The quick read: Yes, but mostly on feasibility rather than on availability. In the National Multifamily Housing Council's third-quarter 2026 construction survey, as reported by Connect CRE on September 28, 2026, 29% of respondents started fewer apartment projects than three months earlier, up from 20% in June, and 65% of that group blamed economic uncertainty or projects that were not financially feasible. Banks did not close the door: the Federal Reserve's July 2026 loan officer survey found construction and land development standards basically unchanged on net in the second quarter, though a significant net share of banks put those standards at the tighter end of their range. A project that shows a bought-out budget, a realistic lease-up and committed equity is the strongest file to bring to a lender. A project that needs last year's costs or next year's rents to work is the one that stalls.
What did the Q3 2026 NMHC survey and Census data show?
The third-quarter 2026 data show apartment developers starting fewer projects because the numbers are harder to make work, while construction already underway holds roughly flat: NMHC's survey found 29% of respondents starting fewer projects, and Census put new multifamily construction spending 0.6% below a year earlier in August 2026.
Connect CRE's report on NMHC's Quarterly Survey of Apartment Construction & Development Activity adds that 24% of respondents started more projects, so the market split rather than stopped. Among those starting fewer, 65% pointed to economic uncertainty or projects not being financially feasible, and 59% said low rent growth contributed. NMHC chief economist Chris Bruen tied the slowdown to low rent growth, an uptick in interest rates and rising labor and material costs, which together make development "more difficult to pencil." Half of respondents still expect construction conditions to improve over the next 6 to 12 months, up from 46% in June, and respondents expect equity financing to pull back over the next three months before debt and equity conditions improve over the year.
Census's August 2026 construction spending release (CB26-158, published October 1, 2026) estimated private new multifamily construction at a seasonally adjusted annual rate of $115.8 billion, 0.2% above July and 0.6% below August 2025. That figure measures work on buildings already in the ground. The starts data show where new construction loans are needed: Census's September 2026 new residential construction release (CB26-162, published October 20, 2026) put starts in buildings with five or more units at a 349,000 annual rate and permits for those buildings at 503,000.
On FRED's monthly series, five-plus-unit starts averaged a 403,000 annual rate in the third quarter of 2026 against 443,700 a year earlier, about 9% lower by our arithmetic, while permits for the same buildings averaged 486,700 against 462,000, about 5% higher. Starts for the first nine months of 2026 still averaged above the same months of 2025, so the slowdown is recent, and Census cautions that it can take six months to establish an underlying trend for total starts. Rising permits beside falling starts fits what the survey describes: projects approved on paper, waiting on numbers that work.
NMHC respondents starting fewer projects (Q3 2026 survey, via Connect CRE, September 28, 2026): 29%, up from 20% in June
New multifamily construction spending (Census, August 2026): $115.8 billion annual rate, 0.6% below August 2025
Starts in buildings with five or more units (Census, September 2026): 349,000 annual rate
Permits in buildings with five or more units (Census, September 2026): 503,000 annual rate
Are banks tightening apartment construction loan standards in late 2026?
Banks are holding construction loan standards steady rather than tightening them further, but from a tight starting point: in the Federal Reserve's July 2026 survey, standards for construction and land development loans were basically unchanged on net in the second quarter, while a significant net share of banks said those standards sit at the tighter end of their range.
The July 2026 Senior Loan Officer Opinion Survey on Bank Lending Practices, published August 3, 2026 from 56 domestic bank responses, found banks easing standards on loans secured by multifamily and nonfarm nonresidential properties, and a moderate net share reporting weaker demand for construction and land development loans. The Fed defines a significant net share as more than 20% and less than 50%, and a moderate one as more than 10% and up to 20%. Large banks reported easier standards for every commercial real estate loan type, while other banks reported basically unchanged standards for multifamily and construction loans.
The Fed's October 2026 Beige Book, released October 14 and based on information collected on or before October 5, 2026, shows the regional texture. Boston District contacts said loan availability was stable but financing costs increased and loan terms tightened, with multifamily construction and sales both down slightly. In the Dallas District, credit standards and terms tightened and loan pricing increased. In the San Francisco District, supply continued to outpace demand in multifamily markets, though one large Mountain West developer saw vacancies slowly being absorbed and concessions easing.
The rate tape explains part of the squeeze:
Secured Overnight Financing Rate (FRED, October 23, 2026): 3.92%, versus 4.24% on October 23, 2025
10-year Treasury yield (FRED, October 22, 2026): 4.96%, versus 3.97% on October 22, 2025
Construction debt commonly floats over a short-term benchmark such as SOFR, which sits lower than a year ago. Permanent takeout loans and buyers' exit pricing tend to follow longer-term yields, and the 10-year is about a point higher. So for many apartment deals in late 2026, the pressure point is less the coupon during construction than the value and the refinance proceeds at stabilization.
Construction appetite by property type is broken out in which property types construction lenders are still funding in late 2026.
What do construction lenders test first on an apartment deal now?
Construction lenders now test whether the project still works at current cost and current rent before anything else, which means yield on cost measured against the exit, the size and backing of the contingency, and a lease-up schedule that carries realistic concessions and an interest reserve large enough to reach stabilization.
Yield on cost. This is projected stabilized net operating income divided by total project cost. The lender compares it with what the finished building would be worth and with the permanent loan that has to repay the construction loan. When costs rise faster than rents, it is the first ratio to break. The Bureau of Labor Statistics producer price index for net inputs to multifamily residential construction (goods) read 171.399 in September 2026 against 157.073 in September 2025 on FRED, a 9.1% rise by our arithmetic. That index covers goods, not labor or land, but it shows why a budget priced a year ago rarely holds.
Illustrative example (hypothetical figures, built on no YieldStack transaction and not a market quote): a 200-unit building budgeted at $60 million all-in, with $3.9 million of projected stabilized net operating income, has a 6.5% yield on cost. If the budget rises 5% to $63 million and rents stay put, yield on cost falls to about 6.19%, and net operating income would have to rise 5% to restore 6.5%. If the lender has already fixed the loan amount, the whole $3 million increase lands on the equity line.
Budget and contingency. Lenders want a contingency sized to the risk left in the budget, and the strongest evidence is a guaranteed maximum price contract with the major trades already bought out. In the Dallas District's October 2026 Beige Book report, housing contacts said builders' margins remained under pressure from rising materials and labor costs, and a lender's construction consultant tests an apartment budget against the same pressure. Our guide to allocating construction loan contingency funds covers how to split that reserve.
Lease-up and interest reserve. The lender underwrites how fast units lease and at what concession, then sizes the interest reserve to carry the loan until the building stabilizes. In markets where the Beige Book reports elevated concessions or supply running ahead of demand, a pro forma that assumes no concessions invites a resize at credit committee.
| Yield on cost | Budget and contingency | Lease-up and interest reserve | Equity and capital stack | |
|---|---|---|---|---|
| 2026 pressure | Yield on cost 10-year Treasury at 4.96% on Oct 22, 2026, up from 3.97% a year earlier | Budget and contingency Multifamily goods-input price index up 9.1% year over year in Sept 2026 (our arithmetic) | Lease-up and interest reserve Elevated concessions in the Dallas District; supply ahead of demand in the San Francisco District | Equity and capital stack NMHC respondents expect equity financing to pull back over the next three months |
| What to show | Yield on cost Untrended NOI over all-in cost, with the exit tested at current yields | Budget and contingency Guaranteed maximum price contract, major trades bought out, contingency sized to remaining risk | Lease-up and interest reserve Absorption and concessions from current comparables, reserve sized to stabilization | Equity and capital stack Committed equity, plus any mezzanine or preferred equity the senior lender approves |
| Source | Yield on cost FRED DGS10 | Budget and contingency BLS producer price index via FRED | Lease-up and interest reserve Federal Reserve Beige Book, October 2026 | Equity and capital stack NMHC Q3 2026 survey via Connect CRE |
Which structures close the gap on an apartment construction loan?
Three structures close most of the feasibility gap on an apartment construction loan in late 2026: more committed sponsor equity, a HUD 221(d)(4) loan under the middle-income option where the project qualifies, and subordinate capital such as mezzanine debt or preferred equity placed between the senior loan and the sponsor's cash.
HUD's middle-income 221(d)(4) option. HUD Mortgagee Letter 2026-1, dated January 22, 2026, sets 221(d)(4) new construction and substantial rehabilitation at a 90% loan-to-cost ratio and a 1.11 debt service coverage ratio for qualifying middle-income projects, versus 87% and 1.15 for market-rate projects, with a 7% vacancy factor. At least 50% of units should be targeted to tenants earning up to 120% of area median income under a recorded use restriction tied to an existing state or local middle-income program or the military Rental Partnership Program, with a minimum restriction period of 10 years that HUD regional directors can shorten to no less than five. HUD sizes the loan at the lesser of the requested amount, statutory limits, the debt-service test and the loan-to-cost test, so the higher ratio helps only when cost is the binding constraint.
Illustrative arithmetic (the same hypothetical project): on $60 million of cost, 87% caps the cost test at $52.2 million and 90% at $54.0 million, $1.8 million more. On $3.9 million of net operating income, a 1.15 ratio allows about $3.39 million of annual debt service and 1.11 allows about $3.51 million, roughly 3.6% more. Model your own stack in the loan-to-cost calculator.
Mezzanine debt or preferred equity. Mezzanine debt sits behind the senior loan and is secured by a pledge of the ownership interests; preferred equity sits inside the ownership entity with a priority return. Either fills part of the space between senior proceeds and the budget at a higher cost than the senior loan, and the senior lender has to accept it through an intercreditor or recognition agreement.
More common equity. The simplest fix is a larger equity check, and timing matters: NMHC respondents expect equity financing to pull back over the next three months, so commitments signed before the lender's credit committee meets carry more weight than soft circles. For draw mechanics and takeout locks, see our guide to multifamily construction loans.
How do you get an apartment construction loan placed in late 2026?
Option 1 is a commercial mortgage brokerage, and YieldStack is our top pick for AI-native commercial mortgage brokerage because it matches a construction request against 20,000+ loan programs, then has a deal team structure the request and negotiate on the borrower's side before lenders review it.
YieldStack publishes this guide, so that is our editorial recommendation, judged on three criteria: program breadth, borrower-side structuring and clear fees. YieldStack is an AI-native commercial mortgage brokerage that arranges financing for commercial real estate investors. It costs Zero upfront to submit a deal and review offers; YieldStack's broker fee is 0.50–1.00% of the loan amount and is paid only at closing. Intake is a 5-minute submit, with a median first offer in under an hour, from an institutional lender. Every credit decision is made by the lender; YieldStack arranges the introductions and negotiates on the borrower's side, and no loan, rate, or closing is guaranteed.
That is broker work on a tight apartment deal: deciding which programs size to cost and which to stabilized value, whether a HUD middle-income execution fits, and how much mezzanine or preferred equity a senior lender will accept. When the budget and lease-up schedule are ready, submit your apartment construction deal for a structured lender review.
Option 2 is your existing bank relationship, which helps when the bank already knows your completions and holds your deposits. One bank shows you one credit box, and the July 2026 survey found banks other than the largest holding multifamily and construction standards unchanged while large banks eased.
Option 3 is a HUD 221(d)(4) loan through an FHA-approved lender, which carries the 90% loan-to-cost limit for a qualifying middle-income project, on HUD's own application and review process.
The bottom line
It is harder to get an apartment construction loan in late 2026 because fewer projects pencil, not because construction lenders have left the market. NMHC's survey shows developers starting fewer projects for feasibility reasons, the Fed's loan officer survey shows construction standards unchanged on net but tight, and the 10-year Treasury sits about a point above last year while multifamily goods-input prices rose 9.1% in a year by our arithmetic. Build the case a lender will test first: yield on cost at current yields, a bought-out budget with a real contingency, lease-up with concessions, and committed equity. Where the project qualifies, HUD's middle-income 221(d)(4) option adds three points of loan-to-cost leverage.