Self-Storage Underwriting Method: Square Feet per Capita, Lease-Up and the Class's SBA Record

A self-storage facility earns its income from many small rentals of secure space, each one a customer who stores goods and comes back for them ^[1]. No anchor tenant signs a long lease and no room is sold night by night, so the credit question comes down to three quantities a study can measure: how much storage space the households and businesses within a short drive will occupy, how much of that space is already built or on its way, and how quickly a new facility will fill the part left over. The sections below take the three in the order a lender's reviewer meets them, then turn to the class's record in the SBA loan data.

The general tools live in other articles of this series, and this one only points to them: counting competitors and staging the pipeline in Supply and Comparable Analysis: Counting Competitors, the Pipeline and the Properties a Study Leans On, building and testing a ramp in Ramp-Up and Absorption: How Long a New Project Takes to Stabilise, and How a Study Proves It, and drawing a trade area in Trade Area and Demand Methods: Capture Rates, Penetration Rates and Gravity Models. The space here goes to what storage adds, from a demand model written out term by term to the class's 7(a) and 504 record and a short worked example. The full worked example for the code's other occupant, outdoor RV and boat storage, is Worked Example: SBA 504 RV and Boat Storage Feasibility Study in New Jersey.

What Sits Inside Code 531130

The loan dataset assigns each SBA approval to an asset class by its industry code, and the self-storage class is a single code, 531130. The NAICS manual describes that industry as establishments mainly renting or leasing space for self-storage, and it names the forms the secure space can take: rooms, compartments, lockers, containers or outdoor space ^[1]. The last of those forms decides how the class's loan record has to be read. A fenced lot where boats, trailers and motor homes are parked by the month is self-storage under this definition, and the dataset's asset map is explicit about what follows: the vehicle storage trade has no separate code, so its loans are coded either here or to other warehousing (493190), and those coded here sit inside this class and cannot be pulled back out. Every class figure in this article is therefore the combined record of enclosed storage buildings, outdoor vehicle storage and facilities that offer both, and none of it can be split by product. The series method for RV and boat storage reads the same cells from the other side of the shared code.

The boundary on the other side runs to warehousing. NAICS places public and contract warehousing and storage, other than self-storage, in Industry Group 4931 ^[2], and the dataset holds general warehousing and cold storage as classes of their own. A facility that also lets business bays on longer terms, or that runs a records or wine vault under contract, earns part of its income outside the class. A study for such a project should split revenue by activity, so that a lender who sets the project beside the class record compares like with like. Why a shared code can be described but never divided is the subject of From NAICS Code to Asset Class: How SBA Loans Are Assigned, and Where Two Industries Share One Code.

The code also fixes the size test. SBA's size standard for 531130 is $34.0 million in annual receipts ^[3]. A 7(a) or 504 applicant above that line may still qualify under the alternative standard of 13 CFR 121.301(b)(2), which opens, ahead of its instructions on computing the tax figures, with this test: "Including its affiliates, tangible net worth not in excess of $20 million, and average net income after Federal income taxes (excluding any carry over losses) for the preceding two completed fiscal years not in excess of $6.5 million" ^[4]. Both measures take in affiliates, so for an owner who holds each facility in its own entity the test is run on the group, not on the subject alone. A study should list every facility the sponsor and its owners control, with the entity that holds each one, so the lender can run the size test and the affiliate review from a single schedule.

The Demand Model: Households, Businesses and the Space They Occupy

Storage demand is a stock of occupied space rather than a flow of visits, and the model that fits a stock is a penetration model: count the households and businesses in the trade area, apply the share of each group that rents storage, and multiply by the space each renter occupies. Written out, with every term an input a lender can see and change:

1. Households: H = P / s, where P is the trade area's resident population and s its average household size.
2. Households by tenure: H_o = H x t and H_r = H x (1 - t), where t is the owner share of households.
3. Households renting storage: N = H_o x u_o + H_r x u_r, where u_o and u_r are the shares of owner and renter households that rent space.
4. Resident demand: D_res = N x a, where a is the average occupied area per renting household.
5. Business demand: D_bus = E x u_b x a_b, where E counts establishments, u_b is the share of them that rent space and a_b the area each occupies.
6. Occupied space the trade area supports: D = D_res + D_bus.
7. Implied market occupancy: D / S against the existing competitive supply S, and D / (S + Q) once the subject's net rentable area Q is added.

Public data carries the first terms. Population comes from the Census Bureau's estimates, or from the American Community Survey for the tracts and block groups that make up the drive-time area, and household size and tenure come from the same survey; the series piece on trade areas covers how those small-area figures are assembled and why their margins of error travel with them. Household size and tenure both move the answer, and tenure differs by region. For the third quarter of 2026 the Census Bureau's Housing Vacancy Survey puts the national homeownership rate at 65.5 percent ^[5] and the West region's at 60.6 percent ^[6]. The model below uses the national figure for its base column and the western one for a comparison, as stand-ins for the survey figures of a real trade area, which a study would use in their place.

Business demand starts from a count of establishments. Establishments with a payroll appear in County Business Patterns ^[7], and those without one appear in the Census Bureau's Nonemployer Statistics ^[8], so a study needs both to count the contractors, small shops and home businesses that rent units. The supply article explains what each series leaves out.

The remaining terms, the use rates and the space per renter, have no public series and come from the field. Take the occupied square feet the field census records in each competing facility, divide the total among the trade area's households and businesses, and the result is the use the market already shows; the comparables' unit mix gives the average space per renter. A study should print that calibration with its date rather than borrow use rates from a national survey, because a national rate knows nothing of the local mix of apartments, small lots and businesses.

The inputs for both models in this article are in Table 1. The first twelve rows, with the subject's size in the thirteenth, feed the demand model; the rest belong to the worked example near the end.

Rounded as shown; lines may not foot by one unit in the last place.

In the base column of Table 2, 64,000 residents in households of 2.5 persons make 25,600 households, of which 3,267 rent storage. Their resident demand is 408,320 occupied square feet, business users add 17,000, and the trade area supports 425,320 square feet in all. Set against the 470,000 square feet the field census counts, that implies market occupancy of 90.5 percent, close to the balanced occupancy the study adopts. Adding the subject's 52,000 square feet pulls implied occupancy down to 81.5 percent. That last figure is the one to carry forward: unless the subject wins customers from older facilities or the trade area grows, the market as a whole cannot fill a new building to the stabilised occupancy the worked example assumes later in this article.

Square Feet per Capita Is an Output, Not a Benchmark

Divide a trade area's competitive storage supply by the people who live in it and the result is square feet per capita; the site's archived ranking of metropolitan storage markets by that ratio and by rent growth treats it as a saturation measure. The demand model shows why the ratio cannot carry a lending decision alone. In Table 2 the first three columns share one population and the same 470,000 square feet of supply, so the ratio is 7.34 square feet per resident in each of them, and yet they lead to opposite conclusions.

• Base. Demand per resident is 6.65 square feet, and the ratio at which the market would sit exactly at its balanced occupancy is 7.38. Existing supply is almost in balance: the market could take 2,578 more square feet before falling below it.
• West region tenure. With more renter households, demand per resident rises to 6.84 and the balanced ratio to 7.60. Implied occupancy is 93.2 percent, and the market has room for 16,516 square feet.
• Larger households. The same residents in fewer, larger households rent less space: demand per resident falls to 5.77, the balanced ratio to 6.41 and implied occupancy to 78.5 percent. This market is oversupplied by 60,000 square feet before the subject opens.

The fourth column moves population instead. With residents and households both up and supply unchanged, the ratio falls to 6.68, implied occupancy rises to 99.2 percent and the market could take 47,947 more square feet. Growth is therefore an input a study has to source: the Census estimates for the years already past, and a named, dated forecast for the years ahead.

Three lessons follow for a credit officer. First, the benchmark that matters is the trade area's own balanced ratio, which is an output of the demand model, not a constant brought in from a national ranking. A market at 7.34 square feet per resident can be short of space, or long by 60,000 square feet, depending on who lives in it. Second, a ratio is only as good as its two counts. None of the federal series this article cites counts rentable storage space: the construction series below measures dollars put in place ^[13] and the business series count establishments, so the numerator of any per capita figure comes from a field census or a private inventory, and its source and date belong beside it. Third, ratios compare trade areas only when both are drawn the same way, a trap that the supply article works through with a boundary example. Read in that light, the ratio is a useful screen between candidate sites and a poor substitute for the model.

The Federal Record of New Storage Construction

The Census Bureau's construction spending survey carries a line for storage building inside its warehouse category: the mini-storage line includes mini-storage centers and self storage centers ^[14]. Its annual tables give the national value of private mini-storage construction put in place, and Table 3 prints the line for every year from 2012 to 2025.

Read down the column and the last building cycle is plain. Spending on mini-storage construction was $365 million in 2012 and $4,893 million in 2018 in the earlier table ^[15], and it reached $7,836 million in 2023 before easing to $6,359 million in 2025 in the later one ^[16]. Three cautions travel with the series. Its unit is dollars put in place ^[13], so a rise can mean more building, dearer building or both, and the table cannot say which. The class line is a small and noisy estimate: the Census Bureau puts its coefficient of variation at 5 percent, against 1.9 percent for general commercial warehouses ^[17]. And the two halves come from separate annual releases, printed here as published; a study that joins them into one series should say that it has done so.

What the series cannot do is locate a project. A national total says nothing about whether a facility is rising two miles from the subject; the local pipeline still has to come from planning and permit records, staged as the supply article describes. The interagency real estate lending guidelines list current and projected construction rates among the market factors an institution should consider ^[18], and a national line that has run several times higher than its 2012 level gives a reviewer reason to ask when each comparable opened, what opened near it in the same years, and whether its lease-up history was earned against unusual competition.

Lease-Up: Occupancy and Rent Rise on Separate Schedules

A new storage facility opens empty and fills one rental at a time, and its revenue in any month is the product of two curves rather than one. The first is physical occupancy, the share of net rentable area, or of units, under rental. The second is the rent those occupied units actually pay, as a share of the street rate the facility quotes to new customers. Opening discounts and free months hold the second curve down early, and in this article's model it climbs as introductory terms run off. A study that runs only the occupancy curve, at full street rates, books revenue that any opening discounts will keep the facility from collecting. The worked example runs both: physical occupancy averages 40 percent in the first year of operation, 70 percent in the second and 88 percent once stabilised, while in-place rent runs at 80, 90 and 95 percent of the street rate, as Table 1 records.

The evidence for both curves comes from facilities that have already leased up, ranked by weight in the way the ramp-up article ranks it. For storage the strongest evidence is the rent roll history of comparable facilities that opened in the trade area, or in a similar one, in recent years: occupied area and in-place rent at each month end from opening to stabilisation, with the promotions offered at each stage. Interviews with the operators of those facilities come next, and a record of move-ins and move-outs shows whether occupancy was bought with discounts. The federal absorption survey is no help for this class, because it samples only residential buildings with five or more units ^[19]. A study should also count the facilities that will be leasing up at the same time as the subject, since each draws on the same new demand.

Stabilisation needs a definition the reader can test. A study should state it as an occupancy held for a stated period, show that the comparables reached it and how long they took, and give the month in which the subject is expected to reach it. Projected vacancy and absorption sit on the guidelines' list of market factors ^[18]; in a storage study they take the concrete form of that dated month.

Lease-up is also where a storage loan meets the coverage test hardest.

Conventions. This article follows the conventions of the Loan Analytics worked examples. Coverage is tested against the program floor the site's program page states, where the program has one, and against a stated 1.25x lender assumption, and the higher of the two binds. Equity is measured at the loan the cash flow supports, with fees recomputed on that loan. An acquisition is tested on the seller's historical earnings, not on projections.

In the base case of the worked example, coverage is 0.08x in the first year of operation and 0.80x in the second, so debt service has to be met from outside the facility's income until it stabilises, and the operating shortfall over those two years comes to $354,413 before any interest during construction. A lender should see who carries it, whether an interest reserve inside the loan, sponsor cash set aside at closing or support from a guarantor, and should see it sized against the slower lease-up case, which raises the shortfall to $576,133.

The Metrics a Credit Officer Reads on a Storage Loan

The interagency guidelines on real estate lending list the market supply and demand factors an institution should consider, among them current and projected vacancy, construction and absorption rates, lease terms and rental rates including concessions, and operating expenses for different types of projects ^[18]. For a storage facility each factor takes a particular form, is proved by a particular document and can be overstated in a particular way. Table 4 sets them out.

The gap between the two occupancies is the first thing to check. A facility can report high physical occupancy because it filled units at deep introductory rents, and its collected revenue then tells the truer story. In the worked example the stabilised in-place share of 95 percent, applied to physical occupancy of 88 percent, carries as much weight as the street rate, because stabilised revenue moves one for one with either, and a study should show the comparables' own ratio of collected to potential rent before adopting it. A street rate means little without the unit size it applies to: a study should weight the comparables' quoted rates by the subject's own unit mix and print the mix.

In storage the remedy against a customer who stops paying is set by state statute, so delinquency runs on a legal timetable a study can read. Florida's lien statute, as one example, requires the notice to a tenant in default to demand payment within a specified time of not less than 14 days after delivery ^[20], then requires the sale to be advertised once a week for 2 consecutive weeks in a newspaper of general circulation where the facility is located ^[24], and lets the sale take place no sooner than 15 days after the first publication ^[25]. On that newspaper route a sale cannot follow the notice by fewer than 29 days, and the unit earns nothing in the meantime. The statute also allows the sale to be held on a public website that customarily conducts personal property auctions ^[26]. A study should take its collection-loss assumption from the subject state's statute and the comparables' records, not from another state's timetable.

Value is the appraiser's question, not the study's, as the article on feasibility studies and appraisals explains. The supervisory loan-to-value ceilings a bank works within stand at 80 percent while the loan finances construction of commercial or other nonresidential property, and at 85 percent for an improved property loan ^[22]. A study helps that test only indirectly, by supporting the net operating income an appraiser capitalises.

Eligibility Under 13 CFR 120: Operating Business or Passive Landlord

The first SBA question for a storage borrower is whether it counts as a business at all. Of the types 13 CFR 120.110 shuts out, the third on the list, paragraph (c), is "Passive businesses owned by developers and landlords that do not actively use or occupy the assets acquired or improved with the loan proceeds"; paragraph (c) itself then carves out Eligible Passive Companies ^[27]. Part 120 never names self-storage, so the question for the class is the general one: does the borrower use the facility to run a business, or own it and collect rent? The loan record answers only the narrow question of fact, that the programs have approved loans to the class in number, 1,584 ^[28] 7(a) and 946 ^[29] 504 approvals over fiscal years 2017 to 2026. It does not record the reasoning on which each borrower was found eligible, and a study should not offer reasoning of its own as if it were SBA's. A study's part is to lay out the operating facts on which the lender, the CDC and SBA decide: who manages the facility day to day, what the business sells besides space, how its rental agreements are written and how it collects.

Part 120 measures occupancy in Rentable Property, which it defines in these terms: "the total square footage of all buildings or facilities used for business operations" ^[30]. When the SBA loan pays for a new building, at least 60 percent of that space must be permanently occupied and used by the borrower and at most 20 percent may be let permanently to tenants ^[31]. The borrower must also hold plans "to permanently occupy and use within three years some of the remaining space not immediately occupied and not permanently leased" ^[32], and a ten-year plan for "all of the remaining space not permanently leased" ^[33]; an existing building moves the thresholds to 51 percent occupied and as much as 49 percent leased ^[34]. Those rules were written for a business that occupies part of a building and lets the rest. A storage operator's rentals are its business, and Florida's statute calls the storage customer a tenant ^[20]. Whether SBA counts space under customer rental as space the borrower uses in its business, or as space leased to others, is not settled by the text of Part 120, and this article does not settle it. A study should describe the space so that either reading can be applied: the area under customer rental agreements, any area let on longer terms to other businesses, such as a retail counter or a set of warehouse bays, and the area the operator uses for its own office and stock.

Where the real estate sits in one entity and the storage business in another, the loan runs through Part 120's passive company route. The property entity owns the facility and lets it to the Operating Company that runs the storage business, and it may not itself engage in "regular and continuous business activity" ^[35]. Two of the route's conditions reach straight into a study's numbers. One caps the intercompany rent: "The rent or lease payments cannot exceed the amount necessary to make the loan payment to the lender, and an additional amount to cover the Eligible Passive Company's direct expenses of holding the property, such as maintenance, insurance and property taxes" ^[36]. A rent pinned to the loan in that way proves nothing about repayment, so the study should measure coverage at the operating company, where customers pay for their units. The other condition sets the lease's length, which, "including options to renew exercisable solely by the Operating Company, must have a remaining term at least equal to the term of the loan" ^[37]. And when the property entity lets all of a new building to the operating side, the operating company, or the operating companies together, carry the occupancy tests ^[38].

In the 504 program the borrower's own cash steps up with two conditions, a short operating history and a special-purpose building. With neither, the floor is 10 percent; with either, 15 percent; with both, 20 percent ^{[39] [40]}. The history condition reaches a borrower, or the operating company behind a passive borrower, that has run for two years or less; the building condition is met by any "acquisition, construction, conversion, or expansion of a limited or single purpose building or structure" ^[40]. Whether a storage building falls in that category is for the CDC and SBA to classify. A study should state which contribution its sources and uses assume, and why, without treating the answer as settled; the 504 method article and the article on equity injection cover the rest of the structure.

B&I, the Rural Test and the State Lien Statutes

Outside the SBA programs, a rural storage facility may look to the USDA Business and Industry guarantee, whose tests are set out in the B&I method article. The rural test there starts from a population line: a place of more than 50,000 inhabitants falls outside it, and so, as a general rule, does the urbanized area contiguous to such a town ^[41]. That second exclusion has an exception. The Under Secretary of Rural Development may find a piece of such a fringe rural in character, and a piece with that finding stays inside the definition ^[42]; each finding is made for a particular project ^[43]. A study for a storage site at the edge of a larger town should therefore cite the determination, or state that none has been requested, rather than read the site in or out of the program from a map. Two of the ineligible project rules in 7 CFR 5001.115 bear on storage. One shuts out speculative real estate investment, together with investment or arbitrage generally, subject to stated exceptions ^[44]. The other shuts out properties whose intended use is "primarily commercial rental when the borrower has no control over tenants and services offered", unless the project is industrial-site infrastructure development ^[45]. That second rule puts, from the USDA side, a question close to SBA's passive-business test. A study for a B&I storage loan should therefore document the borrower's control in concrete terms: who sets the rental terms, who accepts and removes customers, who staffs the office and who provides each service sold. The first rule is answered by demand evidence. A project shown to fill on the strength of a calibrated demand model and dated comparables is harder to read as a bet on land or on future growth alone.

At the state level, the rules a study has to read are the self-storage lien statutes, which govern what an operator may do when a customer stops paying. Florida's, used above, also carries a rule that matters for the vehicle storage sharing the class's code: when a motor vehicle or watercraft is held and its "rent and other charges related to the property remain unpaid or unsatisfied for 60 days after the maturity of the obligation", the owner may either sell it by the statute's procedure or arrange for it to be towed ^[46]. A facility that mixes enclosed units with boat and RV spaces therefore runs two collection clocks in one rent roll, and its loss assumption should be built for each. This article has not surveyed every state's statute, and a study should cite the one that governs the subject, together with any state or local registration, zoning or fire code approval the facility needs to open, each taken from the issuing authority's own text.

The Class's Record in the SBA Program Cells

Over fiscal years 2017 to 2026 the class holds 1,584 ^[28] 7(a) approvals worth $2,313,999,900 ^[28] and 946 ^[29] 504 approvals whose debentures total $1,218,821,000 ^[29]. The 504 dollars are the CDC debenture alone, so they understate the projects those loans helped finance. The median 7(a) loan is $975,000 ^[28]; a quarter of the class's 7(a) loans are below $450,375 ^[28] and a quarter above $2,000,000 ^[28]. The median 504 debenture is $890,000 ^[29], with its lower quartile at $432,500 ^[29] and its upper quartile at $1,710,750 ^[29]. The 7(a) approvals are spread across 47 ^[28] states or territories, and the 504 approvals across 47 ^[29]. Every count includes approvals later cancelled, because the dataset counts gross approvals.

Few approvals in the class carry a franchise code: 6.82 percent ^[28] of its 7(a) approvals do, and 0.74 percent ^[29] of its 504 approvals. For a study the practical consequence is that the borrower's own operating record, or the record of a named third-party manager, has to supply the evidence of operating competence that a franchise system's documents would supply in a branded class.

The yearly series moves a long way. The class's 7(a) approvals numbered 172 ^[28] in FY2017, 296 ^[28] in FY2021, 124 ^[28] in FY2023 and 109 ^[28] in FY2026; its 504 approvals numbered 71 ^[29], 181 ^[29], 91 ^[29] and 59 ^[29] in the same years. Across the whole 7(a) program, approvals went from 51,856 ^[47] in FY2021 to 56,130 ^[48] in FY2026, so the class's 7(a) count fell over a span in which the program's rose, while 504 approvals as a whole went from 9,676 ^[49] to 6,075 ^[50] and the class's 504 count fell with them. The construction line in Table 3 peaked in dollars in calendar 2023, while the class's approvals in both programs peaked in FY2021. Nothing in the dataset explains these moves, and this article puts forward no explanation of its own; a study should not cite the class's approval trend as evidence of local demand or of lenders' appetite for the class.

Because the code is shared, every figure in this section includes RV and boat storage and cannot be read as the record of enclosed storage alone. The program pages for the class, self-storage 7(a) loan data and self-storage 504 loan data, carry the same cells, and How to Read a Loan-Outcome Page: Counts, Floors, Ranks and the Keys Behind Every Figure explains each figure on them.

What the Charge-Off and Cancellation Floors Show, and Where They Stop

Take the class's disbursed 7(a) loans from the approvals of fiscal years 2017 to 2021. By the end of September 2026, 4 ^[28] of the 806 ^[28] had been charged off. That makes the class's approval-cohort floor 0.50 percent ^[28]; across the entire 7(a) book of those approval years the floor is 6.33 percent ^[51], more than ten times as high. Two narrower measures point the same way. Counting only the charge-offs that arrived within five years of approval puts every loan on an equal footing, and on that count the storage class sits at 0.12 percent ^[28], the program as a whole at 3.72 percent ^[52]. In dollars, charge-offs so far come to 0.04 percent ^[28] of the class's approved amount and 1.95 percent ^[53] of the program's. In 504 the class shows 0 ^[29] charge-offs among 412 ^[29] disbursed loans from those approval years, so its floor is 0.00 percent ^[29], where the 504 program as a whole shows 0.35 percent ^[54].

They are floors because a disbursed loan that is still running shows no status in the public file ^[55]. Open loans numbered 250 ^[28] inside the class's 7(a) denominator and 267 ^[29] inside its 504 denominator. Each of those can still end in a charge-off, so the class's floors can move up and never down. Why a published loss rate built this way can only understate, and how the cohort and the horizon are chosen, is the business of The Approval-Cohort Charge-Off Floor: Why a Published SBA Loss Rate Can Only Understate.

The floors record what happened to a set of approved loans; they do not say why. No cause for the low floors appears in the report that documents the dataset, and this article supplies none. Three limits follow for anyone reading them into a single loan. A few charge-offs are a thin base for any rate. The cohort is a set of loans approved in particular years and observed through particular conditions, and a facility approved into today's market is not in it. And the code mixes enclosed buildings with outdoor vehicle lots, so the floor belongs to neither product alone. A low class floor is the record of a population of loans, each of which passed its own underwriting; a lender should not accept it as a substitute for the subject's demand, lease-up and coverage evidence, and a study should not offer it as one. The failure a study has to test for is the one the cohort record cannot show for a new project: a lease-up slower than planned, a street rate lower than surveyed, or a pipeline that opens alongside the subject.

Cancellations tell a different story in the two programs. For the approval cohort of fiscal years 2017 to 2024, the class's 7(a) cancellation floor is 11.14 percent ^[28], close to the 11.24 percent ^[56] for the program. In 504 the gap is wider. Against a program floor of 13.60 percent ^[57], the storage class's floor reads 19.15 percent ^[29], and the class's ceiling, which counts the 38 ^[29] approvals still waiting to disburse as if they had cancelled, is 23.76 percent ^[29]. A cancelled approval never reached disbursement, and the file is silent on whether its project was abandoned, financed elsewhere or built with other money; Cancellation Is Not Charge-Off: The Two Ways an SBA Approval Can End Without Repayment explains what the status can and cannot carry. For a storage study the 504 figure is a prompt rather than a finding: a reason to show the lender that the site, its approvals and the cost estimate are firm enough for the approval to fund.

Every loan figure used above is collected in Table 5, with its basis and its key.

Loan Data Behind This Article

Every figure in this table comes from the Loan Analytics loan dataset, analytics-loan-data freeze-3 (commit 360c856), and carries its key. SBA rates are approval-cohort floors and are withheld when fewer than 30 loans stand behind them.

A Mini-Example: One New Facility Through Its First Three Years

The example is a ground-up facility with 52,000 net rentable square feet. Its 7(a) financing, $2,950,000, sits well short of the program's ceiling on a single loan, which is $5,000,000 ^[58]. Every market and cost input is an assumption listed in Table 1, chosen to show the mechanics rather than to describe a site. Interest is the prime rate, 7.00 percent on the H.15 release covering the opening days of November 2026 ^[9], plus an assumed margin of 2.75 points. The variable-rate ceiling sits 3.0 points over the base rate, and the rule applies it to every 7(a) loan above $350,000 ^[10], so the assumed margin fits beneath it. Repayment is level and monthly over 25 years, the most the rule allows for real property before any construction period is added ^[11], and carries no balloon, which SBA does not allow ^[59].

Table 6 carries the facility through its first two years of operation and into its stabilised year, then moves one input at a time.

In the base column, stabilised net operating income is $405,059 against annual debt service of $315,463, coverage of 1.28x. Two minimums apply: the 1.25x lender assumption, and the 1.15x floor that the site's program page attaches to standard 7(a) loans ^[12]. The assumption is the stricter of the two, so it governs, and at that level the stabilised income would carry a loan of $3,030,276, a little more than the loan assumed. Break-even physical occupancy at stabilised rents is 75.2 percent, and the occupancy needed for the binding minimum is 86.5 percent, less than two points under the 88 percent the example assumes.

The lease-up years are the weak point. Coverage is 0.08x in the first year of operation and 0.80x in the second, and the shortfall to be carried over those two years is $354,413. The site's SBA program page puts the startup rule under SOP 50 10 8.1 in these words: "standard 7(a) loans need at least 1.15x business coverage, and startups must reach it within two years of funding" ^[12]. With the assumed twelve-month build, only 12 months of operation fall inside that window, and the example's coverage over them is 0.08x. Read against the statement as the site's page gives it, the base case does not reach the floor inside the window. A study for such a project should say so plainly, show what would change the answer (a shorter build, a phased opening or a smaller first loan), and leave the structure to the lender.

The shock columns show which assumptions carry the loan. A street rate one tenth lower cuts stabilised coverage to 1.09x and the supportable loan to $2,568,051, and lifts the occupancy needed for the binding minimum to 96.1 percent. A slower lease-up leaves stabilised coverage where it was but takes first-year net operating income below zero and raises the shortfall to carry to $576,133. Stabilising five points lower leaves coverage at 1.17x, which clears the program floor and misses the lender assumption, and supports $2,767,648. Two points more on prime brings coverage to 1.11x and the supportable loan to $2,609,583, and moves break-even occupancy to 82.5 percent. How large each shock should be, and why, is set out in Sensitivity and Stress Testing: Break-Even Points, Shock Sizes and What a Lender Should See.

The demand model points to the column with the lower stabilisation. In the base trade area of Table 2, implied market occupancy with the subject added is 81.5 percent, well under the 88 percent the lease-up assumes for the subject. A study can close that gap only with evidence: comparables showing that newer facilities in the area hold occupancy above the market average at the expense of older ones, or a sourced population forecast of the kind the growth column tests. Without that evidence, the lower stabilisation is the case the study should present, and the case the loan should be sized on.

Two pointers complete the example. The full version, with a 504 structure, a sourced market and a complete sources and uses, is Worked Example: SBA 504 RV and Boat Storage Feasibility Study in New Jersey. And the purchase of an existing facility is a different test: the site's program page states that initial acquisitions need 1.25x on historical earnings ^[60], a program test the page reports and a separate matter from the 1.25x lender assumption of this example. Buying a Business With a Loan: Why Coverage Is Tested on Historical Earnings, Not the Buyer's Plan explains why an acquisition is sized on what the business has earned, which for a facility bought before it stabilises means its trailing results rather than its lease-up projection.

Questions to Put to a Self-Storage Study

A reviewer can test a storage study quickly against the questions below. Each should already be answered on the study's own pages.

• The code. Does the study say whether the facility is enclosed storage, outdoor vehicle storage or both, and does it separate any income earned in warehousing or other activities outside 531130 ^[2]?
• The demand model. Are population, household size and tenure taken from Census sources for the trade area as drawn, and are the use rates and space per renter calibrated against the field census, with the calibration printed and dated?
• The ratio. If square feet per capita is printed, is the trade area's own balanced ratio printed beside it, and were the supply count and the population drawn on the same map and dated?
• The pipeline. Which planning and permit records were searched, on what date, and is the national construction record ^[16] used only as context, never as a local count?
• The lease-up. Are physical occupancy and in-place rent modelled separately, each from named comparables' histories, and is stabilisation defined and given a month?
• The collections. Does the delinquency and collection-loss assumption follow the lien statute of the subject's state ^[20], with a separate clock for any vehicles and boats?
• Eligibility. Are the facts that bear on the passive-business test ^[27] set out, including the entity structure, the intercompany lease and the services the business sells?
• Coverage. Is coverage shown for each lease-up year as well as at stabilisation, tested against whichever is stricter, the program's floor or the lender's own minimum, and does the study name who funds the shortfall while repayment depends on it ^[21]?
• The loan record. Is every class rate printed as a floor, with its cohort, and is the class record presented as context, never as evidence that the subject will perform?

Where a storage study leaves one of the nine open, the reviewer's next steps are concrete. Pull the rent rolls of two or three comparable facilities and set their collected rent against their quoted street rates. Ask the planning office for every storage project filed since the study's search date. Then rerun the lease-up table at the occupancy the demand model implies once the subject is added, and read off the loan that occupancy supports. Each of those checks can change the size of the loan, and each can be finished before the credit memorandum is written.

Related Pages on Loan Analytics

• The Loan Analytics methodology series (all articles)
• How a feasibility study is done: the Loan Analytics methodology
• SBA feasibility studies (7(a) and 504)
• USDA feasibility studies
• Bankable feasibility studies for conventional loans
• Self-Storage Feasibility Study
• Loan data by state
• Loan data by asset class
• Worked examples (case studies)
• Worked Example: SBA 504 RV and Boat Storage Feasibility Study in New Jersey
• Self‑Storage Saturation Index (Supply per Capita vs Rent Growth)
• Self-storage development feasibility in 2026

Commission a Feasibility Study

Loan Analytics prepares lender-grade feasibility studies for SBA 7(a), SBA 504, USDA Business and Industry, conventional and EB-5 financing, built on the methods this series sets out. A study is not an appraisal: it tests whether a project's market, costs and cash flow can carry the debt a lender is asked to approve. To discuss a study for your project, contact Loan Analytics.

Daniel Smith, MAI, Loan Analytics