Self-Storage Feasibility Study: The Consultant's Comp Set vs the Broker's Feasibility Report

Every self-storage development file arrives with a feasibility report already in it. The developer commissioned it, a broker or consultant with a storage practice wrote it, and it concludes that the trade area is undersupplied. Sometimes it is right. The lender's problem is that it cannot tell from the document, because the report was written to get the deal financed, its comp set was assembled to make the per-capita arithmetic come out, and its lease-up was borrowed from the last cycle. That is why SBA lenders and CDCs order an independent study on self-storage new builds, and why the independent study starts by rebuilding the comp set rather than adopting it.

This is the reading we give lenders on self-storage files, current as of 7 September 2026, with the SOP 50 10 8.1 acquisition rules noted where they apply from 1 October. It draws on the Yardi Matrix national reports through August 2026, the public REITs' 2025 and 2026 filings, the SBA's loan-level FOIA releases, and the municipal record.

Why self-storage is an SBA asset at all

Self-storage looks like passive real estate, and 13 CFR 120.110(c) makes "passive businesses owned by developers and landlords that do not actively use or occupy the assets" ineligible for SBA financing. A self-storage facility is eligible because, and only because, the borrower operates it: staffing and management, security and access control, unit turnover, tenant administration, and ancillary sales of locks, boxes and insurance. An owner who leases the building to an operator, or who holds units under long-term leases and collects rent, is on the wrong side of the line. The distinction is the first question a credit officer asks, and a study that describes the subject as an investment property rather than an operating business has answered it badly.

Self-storage does not appear on SOP 50 10 8's example list of special-purpose property, which runs from hotels and gas stations to bowling alleys and cold storage, but the list is non-exhaustive and lenders and CDCs differ on whether they treat a storage facility as limited-market collateral for the 504 borrower contribution. The classification, and therefore whether the equity ask is 10% or 15%, or 20% on a start-up, should be confirmed with the lender at engagement rather than discovered at commitment.

The study itself is at lender discretion. SBA "may require" one under 13 CFR 120.160(b); the 504 program's stated circumstances for obtaining one, market saturation, an unproven concept, a specialized property, a project disproportionate to the applicant's size, and rapid growth on unseasoned debt, describe nearly every ground-up storage project, and market saturation is the one the asset class is now known for. On an acquisition numbered on or after 1 October 2026, SOP 50 10 8.1 requires 1.25x coverage on trailing EBITDA with no projections credited, total debt capped at the appraised business value, and a lender-ordered quality of earnings report at a $3 million business price net of real estate.

The market the study is written into

The self-storage cycle turned in 2023 and has not fully turned back. Yardi Matrix's national reports show advertised street rates for the combined 10-by-10 climate and non-climate mix falling 2.0% year over year in March 2026 to $16.07 a square foot, and still down 1.5% in July at $16.47, with 26 of the top 30 metros negative on an annual basis in June and a national average of 7.8 net rentable square feet per capita. Public Storage's average annual move-in contract rent fell to $11.60 a square foot from $12.97 a year earlier while its realized rent per occupied square foot held at $22.50, which is the gap between the street rate that attracts a new tenant and the in-place rent that existing-customer increases support. The four largest REITs reported same-store NOI between negative 1.5% and positive 0.1% in the fourth quarter of 2025, a cyclical trough, before Public Storage reported in the second quarter of 2026 that both move-in rate and occupancy were positive year over year for the first time since 2021, with move-in rates up about 4% in June.

Supply is the reason. The industry added 55 million to 58 million net rentable square feet in 2025 to a national inventory above 2.1 billion, and Yardi Matrix forecasts 51.1 million in 2026, 44 million in 2027 and about 38 million in 2028, taking new supply from 3.0% of stock in 2025 to 2.4% in 2026 against a long-term average of 4.2%. The under-construction pipeline stood at 595 properties and 44.1 million square feet in July 2026, 2.1% of stock, and the slowdown is real: Extra Space disclosed that the share of its same-store properties facing new competitive supply fell from the high 20% range in 2021 to 2023 to 8% in 2025 and an expected 6% in 2026. The pressure is concentrated. Phoenix carried 6.6% of inventory under construction in July, Orlando 5.1% and Sarasota-Cape Coral above 5% on top of 11.4 square feet per capita; Jacksonville carries 10.5 square feet per capita with 7% of inventory scheduled for 2026 delivery; Milwaukee carries 3.8 and posted 6.1% annual rent growth. Consolidation followed the trough: Public Storage's roughly $10.5 billion all-stock acquisition of National Storage Affiliates, announced in March 2026, creates a combined portfolio of about 4,600 facilities and 328 million square feet.

The number a lender should keep in view is the occupancy gap. At the end of the second quarter of 2026 the REITs' weighted average occupancy was 92.9%; the tracked non-REIT universe ran 83.4%, a gap of about 9.5 points that reflects pricing sophistication, marketing spend and operating platform, not location. An independent developer's pro forma that stabilizes at REIT occupancy is stabilizing at a number the independent universe does not reach.

Where the broker's report fails, and what the independent study does instead

The square-feet-per-capita headline. The developer's report almost always rests on one calculation: the trade area's existing rentable square feet divided by its population, compared to a national or regional benchmark, with the shortfall presented as unmet demand. The calculation is legitimate and it is where most self-storage studies go wrong, for three reasons. The benchmark moves: 7.8 square feet per capita is a national average across markets ranging from 2 in dense urban cores to above 20 in some Sun Belt metros, and it is not a saturation threshold. The numerator is usually wrong: reports count gross square feet, omit facilities outside a hard radius that draw from the same customers, and miss the pipeline entirely. And product type is blended: a market that is undersupplied in climate-controlled units and oversupplied in drive-up non-climate units reads as balanced on a single per-capita figure, and a study that builds climate-controlled units into that market on the strength of the blended number is wrong in the direction that matters.

The independent study disaggregates. It counts net rentable square feet by facility and by product type, climate-controlled against non-climate, drive-up against multi-storey, vehicle and RV storage separately, and it computes per-capita supply for each against the demand the subject's product actually serves. It adds every project under construction, planned and prospective inside the trade area, sourced and dated, because the pipeline is where the 2023 to 2025 oversupply came from and where the developer's report is silent.

The radius. The 1, 3 and 5 mile ring convention is the working standard, with 3 miles as the primary urban and suburban trade area, and it is also the easiest lever to pull. A radius that stops at 3 miles because the fourth mile holds a 900-unit competitor is a radius chosen to hide the competitor. The independent study justifies its trade area from drive time, arterial access and barriers, states the population and household base it produces, and shows what the conclusion looks like at the alternative boundary. Where the ring and the drive-time polygon disagree by a large factor, the study says so, because the difference propagates directly into the demand estimate and the coverage ratio.

The comp set. The report's comp set is the set that supports its rate. The study's comp set is every facility competing for the subject's customers, each verified against its own current source: the operator's own website and pricing page on a stated date, a call or visit to confirm occupancy, unit mix and specials, and the facility's product type and age. It states which figures are primary-verified and which are carried from aggregators, and it reconciles the asking rates it uses to the Yardi Matrix metro figures so the lender can see whether the subject's projected rate sits above or below the market it is entering. In a market where 26 of 30 metros were negative year over year in June 2026, a study that projects rate growth from opening has to explain which market it is in.

The lease-up. The last cycle's lease-ups were fast because the market was empty. Facilities opening into 2026 open into a national pipeline of 2.1% of stock, into metros where the under-construction share is 5% to 7%, and into a street-rate environment that turned positive only in the summer. The developer's report typically models absorption at a fixed number of units a month to stabilization inside a year. The independent study models absorption from comparable openings in the same metro over the same cycle, states the stabilized occupancy it is using against the REIT and non-REIT benchmarks, tests the ramp against the interest reserve and the SBA coverage floor by year, 1.15x for a Standard 7(a) loan and 1.10x for a Small Loan, and shows what a twelve-month delay in stabilization does to coverage. On a 504 construction loan the 15% contingency and the completion-value appraisal have to reconcile to the same budget.

The ordinance layer. A rising number of jurisdictions have stopped treating self-storage as a by-right commercial use. Atlanta's mayor signed a moratorium on new self-storage permits by executive order on 24 June 2026, the City Council extended it for 180 days on 6 July, a companion ordinance requiring a special use permit for any new facility went to the Zoning Review Board, and the city had already barred self-storage within 500 feet of the Beltline corridor. Coos Bay, Oregon adopted a twelve-month moratorium by Resolution 25-29 on 19 August 2025 after finding thirteen facilities providing more than twice the storage a community of its size typically requires. Prattville, Alabama adopted a temporary moratorium on mini-warehouses and self-storage in June 2025. Cape Coral, Florida let its moratorium lapse in April 2025 only after imposing a one-mile separation between facilities and a 500-foot setback from major intersections. Elk Grove, California took a citywide moratorium to its planning commission in August 2026 with nineteen facilities already in the city. For a proposed project the ordinance layer is a gating condition the study states first; for an entitled site inside a restrictive jurisdiction it is a frozen competitive set the study should claim.

What the loan tape says

SBA's loan-level FOIA releases for NAICS 531130, lessors of miniwarehouses and self-storage units, contain the fields to compute loan count, approval volume and charge-off rates by program and fiscal year, and a lender reading a self-storage study will have run them. The study should run them too, present the resolved-loan and dollar-weighted rates side by side with their denominators, separate the 504 construction lane from the 7(a) acquisition lane, and note that the 2021 to 2023 construction vintages, underwritten into the fastest supply growth the sector has recorded, are the cohort still seasoning. A study that quotes a lifetime charge-off rate without the vintage caveat will be corrected by a reviewer who has the same file.

What a lender should require

1. A statement of the operating model that satisfies 13 CFR 120.110(c), and the lender's classification of the collateral for the 504 contribution, confirmed at engagement.
2. Supply counted in net rentable square feet by facility and by product type, with the pipeline under construction, planned and prospective inside the trade area, sourced and dated.
3. The trade area justified from drive time, access and barriers, with the population base stated and the conclusion tested at the alternative boundary.
4. A comp set verified against each facility's own current source, with primary-verified and carried figures labeled, and asking rates reconciled to the metro benchmark.
5. Stabilized occupancy stated against the REIT and non-REIT benchmarks, with the reason the subject sits where it does.
6. Absorption drawn from comparable openings in the same metro over the current cycle, coverage by year at the applicable floor, and the sensitivity to a twelve-month stabilization delay.
7. The ordinance layer stated as a gating condition, with the entitlement status of the subject and of every competitor in the pipeline.
8. The SBA default record for the asset class with denominators and the construction-vintage caveat.
9. A signed certification of independence and qualifications, and a statement that the study was not prepared by or for a party compensated on the closing.