Case Study · Arizona · Multifamily · Agency / Conventional
Multifamily Feasibility Study, Arizona — An Agency / Conventional Worked Case
This is how our independent feasibility study company and apartment feasibility consultant team analyzed a new market-rate community underwritten to a conventional bank construction loan and a Fannie Mae/Freddie Mac agency permanent takeout, from primary-market absorption and capture through the debt-service coverage the takeout must document. It is a representative, anonymized worked example of the methodology — not a specific client deal — set at a roughly 220-unit garden community in a high-growth suburban submarket of metro Phoenix.
A ground-up garden community on the Valley's growth edge.
A merchant developer came to our feasibility study company with a ground-up apartment project and a two-part capital plan: a conventional bank construction loan to build and lease, then a Fannie Mae DUS or Freddie Mac Optigo agency loan to take out the construction debt at stabilization. Both lenders needed the projected cash flow independently tested before they would commit. The subject is a roughly 10-acre site on a suburban arterial in a fast-growing outer-ring submarket of metro Phoenix, programmed for about 220 class-A garden and wrap units averaging near 925 square feet, with the amenity package new renters in the Valley now expect.
Because market-rate apartments are an investment property rather than an owner-occupied business, this deal does not route through SBA at all — both the 7(a) and 504 programs require owner-occupancy, so standard multifamily falls to the agency, HUD-FHA, life-company, CMBS, and conventional bank channels.13 And in agency and conventional lending the deliverable reviewers require is a market study — a demand, supply, comparable, and absorption analysis that informs the lender's own underwriting — with the feasibility study adding the site-specific financial model and coverage conclusion around it.11 Our scope was that combined work: the independent absorption, capture, effective-rent, and debt-service analysis that supports both the construction credit and the agency takeout.
Representative and anonymized. Every figure below is illustrative of a typical engagement of this type; the site, submarket, and parties are composited, not a real named borrower, address, or completed transaction.
Absorption, capture, and fair share.
The demand read starts with income- and size-qualified renter households in a defined primary market area, not a rent applied to a population count. Metro Phoenix added more than 60,000 apartment units since January 2022, so the only defensible question is whether qualified demand at the subject's rent point clears the units delivering into its lease-up window.
Absorption is derived across four layers. Demand drivers — employment, household formation, in-migration, and income by tier — establish annual rental demand at the subject's rent point; the supply pipeline sets the competitive constraint, counting the units delivering during lease-up rather than a standing snapshot; capture-rate analysis allocates that demand across the subject and its competitors; and the resulting monthly lease-up schedule sizes the interest reserve.11 The subject's capture rate is its unit count divided by income- and size-qualified annual demand; its fair share is its unit count divided by competitive units in the primary market area. Here the read is supported: metro Phoenix recorded trailing-twelve-month net absorption of roughly 21,491 units, a market record that finally outpaced deliveries for the first time since early 2021, so demand is present — but so is competing supply.3 We modeled a stabilized lease-up pace near 18 units per month, inside the 12-to-25-unit band typical of garden and mid-rise product, for a roughly twelve-month absorption to stabilized occupancy.11
| Demand driver | Basis | Supported figure |
|---|---|---|
| Primary market area renter demand | Qualified renter households at the subject's rent point, growing with in-migration1 | Rising qualified base |
| Net absorption (metro anchor) | ~21,491 units TTM, a record, absorption > deliveries3 | Demand present |
| Subject capture rate | 220 units ÷ income- and size-qualified annual demand | Defensible, single-digit |
| Fair share vs. competitive set | 220 units ÷ competing PMA units in lease-up | Below proportional draw |
| Absorption pace | ~18 units/mo (12–25 band for garden product)11 | ≈ 12-mo lease-up |
Capture and absorption logic grounded in NCHMA demand methodology and current Phoenix absorption data; see sources 3 and 11. Figures are illustrative of the engagement type.
The pipeline delivering into lease-up.
The correct competitive denominator is not today's standing stock — it is the units leasing alongside the subject. Metro Phoenix carried roughly 16,399 units under construction as of Q1 2026, down about 30 percent year over year, so the pipeline is thinning, but concessions are still clearing new lease-ups across the Valley.
| Comp | Vintage / status | Units | Asking rent | Read |
|---|---|---|---|---|
| Comp A | Delivered 2024, leasing | 288 | ~$1,780 | ~8 weeks free; direct new-supply rival |
| Comp B | Delivering 2026 | 246 | ~$1,860 (pro forma) | Overlaps subject lease-up window |
| Comp C | Stabilized 2019 | 312 | ~$1,585 | Older; rent-gap floor, weak amenities |
| Comp D | Stabilized 2021 | 264 | ~$1,690 | Closest amenity match, no new lease-up |
| Comp E | Planned / permitted | ~300 | n/a | Scanned, not yet vertical; timing risk |
Competitive set surveyed and field-verified for the engagement; anonymized. Announced and permitted supply was scanned, not just the standing set, consistent with NCHMA field-verification practice. Asking rents illustrative; the model uses effective, not asking, rents.
Two of the five comps are new-supply rivals: one delivered in 2024 and still leasing on roughly eight weeks free, and one delivering in 2026 that overlaps the subject's lease-up window directly. That is exactly the denominator a snapshot misses. A rigorous study does not stop at the standing set; it scans announced and permitted supply — Comp E here — so the capture forecast is not quietly overstated by units the trailing data cannot yet see.11 The offsetting read is that the pipeline is genuinely thinning: Phoenix starts have fallen sharply, roughly 16,399 units remained under construction against that record absorption, and Northmarq counted only about 1,800 units delivered in Q1 — so the subject leases into a decelerating supply curve, not the peak of the wave.45 The model credits the subject with a below-proportional fair share during lease-up and prices to effective, concession-inclusive rents rather than the posted asks above.
Arizona macro: digesting a record cycle, not deteriorating.
The state backdrop is a demand tailwind tempered by near-term oversupply. Metro Phoenix is among the most oversupplied U.S. apartment markets after the 2022–2024 building boom, yet it is also the site of the transformative TSMC megafab and one of the nation's largest data-center clusters, and its supply is now inflecting toward balance.
| Reading | Metro Phoenix | Tucson | Signal for the subject |
|---|---|---|---|
| Apartment vacancy | Digesting11.5–12.5% (Q1 2026) | Balanced8.75%, improving | Price to soft occupancy, not 2021 comps |
| YoY rent | Negative−2.3% to −4.1% | Positive~+$12 QoQ | Model effective rent, expect burn-off |
| Deliveries vs. absorption | InflectingAbsorption > deliveries, TTM record | Balanced | Supply curve decelerating into lease-up |
| Under construction | Thinning16,399 units, −30% YoY | Light | Fewer 2027 rivals for the exit |
| Structural demand anchor | StrongTSMC $165B; data centers #2 in N.A. | University / defense | Employment base for renter demand |
Readings compiled from sources 2–7 and 15 below. Metro reads are Q1–mid 2026.
The Valley is the signature Arizona oversupply story, alongside Austin, but it is digesting rather than deteriorating. Yardi Matrix put Phoenix asking rents down 4.1 percent year over year through November 2025 with stabilized occupancy at 93.4 percent, after more than 60,000 units were added since January 2022; CoStar showed 11.5 percent vacancy and −2.3 percent rent growth, and a Kidder Mathews cut reported 11.8 percent vacancy at a $1,535 average asking rent.234 The inflection is real: trailing absorption of roughly 21,491 units finally outpaced deliveries, and the under-construction count is down about 30 percent year over year, so the read is toward balanced by 2027, not toward a deeper glut. Nationally, apartment completions crested in 2024 at the most since 1986 before falling sharply, and Phoenix is riding that same supply curve down.8 Underwriting 2021-era rent growth here is the classic Arizona failure mode; the model instead prices to effective rents and grades the recovery. Tucson, the state's balanced counterpoint, tightened to about 8.75 percent vacancy with rents rising over the same window — a reminder that a single statewide Arizona rent assumption is indefensible.6
The demand base underneath is structural, which is what separates Phoenix from a pure overbuild. TSMC has pledged $165 billion across its Arizona fabs, the largest foreign direct investment in U.S. history, and metro Phoenix ranks second in North America for planned data-center development, layered over sustained in-migration — the Phoenix–Mesa–Chandler MSA reached 5,186,958 and Maricopa County 4,673,096 as of July 1, 2024.151 Two Arizona-specific factors also cut in the subject's favor rather than against it: the state has no general Certificate of Need regime, so housing supply is market-set rather than permit-gated, and the subject sits inside an established municipal service area rather than the water-gated fringe — which the next section takes up directly.
Why the submarket captures the renter.
Household formation, employment access, and an Assured Water Supply designation all point the same direction, and the site geometry converts that demand into leases.
The primary market area carries a renter base growing with in-migration and household formation, anchored by the Southeast Valley's semiconductor and data-center employment corridor — the exact high-wage job growth that supports class-A rents and a durable renter pool. Trailing Census counts understate the captive base in a fast-growing outer-ring submarket, a common distortion a careful study corrects for rather than extrapolates.1 Because for-sale housing affordability has eroded across the Valley, the renter-by-necessity cohort is deep, supporting absorption even as concessions clear the newest lease-ups.
One Arizona-specific gate decides where you can build at all: water. Under the 1980 Groundwater Management Act, projects in the Phoenix Active Management Area must demonstrate a 100-year Assured Water Supply, and after the Arizona Department of Water Resources found a roughly 4.86 million acre-foot 100-year groundwater shortfall in the Phoenix AMA in June 2023, it paused new groundwater-reliant determinations — constraining fringe subdivisions in Buckeye, Queen Creek, and Pinal County.14 The subject clears that gate by design: it is served by an established municipal provider carrying a Designation of Assured Water Supply, so the study verifies the specific provider's status rather than assuming it. A feasibility consultant who skips that verification can underwrite a Phoenix-area site that cannot, in fact, be served — a uniquely Arizona way for a study to fail review.
Bank construction into an agency permanent takeout.
Total development cost lands at $50.6 million. The structure is a two-stage capital plan common to ground-up multifamily: a conventional bank construction loan at 65 percent loan-to-cost to build and lease, then a Fannie/Freddie agency loan to retire the construction debt once the asset stabilizes.
| Cost component | Amount |
|---|---|
| Land (~10 acres) | $5.30M |
| Hard construction costs | $35.60M |
| Site work, utilities & offsites | $2.70M |
| Amenities, FF&E & landscaping | $1.60M |
| A&E, permits & impact fees | $2.40M |
| Financing costs & construction-period interest | $1.80M |
| Developer fee & contingency | $1.20M |
| Total development cost | $50.60M |
| Item | Figure |
|---|---|
| Senior construction loan (65% LTC) | $32.89M |
| Sponsor & JV equity (35%) | $17.71M |
| Construction term / structure | 36-month term, interest-only, funded interest reserve |
| Illustrative construction rate | ~8.25% (SOFR + 3.25%) |
| Agency permanent takeout (Fannie/Freddie) | ~$30.5M |
| Permanent terms | ~6.00% fixed, 30-year amortization |
| Permanent sizing constraint | Lesser of ~65% LTV and ~1.25x DSCR — coverage-bound |
Agency conventions per Fannie DUS / Freddie Optigo; market-rate apartments are SBA-ineligible on owner-occupancy grounds and route here instead. See sources 12 and 13.
The takeout is where this deal earns its study. The agency loan is sized to the lesser of a ~65 percent LTV ceiling and a ~1.25x coverage minimum on stabilized net operating income, and in a higher-rate environment coverage binds first.12 Sized to a 1.30x stabilized coverage, the permanent loan comes in near $30.5 million — only about 56 percent of the ~$54.3 million stabilized value, well inside the 65 percent LTV ceiling, at a debt yield near 9.3 percent. That is the DSCR-constrained takeout the whole capital plan turns on: because coverage — not leverage — governs, the agency loan retires most, but not all, of the $32.89 million construction balance, leaving a roughly $2.4 million paydown the sponsor funds at conversion. Discovering that gap at conversion is the dominant execution risk on newly built lease-up assets; the feasibility study sizes it up front, which is precisely why the absorption and effective-rent conclusions matter more than headline value. The agencies entered 2026 with roughly $176 billion of combined purchase caps, positioning them as the refinance backstop this takeout relies on.10
Feasible and bankable, on coverage the takeout can document.
The stabilized model builds effective gross income from concession-inclusive rents, nets a full-load operating expense to a ~60 percent margin, and carries coverage from a sub-1.0 lease-up year to a 1.30x stabilized DSCR that clears the agency floor.
| Line | Basis | Amount |
|---|---|---|
| Gross potential rent | 220 units × ~$1,825/mo effective4 | $4.818M |
| Other income | Parking, RUBS reimbursement, fees | $0.282M |
| Gross potential income | Rent + other income | $5.100M |
| Vacancy, concession & credit loss | ~6.9% economic, concessions burning off7 | ($0.350M) |
| Effective gross income (EGI) | Collected income | $4.750M |
| Operating expenses | Taxes, insurance, payroll, R&M, mgmt, utilities (~40% of EGI)7 | ($1.900M) |
| Net operating income (NOI) | ~60% NOI margin | $2.850M |
Effective rent reflects a premium to the ~$1,535 metro-average asking rent for new class-A product, net of concessions; operating expenses run 38–55% of revenue for apartments. See sources 4 and 7. Stabilized value at a ~5.25% cap ≈ $54.3M.
| Year | Stage | NOI | Debt-service basis | DSCR |
|---|---|---|---|---|
| Year 1 | Lease-up (interest reserve) | ~$1.65M | Construction I/O, reserve-funded | 0.75 |
| Year 2 | Post-conversion, absorbing | ~$2.52M | Perm, full amortizing ~$2.19M | 1.15 |
| Year 3 | Stabilized | ~$2.85M | Perm, full amortizing ~$2.19M | 1.30 |
DSCR computed as NOI divided by the period debt-service obligation. See source 12 for the ~1.25x agency coverage convention.
The stabilized 1.30x coverage is the figure the agency documents, and it clears the ~1.25x floor with headroom while sitting at only ~56 percent LTV.12 By Year 2 the project already covers fully amortizing debt service at 1.15x. The Year 1 figure of 0.75x is intentionally below 1.0 — it is the lease-up year — which is exactly why the construction facility carries a funded interest reserve: the reserve carries the interest-only ramp, and permanent, fully amortizing coverage is measured only once the community reaches stabilized occupancy. Modeling posted asking rents instead of effective rents, or best-in-class absorption on day one, is among the most common ways apartment pro formas fail review; the ramp here is deliberately graded and priced to concession-inclusive rents.7
On the equity side, the roughly $20.1 million invested — the $17.71 million construction injection plus the ~$2.4 million conversion paydown — earns a modest stabilized cash-on-cash near 3 percent that builds as effective rents recover off their trough, consistent with a Valley digesting toward balance by 2027 rather than a return to 2021 rent growth.3 The value case rests on a positive development spread — a stabilized yield-on-cost near 5.6 percent against a ~5.25 percent exit cap16 — realized through amortization of the agency loan and a disciplined exit near a stabilized cap, with per-unit trade pricing still running below replacement cost across the metro.9 Blended over the hold, the result is an illustrative levered equity IRR in the mid-teens, roughly 15 percent — a return that is real but sensitive: continued oversupply, deeper or longer concessions, or an exit-cap expansion would pull it toward high single digits, which is why the study stresses those variables rather than assuming them away.
Verdict: financially feasible and bankable, with quantified Phoenix supply risk. On independently derived absorption, a stabilized 1.30x DSCR, and a ~15% illustrative levered IRR, the projections support the construction credit and the agency takeout.
Independent absorption, capture, effective rent, and DSCR stress.
The engagement was scoped the way an agency and a construction credit committee read it, and built to NCHMA Model Content Standards, the recognized methodology framework for multifamily market studies.11 As an independent feasibility consultant, our role is to test the sponsor's projection against the market, not to restate it — the value of the deliverable is precisely that it carries no stake in the outcome. We defined a primary market area by drive time and competitive geography, field-verified the comparable set, derived capture from income- and size-qualified demand rather than a metro average applied to a submarket, and priced to effective rents net of the concessions clearing new Phoenix lease-ups today.
The coverage analysis was then stress-tested on the two variables a lease-up asset is most exposed to: absorption pace and effective rent. We ran the DSCR-constrained takeout against a slower lease-up and a deeper concession, and against an exit-cap expansion, to confirm the construction loan still converts and the equity still clears its floor when the Valley digests more slowly than the base case. One scope boundary is worth stating plainly: as the feasibility consultant we reference, but do not perform, the appraisal and the Phase I environmental site assessment, and a market study is not an appraisal under USPAP.12 That combination — independent absorption, capture, effective rent, and a stressed, DSCR-constrained takeout — is what lets both lenders rely on the file.
Underwriting an Arizona apartment project for an agency takeout? Start with the market study.
Feasibility Study Company prepares independent multifamily feasibility and market studies for agency, HUD-FHA, life-company, CMBS, and conventional bank construction credits, built to the NCHMA and lender coverage standards your capital source must document. A methodology briefing walks through the absorption, capture, effective-rent, and DSCR analysis behind a case like this one, calibrated to your submarket and metro.
Request a methodology briefingData sources and dates.
The deal figures are illustrative of the engagement type; the market data that grounds each dimension is real and sourced, drawn from our standing Arizona, Multifamily, and Conventional & Institutional analyses and the primary authorities they cite.
- U.S. Census Bureau, Vintage 2024 Population Estimates: Phoenix–Mesa–Chandler MSA 5,186,958 and Maricopa County 4,673,096 as of July 1, 2024 (fourth most-populous U.S. county, third-largest numeric county gain); Arizona approximately 7.6 million.
- Yardi Matrix, Phoenix multifamily report (January 2026): asking rents −4.1% year over year through November 2025, second-lowest among top-30 metros; stabilized occupancy 93.4% in October 2025; more than 60,000 units added since January 2022.
- CoStar via Solex CRE Phoenix mid-year review (June 2026): multifamily 11.5% vacancy, −2.3% rent growth, trailing-12-month net absorption approximately 21,491 units, a market record outpacing deliveries for the first time since early 2021.
- Kidder Mathews, Phoenix multifamily market report (Q1 2026): 11.8% vacancy, $1,535 average asking rent, 16,399 units under construction (down approximately 30% year over year).
- Getmultifamily analysis of CoStar data (January 2026), Phoenix multifamily 12.5% vacancy; Northmarq, Phoenix multifamily (Q1 2026), approximately 1,800 units delivered against 26,000+ under construction.
- Cushman & Wakefield | PICOR, Tucson multifamily market report (Q1 2026): multifamily vacancy tightening to 8.75% with rents up approximately $12 quarter over quarter.
- RealPage Market Analytics, 4Q 2025 Update and 2025–2026 commentary: national completions, starts, net absorption, and occupancy; roughly 16.9% of stabilized units offering concessions averaging a 10.9% discount (highest since mid-2014); apartment operating expenses of 38–55% of revenue.
- NAHB, Eye on Housing (July 2025), citing the U.S. Census Bureau Survey of Construction: 2024 multifamily completions approximately 608,000, the most since 1986, then falling roughly 30% in 2025.
- CBRE, Q4 2025 Multifamily Underwriting Survey and 2025 investment-volume data: going-in cap rates near 4.75% on core assets, stabilized product broadly low-to-mid 5%; agency coverage conventions of ~1.20x–1.25x; transaction volume rebuilding.
- Federal Housing Finance Agency (November 24, 2025): 2026 multifamily loan purchase caps of $88 billion per Enterprise ($176 billion combined), positioning the agencies as refinance backstops.
- National Council of Housing Market Analysts, Model Content Standards Version 3.1 (September 2025): primary market area definition, demand and capture-rate methodology, absorption benchmarks (commonly 12–25 units per month for garden and mid-rise product), and required field verification of comparables.
- Fannie Mae DUS and Freddie Mac Optigo program terms; CBRE underwriting survey; the DSCR-constrained takeout mechanic (a loan meeting a ~1.25x coverage minimum can fall well below the ~65% LTV ceiling). A market study is not an appraisal and is not conducted under USPAP Standards 3 and 4.
- U.S. Small Business Administration owner-occupancy requirements: both 7(a) and 504 require owner-occupancy (51% existing, 60% new construction), so pure investment apartments are SBA-ineligible and route through agency, HUD-FHA, life-company, CMBS, and conventional bank channels instead.
- Arizona Department of Water Resources, Phoenix AMA groundwater model (June 2023): approximately 4.86 million acre-foot 100-year shortfall and a pause on new groundwater-reliant Assured Water Supply determinations under the 1980 Groundwater Management Act; Designation and Certificate of Assured Water Supply framework.
- TSMC press release, remarks of CEO C.C. Wei (March 4, 2025): total planned Arizona investment of $165 billion, the largest foreign direct investment in U.S. history; JLL, North American Data Center Report — Midyear 2025: metro Phoenix #2 for planned development.
- Yardi Matrix, Winter 2026 outlook (via Multifamily Dive and CRE Daily): per-unit apartment trade pricing near $208,000, below the cost to build in high-cost submarkets where all-in development runs $300,000 per unit or more — the development-margin context for merchant construction.