Food Hall Feasibility Study for SBA, USDA and Bank Loans

A food hall is multi-tenant real estate with an operating overlay, and the lender's borrower is the operator who holds the master lease and the bar. MMCG Invest prepares food hall feasibility studies for SBA 7(a), SBA 504, USDA Business and Industry and conventional financing that underwrite the operator on bar and common-area cash flow with vendor license income haircut for turnover, test demand by daypart against office occupancy rather than against population alone, place the stall economics on the only published worked example, confirm the occupancy rules that decide SBA eligibility, and report the debt service coverage ratio (DSCR) by year with the bar ramp excluded from year one.

The count is growing and the closures are concentrated. Colicchio Consulting counted 458 operating food halls in January 2026 against 343 in early 2023, with at least 114 in the pipeline, and the 2019 projection of 450 by the end of 2020 arrived about five years late. The closures cluster in high-rent urban office districts: Time Out Market closed Boston and Chicago in January 2026 after seven years, displacing 15 and 19 vendors; New York lost Citizens Market Hall, Northend Food Hall and Harry's Table in 2025; The Heights in Chevy Chase closed after twenty months; Seattle's Asean StrEAT was evicted in February 2026 owing $842,209 on a $26,000-a-month lease it had not paid since 2023. The reason one operator gave was "a $23 check average with no alcohol, and the rents that landlords are charging," and early New York halls did 80 percent of their business between 11 a.m. and 2 p.m. on weekdays. The format is a daypart business inside a rent structure, and the study is written around both. The general method and the SOP 50 10 8.1 spine are on the restaurant feasibility study hub; the food truck pod, which is the outdoor cousin of this format, is on the food truck feasibility study page.

The study states which model the subject is, because the borrower changes with it. In the master-lease model the operator leases the shell, builds out, licenses stalls to vendors on one- to three-year agreements, and holds the liquor license; the operator typically pays about half the per-square-foot rent it collects from vendors, and the bar is its margin. In the vertically integrated model one company runs every stall. In the licensed-brand model brands are licensed to and run by a contract food service partner, as at Citizens. A landlord-run hall is the master-lease model with the landlord in the operator's chair, and a fee-managed variant pays the operator a start-up fee, a base management fee and an incentive fee on EBITDA. For SBA purposes the question is whether the borrower is an operating business or a passive lessor: a hall where the operator runs the bar and the common operations supports the operating-business reading; a pure sub-license model does not. For a 504 loan the borrower must occupy 51 percent of an existing building or 60 percent of new construction under 13 CFR 120.131, and the study reports the occupancy arithmetic with the vendor stalls counted as it will be counted. The program rules are on the SBA 7(a) feasibility study and SBA 504 feasibility study pages.

The only first-party worked example in the public record is Politan Row's, and it is the operator's own, so the study uses its structure and tests its numbers. A vendor stall of about 200 square feet does about $650,000 a year, runs labor at 26 percent, food at 24 percent and disposables at 3 percent, pays the hall 28.4 percent of sales in rent, common area and utilities, about $185,000, and nets 18.6 percent. Percentage rent across the industry runs 10 to 30 percent of sales, with some operators quoting 8 to 15 percent plus base rent and others a commission near 30 percent. A stall costs the vendor under $50,000 to open by the operator's account and $75,000 to $200,000 by other reporting. Without a bar the operator nets about 9 percent of collections, about $166,500 on ten vendors; the bar runs 15 to 35 percent of hall volume and about a quarter of cash flow, and with a bar at 22 percent of an $8.3 million hall the operator's net rises to about $616,500. The study carries vendor license income at the published occupancy charge, haircuts it for turnover of about 30 percent a year with vacancy and re-tenanting cost, and gives the bar no ramp credit in year one.

A food hall's trade area is the office and the event, not the rooftop. The study reports daytime population and office occupancy within the walk shed, the residential base within the drive time, the visitor and event generators, and the daypart split the format produces, because a hall that earns its rent between 11 a.m. and 2 p.m. on weekdays is exposed to office attendance in a way a neighborhood restaurant is not. Competitive supply is every hall, food court and fast-casual cluster in the walk shed, plus the permitted pipeline, which at 114 halls nationally is a quarter of the operating count. Openings in 2026 ran from Omakase World Market in South San Francisco and the Mess Hall at the Presidio to the 35,000 square foot Shaver Hall in the former Lord and Taylor building in New York, and the study treats a hall under construction nearby as supply from its expected opening date.

A food hall is a shell build-out with ten to twenty kitchens in it, and the cost follows the restaurant build-out ranges on the independent full-service restaurant feasibility study page multiplied by the kitchen count: $125 to $250 per square foot for second-generation space, $250 to $500 and above for cold shell, before the vendor's own equipment. The operator's capital is the shared infrastructure, grease interception, hood systems, restrooms, the bar and the seating, and the study separates what the operator funds from what each vendor funds. A hall in leased space is a 7(a) project at the 1.15 times floor with the master lease matching the loan term including borrower-only options; a hall whose operator buys the building is a 504 project, and a food hall is not on SBA's special-purpose list, so an established operator contributes 10 percent and a new business 15 percent. A rural hall in a town under 50,000 people is a USDA Business and Industry candidate on the terms on the USDA feasibility study page, with the new-business equity test of 20 percent balance sheet equity or 25 percent of project cost and an independent feasibility study above $1 million. The liquor license, which the operator holds, is covered on the bar and nightclub feasibility study page.

The study reports DSCR as EBITDA over total debt service by year against the program floor, 1.15 times on a standard 7(a) loan, 1.15 times on historical EBITDA for a 504, the lender's covenant under USDA, with the bar excluded from year one and vendor income haircut for turnover. The sensitivities are the ones the record supports: vendor vacancy at two and four stalls of ten, percentage rent at 20 and 25 percent of sales where the base case uses 28.4, office occupancy at 10 and 20 percent below current, bar volume at 15 percent of hall sales, wages 10 percent up and rates 100 basis points up. The master-lease operator's own rent is the fixed line that failed every named closure, and the study shows the hall's break-even in vendor count and in bar sales.

The operator's collateral is thin. Leasehold improvements belong to the landlord at lease end, shared kitchen infrastructure has little value to a next user, vendor licenses are short and terminable, and the liquor license is worth the state fee outside quota states. The loan is a cash-flow loan on the operator's bar and common-area margin, and the study says so.

A MMCG food hall study includes the operating model determination and the SBA occupancy arithmetic, the walk-shed and daypart demand analysis with office occupancy, the competitive census of halls and fast-casual clusters with the pipeline, the stall economics on the published structure with the turnover haircut, the bar projection, the project cost split between operator and vendor, the master lease test against the loan term, the capital stack by program, the DSCR schedule with the vendor-count and bar break-evens and sensitivities, the collateral discussion and a signed conclusion. Standard delivery is nine to sixteen business days; expedited delivery in five to seven is available. Fees begin at $4,900 for a single-site 7(a) study and run $7,500 to $15,000 for a 504 or multi-stall hall. Revisions required by the lender or agency are made at no additional cost under MMCG's written acceptance guarantee. See MMCG's feasibility study methodology and where we work.

The food hall format shares its structure with the competitive socializing venue case, SBA 504, an independent venue with a full bar underwritten on venue-level cash flow, and with the stall-and-bar economics above.