Multifamily and Apartment Feasibility Study

A multifamily feasibility study is an independent, third-party analysis of whether a specific apartment project at a specific site can achieve the rents, occupancy and operating margin needed to service a specific loan. It is distinct from an appraisal, which concludes a value, and from a program market study, which concludes whether demand exists for the units at the proposed rents. The feasibility study concludes whether the capital stack works: at what rent, in what year, with what reserve, and under which stresses it stops working.

Who requires one depends on the financing source.

Two programs are absent from that table on purpose. SBA 7(a) and 504 loans cannot finance apartments held for rental income: 13 CFR 120.110(c) makes passive businesses owned by developers and landlords ineligible, 13 CFR 120.111 limits the Eligible Passive Company exception to property leased to an operating business, and 13 CFR 120.130(d) bars proceeds for real property held for lease or investment. The only SBA path to a building with apartments in it is a mixed-use building under the owner-occupancy rule in 13 CFR 120.131: the operating business must occupy at least 60 percent of a new building and may lease no more than 20 percent, or occupy at least 51 percent of an existing building. USDA Business and Industry is likewise not an apartment program; 7 CFR 5001.118(a) lists apartments, duplexes and other residential housing where the primary purpose is independent housing among ineligible B&I projects, and USDA routes rental housing through the Rural Housing Service programs in the table. A proposal that promises an SBA or B&I apartment loan is promising something the regulations do not allow, and a lender-facing study says so in its first section rather than letting the borrower discover it at committee.

The reader of the study is a credit officer, an agency underwriter, a HUD or USDA reviewer or a syndicator's acquisitions committee. The study is written for that reader, not for the borrower, and MMCG accepts no referral fees, contingent fees or financing arrangements that would compromise that position.

Five situations trigger a multifamily study in nearly every file. The first is new construction of any size, where no operating history exists and the lender must rely on projected rents, a projected lease-up and a projected operating budget. The second is an acquisition of a property still in lease-up or recently delivered, where trailing income does not yet exist and the agency takeout depends on reaching the stabilization test, which Fannie Mae typically sets at 90 percent occupancy for 90 days before funding. The third is a conversion of a hotel, motel or office building to apartments, where the cost basis, the code path and the achievable rent all differ from new construction and the lender has no comparable set of its own. The fourth is a program application: HUD 221(d)(4), USDA Section 538 and every LIHTC allocation require a market study by rule, and the lender's own underwriting still needs the coverage schedule the rule does not supply. The fifth is a small multifamily project of 5 to 50 units financed by a community bank or credit union, where the lender's policy requires a third-party study because the deal is too small for an agency execution and too large for a residential appraisal to carry.

Acquisitions of stabilized properties with trailing operating statements follow a different path. The agency lender underwrites the trailing twelve months and the appraisal carries the market analysis. In those files the feasibility study normalizes the trailing income for concessions, loss to lease, below-market taxes on an old assessment and deferred capital, and tests whether coverage survives the normalization. It does not replace the trailing test; it explains it.

The study is written against a market that is absorbing the largest delivery wave since the mid-1980s and has already stopped starting the next one. On the licensed national data MMCG carries, the United States apartment market held 21.0 million units at 7.8 percent vacancy on October 3, 2026, with a $1,801 asking rent, a $1,726 effective rent and twelve-month asking rent growth of 1.4 percent. Stabilized vacancy, which excludes communities still in lease-up, stood at 6.6 percent, down 40 basis points from its fourth-quarter 2025 peak. Vacancy by quality tier was 9.5 percent for 4 and 5 Star product, 7.9 percent for 3 Star and 6.1 percent for 1 and 2 Star, which is why a new-build study cannot borrow the market-wide vacancy as its own.

Demand is holding and supply is fading. Twelve-month net absorption reached 504,750 units, and third-quarter 2026 absorption of roughly 142,000 units ran 29 percent above the prior year. Deliveries reached nearly 695,000 units in 2024, the highest annual total since the mid-1980s; trailing twelve-month deliveries fell to 435,764; third-quarter starts of about 60,000 units were the lowest quarterly total since 2012 and about 71 percent below the cycle peak; and the national pipeline contracted from nearly 1.2 million units at its 2023 peak to 567,164 units under construction. About 60 percent of the recent supply growth landed in the South and Mountain regions, and seven of the 50 largest markets still carried negative annual rent growth in the third quarter. The forecast MMCG carries holds vacancy at 7.9 percent through 2027 before it eases to 7.6 percent in 2028 and 7.3 percent by 2030, with deliveries of about 448,000 units in 2026, 362,000 in 2027 and 302,000 in 2028, and rent growth of 1.8 to 2.1 percent a year. Source: CoStar Group, United States Multi-Family National Report, October 3, 2026, licensed to MMCG Invest, LLC.

Public sources corroborate the direction. The Census Bureau's Housing Vacancy Survey put rental vacancy at 7.3 percent in the second quarter of 2026. Completions in buildings of five or more units ran 35.7 percent below the prior year in August 2026. The National Multifamily Housing Council's Market Tightness Index rose to 57 in July 2026, its first reading above 50 in a year. The concession load is the figure a lease-up study cannot ignore: RealPage's August 2026 analytics reported 15.4 percent of stabilized units offering concessions at an average discount of 11 percent of asking rent, about 5.7 weeks free on a twelve-month lease.

Debt is more expensive than the market narrative admits. The Federal Open Market Committee raised its target range to 3.75 to 4.00 percent on September 16, 2026, the first increase since 2023, and the 10-year Treasury stood at 5.25 percent on October 1, 2026, with overnight SOFR at 3.87 percent. Agency lender rate sheets on that date indicated ten-year fixed-rate conventional loans at 80 percent loan-to-value and 1.25x coverage at roughly 6.40 to 6.90 percent, HUD 221(d)(4) at roughly 7.15 percent before the 0.25 percent mortgage insurance premium, and bank construction loans at SOFR plus 200 to 375 basis points. At those rates the coverage test, not the loan-to-value test, sizes nearly every permanent loan, and a study that reports proceeds at 80 percent of value without reporting the coverage-constrained loan is reporting a number the lender will not fund.

Capital is available but selective. The Federal Housing Finance Agency set 2026 purchase caps of $88 billion for each of Fannie Mae and Freddie Mac, 20.5 percent above 2025, with at least half mission-driven and with workforce housing loans excluded from the cap entirely. The Mortgage Bankers Association reported $381.8 billion of multifamily originations in 2025 and projects $875 billion of commercial and multifamily mortgage balances maturing in 2026, of which only $39 billion is agency multifamily and health care debt; the maturity wall is a bank, debt fund and CMBS problem, which is why a bridge-to-agency refinance is underwritten to the exit coverage rather than to agency availability. The Federal Reserve's July 2026 Senior Loan Officer Survey found construction lending standards unchanged at a tight level with weaker demand, while permanent multifamily standards eased modestly. CMBS multifamily delinquency stood at 6.53 percent in the second quarter of 2026 against 0.60 percent at Fannie Mae and 0.51 percent at Freddie Mac. Apartment values sit about 20 percent below their 2022 peak on repeat-sale indices, with market cap rates of 6.2 percent overall, 5.6 percent for 4 and 5 Star product and 6.3 percent for 3 Star, which is the exit assumption a study must carry and then stress.

The study defines the primary market area from the subject's actual commute shed and the renter's actual alternatives, not from a radius. For a suburban garden community that is the drive-time ring to the employment nodes its renters work in and the competing communities those renters would otherwise choose; for a downtown podium it is the walkable district and the transit lines; for a rural USDA project it is the county and the one or two towns whose renters would relocate. The study maps the boundary, states why, and holds it fixed for demand and supply alike, because a market area that expands for demand and contracts for competition is the oldest way to manufacture feasibility.

Each demand generator is named and measured. Employment is taken from the employer, the state workforce agency or the institution's own filings, with announced expansions dated and sized by headcount. Military installations are measured by assigned strength and civilian employment, universities by enrollment and on-campus bed capacity, hospitals by employment and bed count, and distribution and manufacturing by permanent headcount rather than construction peak, because a construction crew rents by the week and leaves. Household growth comes from the decennial census and the American Community Survey, and building permits by place and county from the Census Building Permits Survey, so that the supply response to the same demand is visible in the same table.

Total demand is not the number of households in the market area. It is the number of renter households that can pay the subject's rent and would choose the subject's product. The study builds the income-qualified band from the American Community Survey's tenure-by-income tables. At a 30 percent rent-to-income ratio, which is the convention for market-rate product and most agency underwriting, a $1,800 rent requires a household income of $72,000; at the 35 percent ratio several housing finance agencies permit for affordable product, the floor falls to $61,700. The upper bound for market-rate product is the income at which a household would buy instead, set from the local price-to-rent relationship rather than assumed. Within the band the study counts renter households, applies annual turnover, adds net new renter household formation, and adds rent-burdened and substandard households only where the evidence is specific.

The result is expressed the way agency and housing finance agency reviewers expect: a capture rate, meaning the subject's units divided by income-qualified renter demand, and a penetration rate, meaning all competing units in the pipeline plus the subject divided by that demand. Housing finance agencies publish their ceilings. Georgia's Department of Community Affairs caps the overall capture rate at 35 percent in its rural pool and limits absorption to 24 months; Iowa requires justification above a 10 percent overall capture rate; Texas computes a gross capture rate under 10 TAC 11.302(i). Market-rate lenders publish nothing, so the study states the capture rate, compares it to what the competing communities achieved in their own lease-ups, and lets the lender judge. A capture rate below 10 percent in a growing market is routine. One above 20 percent in a market with a visible pipeline is the finding, not a footnote.

The competitive set is built from the communities that compete for the subject's renters, verified property by property against each community's own published unit count, bedroom mix, year built, amenity package, asking rents by floor plan and advertised concessions. Where a figure cannot be verified from the community's own site or a public filing it is marked as not verified in the census rather than carried silently from an aggregator, because the lender's appraiser will find the discrepancy and the study's credibility goes with it.

Announced, permitted and under-construction communities are added with their expected delivery dates, because a new community lowers every competitor's share from the day its leasing office opens. The pipeline is drawn from permits, site plan approvals and the delivery schedules of named projects, and the study states it as a percentage of existing inventory in the market area so that the lender can compare it to the metro figure. A market area where the pipeline has fallen below trailing deliveries is absorbing; one where the pipeline exceeds them is still digging. Program studies add the program inventory: every USDA Section 515 and 538 property, every LIHTC property with its income mix and placed-in-service year, and every project-based assisted property, drawn from the USDA multifamily property list, the HUD LIHTC Database and the state agency's inventory.

New communities do not open at stabilized occupancy, and in 2026 they do not open without concessions. The study builds the lease-up curve from the monthly pace the named competitors actually achieved, caps it at a defensible share of the market area's trailing net absorption, and states the pace in units per month so that the lender can test it. As a frame rather than a rule, garden communities of 200 to 300 units in absorbing suburban markets have been leasing at roughly 15 to 25 units a month and reaching 90 to 93 percent occupancy in 10 to 16 months; podium product in urban submarkets runs slower; build-to-rent communities delivered in phases lease at roughly 8 to 15 homes a month; and age-restricted product can take 18 to 30 months because the renter's decision cycle is long. These are MMCG working ranges from recent lease-ups, not published standards, and the study replaces them with the local evidence in every case.

Concessions are modeled explicitly. The study carries the concession the competing communities are advertising, in weeks free and as a percentage of asking rent, through the lease-up period, burns it off only as fast as the market evidence supports, and reports the gap between asking and effective rent in every year of the projection. The national gap on the licensed data was 4.2 percent in October 2026; in the Sun Belt markets that received the most supply it ran 8 to 10 percent. Every lease-up study runs a three-month and a six-month delay case, because the interest reserve and the operating deficit reserve are sized from them, and a lender that funds the base case without the reserve has funded the wrong number.

The subject's rent is set from the competitive census, not from the market average. New construction commands a premium over the market-wide asking rent because the market average includes forty-year-old product; the question the study answers is the premium over the newest comparable communities, by floor plan, net of their concessions. The study states the subject's rent per square foot against each comparable, positions it on the quality tier the project will actually occupy, and reports the achievable rent rather than the proposed rent when the two differ.

Unit mix is tested against bedroom-level demand rather than left to the site plan. On the licensed national data, three-bedroom apartments carried the highest vacancy of any unit type at roughly 9.3 percent in 2026, against about 7.8 to 8.0 percent for studios, one-bedrooms and two-bedrooms, which is consistent with build-to-rent and single-family rental communities absorbing the larger-household renter. The study therefore tests whether a conventional garden or podium community should carry more than 5 to 10 percent three-bedroom units, and whether larger-household demand in the market area belongs to a horizontal product instead. Where the project is affordable, the mix is tested against the agency's requirements; Georgia, for example, requires at least 10 percent one-bedroom units in family applications and rewards a three-bedroom share of 20 percent or more.

Projections are built line by line, not as a percentage of revenue. Rental income is built from the achievable rent by floor plan, the lease-up curve and the concession schedule. Other income is built from the fees the competing communities actually charge. Vacancy and collection loss is carried at the program convention: HUD underwrites market-rate 221(d)(4) and the Middle Income option at a 7 percent vacancy factor and affordable product at 5 percent; agency lenders underwrite a minimum 5 percent economic vacancy even where the market is tighter; Texas's housing agency underwrites 7.5 percent and Georgia's at least 7 percent for tax credit product. Loss to lease is reported separately where in-place rents sit below market.

Expenses carry the three conventions a lender expects regardless of ownership. A management fee of about 3 percent of effective gross income is included even where the sponsor will self-manage, because the lender underwrites a replaceable operator. A replacement reserve is included at the lender's minimum, typically $250 to $300 per unit per year for new construction and more for older product. Property taxes are computed from the actual levy and the state assessment ratio on the stabilized value, not from the current bill on vacant land, and the study shows the arithmetic: Georgia assesses at 40 percent of market value, so a combined 2026 levy of 26.3 mills inside the city of Tifton works out to about $10.53 per $1,000 of value; South Carolina assesses rental housing at 6 percent against 4 percent for owner-occupied homes and carries full school millage on rentals, so unincorporated Spartanburg County produces $19 to $23 per $1,000 of value on a rental home, roughly four times the owner-occupied burden. Insurance is quoted for the construction type and location and trended; the National Apartment Association's most recent survey put insurance at $777 per unit, up 10.8 percent in a year, inside total operating expenses of $8,657 per unit. Labor is built from local wage evidence, and utilities are allocated between owner and resident per the utility allowance schedule where the program requires one.

Operating break-even for a garden community commonly falls in the mid-30s to low-40s percent occupancy before debt service. That figure is rarely the binding constraint. Debt service is, which is why the study reports break-even both before and after debt service and reports the occupancy at which the program's coverage floor is first met.

For new construction the study presents a project cost estimate by line item: land, site work, hard cost by building system, FF&E, soft cost, financing cost, guarantee fees, reserves and contingency, benchmarked against the current cost surveys and local permit and bid evidence. Rider Levett Bucknall's first-quarter 2026 survey put multifamily hard cost at $185 to $275 per square foot of gross floor area in Austin, $190 to $280 in Dallas, $200 to $295 in Phoenix, $215 to $365 in Charlotte and $225 to $400 in Denver, and its national index rose 4.45 percent in the year to the third quarter of 2026, with steel mill products up 23.4 percent and aluminum shapes up 27.3 percent under the 2026 tariffs. The study escalates the survey figure to the construction midpoint and carries a hard cost contingency of 5 to 6 percent. For a HUD application the budget is reconciled to the statutory per-unit limits at the 270 percent high-cost percentage HUD applies to every listed base city in 2026, with a 315 percent waiver available project by project, and to the Davis-Bacon wage decision for the county.

The capital stack is then stated by program, with the test that sizes the loan identified in each case.

Under every one of these the coverage test binds at 2026 rates before the leverage test does. The study states the coverage-constrained loan, the equity gap it creates against the proposed stack, and the year in which the floor is first met.

The projection is run through the coverage test the program applies, year by year, not at stabilization alone. For a bank construction loan with an agency takeout the study reports the year the agency's 1.25x is first met at the indicated rate and the debt yield at that date; for HUD, the year 1.15x or 1.11x is met and the reserve that carries the project there; for USDA Section 538, the coverage the lender must certify at guarantee issuance; for a tax credit deal, the first-year coverage against the agency floor and the fifteen-year schedule.

Sensitivity is presented as a table of outcomes, not as a narrative. The study stresses rent with zero growth in years one and two, stabilized occupancy by three and five points, concessions by four additional weeks through lease-up, lease-up by three and six months, controllable expenses by 10 percent and insurance and taxes by 8 to 10 percent for two years, the exit cap rate by 50 and 100 basis points, the permanent rate by 100 and 200 basis points and hard cost by 5 and 10 percent, and reports the coverage ratio at each step. Where the demand base includes a finite generator the study runs a post-generator case. Where the project is build-to-rent it runs an exit case under the 2026 federal investor limits. The study concludes feasible, feasible with conditions, or not feasible, and states the condition in the lender's terms: the maximum loan, the minimum equity, the reserve, and the entitlements and letters that must exist before closing.

Garden and three-story walk-up communities on surface parking are the core of bank, agency and USDA lending. Their feasibility turns on the competitive census, the lease-up pace and the property tax computation, and their cost per unit is the lowest of any product. Three Star communities are the right comparable set for most suburban garden product.

Podium, wrap and mid-rise product of four to seven stories over structured parking carries a materially higher cost per unit and per square foot, a longer construction period and a slower lease-up, and its feasibility turns on whether the submarket's rent per square foot supports the parking structure. Four and five Star communities are the comparable set, and their 9.5 percent national vacancy in 2026 is the lease-up frame.

Build-to-rent communities of detached homes, duplexes and attached townhomes on a single parcel under common ownership are the fastest-growing product in the lender pipeline and the one with the newest regulatory question. The 21st Century ROAD to Housing Act, enacted July 11, 2026, bars investors with investment control of 350 or more single-family homes from buying more, defines a single-family home as a structure of two or fewer units, excepts build-to-rent purchases and construction, and imposes no forced-sale period; attached buildings of three or more units fall outside the definition entirely. A BTR study models the product mix against that definition and models the exit as a sale to another institutional holder or a long hold. See the build-to-rent feasibility study page.

Workforce and mixed-income communities serving households at 80 to 120 percent of area median income now carry two program advantages: HUD's Middle Income 221(d)(4) option at 90 percent of cost and 1.11x coverage, and exclusion from the agency purchase caps, both of which require a recorded restriction of at least ten years under a state or local program. The study documents the qualifying program and tests the restricted rents against the market.

Age-restricted 55 and over apartments are underwritten as apartments with a slower lease-up, a smaller unit, lower turnover and no care component. They are lawful under the Housing for Older Persons Act's 80 percent occupancy test and are excepted from the federal investor limits. Anything with licensed care belongs to a different asset class with a different license and is covered on MMCG's assisted living feasibility study page, which this page does not duplicate.

Mixed-use buildings with apartments over retail or office are underwritten as two businesses. The commercial space carries its own lease terms, credit and vacancy, the agency lender caps the commercial share of income and area, and the combined coverage must hold with the commercial space vacant. Mixed-use is also the only SBA path to apartments, under the owner-occupancy rule in 13 CFR 120.131, and the study measures the operating business's occupied share by rentable area because that figure, not the apartment count, decides eligibility. The mixed-use feasibility study page covers the commercial component.

Conversions of hotels, motels and office buildings to apartments are underwritten on a cost basis the market has no comparable for. Hotel and motel conversions avoid the floor-plate and plumbing problems that stop most office conversions, because the rooms already have windows and stacks, and they are the natural product for a lender that already holds the hospitality collateral; MMCG's hotel feasibility study pages cover the hospitality side of that decision. Office conversions depend on plate depth, window access and the local incentive, which in 2026 ranges from Chicago's tax-increment subsidies on LaSalle Street to Colorado's state conversion credit of up to $3 million per project.

Small multifamily of 5 to 50 units is financed by community banks, credit unions and, from $2 million, Freddie Mac's Conventional Small product, which replaced the Small Balance Loan program on April 15, 2026 and left loans below $2 million to the banks. These lenders require a third-party study on new construction because the deal sits below the agency threshold and above the residential appraisal, and the study is shorter but carries the same census, lease-up and levy work.

Student housing is underwritten by the bed, pre-leased by the academic cycle and guaranteed by parents, and its demand turns on enrollment, on-campus capacity and the university's housing policy. It is a specialist product, and MMCG prepares bed-based demand and pre-leasing analysis on request rather than as a standing page.

Bank construction loans remain the entry point for most market-rate projects, at about 55 to 65 percent of cost, with recourse or a completion guaranty, an interest reserve sized to the lease-up and a takeout condition tied to the agency's stabilization test. Fannie Mae DUS and Freddie Mac Optigo conventional executions provide the takeout at 80 percent of value and 1.25x coverage on five to thirty-year terms, with Freddie's Conventional Small covering $2 million to $10 million. HUD 221(d)(4) provides construction and 40-year permanent financing in one loan at 87 percent of cost for market-rate product and 90 percent for Middle Income and affordable product, with a 0.25 percent mortgage insurance premium for applications submitted on or after October 1, 2025 under HUD's final notice of September 23, 2025, a market study whose effective date is the site inspection, and a lender-engaged market analyst who must meet HUD's experience and independence tests but needs no HUD-specific credential or approved list. HUD 223(f) refinances stabilized product at 87 percent of value and 1.15x.

USDA Section 538 guarantees rural rental housing loans up to 90 percent of cost or value for tenants at or below 115 percent of area median income, on 25 to 40-year terms, with $400 million of guarantee authority in fiscal 2026 and a preservation pilot that opens October 9, 2026. Section 515 direct loans have funded preservation, transfer and rehabilitation rather than new construction since fiscal 2011. See the USDA multifamily feasibility study page.

LIHTC 9 percent and 4 percent executions layer tax credit equity, priced at about $0.81 per credit nationally in mid-2026, over a permanent loan sized to the agency's coverage floor and over soft funds. The 2025 federal legislation raised 9 percent allocations by 12 percent from 2026 and lowered the bond test for 4 percent deals to 25 percent. See the LIHTC and affordable housing feasibility study page.

MMCG illustrates its multifamily method through nine model engagements, each performed on a real, publicly marketed parcel with public evidence and written in the register of the study itself, with the determination stated as proposed and as restructured, the full cost table, the loan assumptions for both structures, the line-item operating budget, the five-year pro forma with debt service coverage by year, the break-even and sensitivity tables and the conditions precedent. Each is published as its own page and indexed under multifamily engagements:

An MMCG multifamily feasibility study runs between 90 and 140 pages and includes the market area and demand generator analysis, the income-qualified demand segmentation with capture and penetration rates, the verified competitive census with the pipeline, the lease-up and concession projection, the rent positioning and unit mix test, the line-item operating projection from actual levies, the project cost estimate or acquisition analysis, the capital stack by program, the coverage schedule by year, the sensitivity table and a signed determination. Each competitor and demand generator is checked against its own published source where one exists, and the study states which figures are verified, which are licensed, which are carried from third-party or older sources and which remain conditions.

Standard delivery is 9 to 16 business days from engagement and receipt of the project file, with rush delivery available from 5 business days, for sites in any state on the feasibility study locations list. Fees begin at $4,900 for a single-site market-rate apartment study, and USDA Section 538, HUD, LIHTC and build-to-rent studies are quoted on scope. Payment is 50 percent at engagement and 50 percent at delivery. MMCG has offered a contractual acceptance commitment on every study since founding: revisions required by the lender or agency are made at no additional cost.