different per-unit schedules, and an identity-of-interest deal with construction costs above $2.5 million caps the combined developer fee plus builder OH&P at 18% of total development cost (20% below $2.5 million) on top of that schedule.
- Building a modular cost case on manufacturer marketing claims. New Hampshire's rules specify a bidding procedure for modular construction (three competing modular-contractor bids) but publish no modular cost-savings benchmark — the same gap California's own research corpus flags.
- Leaving the LIHTC allocation fee and the per-unit upfront monitoring fee out of the soft-cost budget. At 7% of the total LIHTC allocation (1% at application, 6% at final allocation) plus a flat $675 (4% LIHTC) or $850 (9% LIHTC) per unit, these are real, sizable NH Housing fees, not rounding errors.
- Assuming the TDC Weighted Average is re-tested against the same ceiling at placed-in-service the way California's high-cost test is. The QAP does not clearly state a placed-in-service re-test against the $397,000/$425,000 figures — confirm the actual closeout mechanics with NH Housing's Construction Services Manager rather than assuming CTCAC's pattern carries over.
- HUD
- LIHTC
- State QAPs
- IRS § 42
- Housing Finance Agencies
