The Most-Financed Business in America Is One of Its Worst to Own

A word on scope

This analysis is built from primary-source public data, joined at the industry level: SBA 7(a) loan records (FOIA, as of March 31, 2026), IRS Statistics of Income (tax year 2023), the Bureau of Labor Statistics' JOLTS series, NYU Stern's Damodaran dataset (2024), the Bureau of Economic Analysis (2024), and the Federal Reserve's H.15 release. It covers Main Street and lower-middle-market acquisitions — the businesses ordinary buyers actually finance, mostly under $5M.

Two of the figures below (the margin and rent ratios) are computed from IRS sole-proprietor tables and describe small Schedule C filers, so they are used as a ranking, not as the exact economics of any one deal. The directional conclusions are sturdy; any single decimal is a signpost, not a survey. Full sourcing and its limits are at the end.

What business do Americans borrow hardest to buy?

Full-service restaurants, by a wide margin. The restaurant is the easiest acquisition for a lender to underwrite: the model is familiar, the equipment list, lease, liquor license, and POS reports are collateralizable and legible, and the file papers cleanly. Approvals follow.

The SBA's 7(a) records are the cleanest available map of where acquisition capital goes. Filtered to change-of-ownership loans only — the ones that fund an actual purchase rather than working capital or new equipment — one category sits on top: full-service restaurants, 7,743 funded acquisition loans pooled from 2018 through early 2026, about 38% ahead of the next category (liquor stores), which leads auto repair. More Americans borrow to buy a restaurant than to buy any other kind of business.

What are they actually buying?

A thin-margin business with unstable labor and a landlord paid first. A restaurant keeps about seven and a half cents of profit on every dollar of sales. Only retail does worse, at six and a half. A professional-services firm — an accounting practice, an engineering shop, a consultancy — keeps around forty-three cents.

These figures come from IRS sole-proprietor returns for 2023, so they describe small operators rather than a two-million-dollar target, and are best read as a ranking: a solo consultant nets high partly because there is no payroll. The order is the point — food sits at the bottom of the profitability stack and knowledge work at the top, across every cut of the data.

A thin margin is survivable only if little goes wrong, and in restaurants the largest variable cost is also the least stable. In 2025, food-service workers quit at 3.9% a month, the highest rate the BLS tracks and nearly double the 2.0% all-industry pace; in 2023 it was 4.8%. At that rate a restaurant turns over close to half its staff in a year. The cost of that churn — continuous hiring, training, and the gaps between a departure and a replacement — does not appear on the seller's add-back schedule.

Rent compounds the problem. Food service pays about 5.3 cents of every sales dollar in rent, second-highest of any sector. Against the 7.6 cents the owner keeps, the landlord earns nearly as much from the restaurant as its owner does, carries none of the operating risk, and is paid first.

Does "the bank approved it" mean it's a good business?

No — those are two different statements. The business Americans borrow hardest to buy ranks close to the worst on the three numbers that most determine whether an acquisition survives: what it earns, whether its people stay, and how much of its revenue is already committed. Loan volume is not tracking quality. It is tracking how easily a business can be recognized and underwritten.

Why does the money concentrate here?

The driver is recognition, not economics. Buyers gravitate to businesses they can readily picture owning — restaurants, liquor stores, auto shops — because the mental model is immediate. Far fewer picture themselves owning a third-party logistics broker or a structural-steel detailer, though those often sit one industry code away with roughly double the margin and a quarter of the churn. Familiarity is easily mistaken for understanding.

The incentive structure reinforces it. The broker surfaces the listing that is easy to explain, because it is easier to sell; the lender favors the deal that is easy to model, because it is easier to approve; the buyer makes an offer on the business that is easy to imagine. Three parties each optimizing for legibility, and capital flows toward the hardest-to-run businesses in the market.

What is the low price really telling you?

That the market discounts earnings it does not expect to survive their owner. Public companies — audited, diversified, professionally managed, built to outlast any one executive — trade at a median of about 12.6 times EBITDA. A Main Street business sells for a fraction of that, often two to three times owner earnings. That discount is the mechanism by which small-business acquisition builds wealth: a dollar of profit acquired for a few dollars rather than twelve.

But the discount prices risk rather than opportunity. The market pays up for earnings that outlast the founder and little for earnings that may leave with them. The multiple is a verdict on transferability — and the most-financed businesses are those the market trusts least to survive a change of owner. A low price is not evidence of a bargain; it is a risk the market has already priced.

Which businesses actually transfer well?

Inverting the pattern describes the businesses that transfer well, and they are unremarkable by design. The highest-margin segment of the small-business economy is professional and technical services, followed by health care; the workforces with the lowest turnover are in utilities, finance, and information, where annual churn is a fraction of a restaurant's. Geography matters too: operating costs run about 27% higher in California than in Arkansas for the same dollar of overhead (BEA 2024 price parities), so a location-flexible service firm based in a low-cost state retains margin that a coastal, rent-heavy business surrenders to its landlord and labor market.

These businesses score well on the four factors that most determine whether a deal works — real margin, a stable team, low fixed obligations, and an owner who can be replaced. That last factor is the decisive one in every category, and the one least likely to appear in a CIM. Where the business is effectively the owner — the relationships, the know-how, the reason customers return — there is little underneath to acquire beyond a job carrying debt, currently at roughly a 9.75% effective SBA rate, a 1.25 coverage requirement, and a 10% equity injection. A forty-cent-margin business tolerates error for a period; a seven-cent one requires near-perfect execution from the first month, in a business that turns over half its staff a year.

Reading the financing signal

None of this argues against buying a restaurant; many are sound, and a strong operator can make a thin-margin business work. It argues against letting the loan approval, the broker's enthusiasm, or the ease of imagining the business substitute for analysis. The ability to vividly picture running a business is closer to a caution than a confirmation — it usually signals a common category, and common categories tend to be thin and contested.

The disciplined read is to underwrite what the business actually earns, whether its people stay, who is paid first, and what survives the seller. Financing availability and business quality are distinct signals, and in this market they point in opposite directions. The skill is telling them apart.

Sources & method

The figures come from primary public sources, joined at the industry level — directionally solid, not deal-level precise:

• SBA 7(a) FOIA records as of March 31, 2026 — funded change-of-ownership loans, pooled fiscal 2018–2026.
• IRS Statistics of Income, sole-proprietor returns for 2023 — profit-margin and rent ratios; these skew small and are used for ranking, not as the exact economics of any one deal.
• BLS JOLTS — quit rates for accommodation and food services.
• NYU Stern (Damodaran) dataset, 2024 — public-company EBITDA multiples, used to illustrate the size discount, not as a like-for-like comparison.
• Bureau of Economic Analysis, 2024 — regional price parities.
• Federal Reserve H.15 — prime rate (6.75% as of June 3, 2026); SBA pricing assumes standard 7(a) terms.