Where the regulator's definition sits, and why the placement matters. The Bureau's definition of a merchant cash advance appears in 12 CFR 1002.104(b), which is a list of excluded transactions. In other words, the federal regulator defines the product in the course of exempting it from the small business lending data collection that Congress required under section 1071 of the Dodd-Frank Act.
That exclusion is new, and it reverses the earlier position. The 2023 rule covered merchant cash advances: as the Bureau put it, "the 2023 final rule encompassed a wide range of credit products, including merchant cash advances and agricultural credit." In the rule published on May 1, 2026, the Bureau narrowed its focus, finding that "the rule should focus on core, generally applicable lending products that are most likely to be foundational to small businesses' formation and operation, loans, lines of credit, and credit cards, before determining whether to expand the scope of the rule to include more niche or specialty lending products." It added merchant cash advances, agricultural lending and credit under $1,000 to the exclusion list. The section's own source note records both dates.
The Bureau gave a data-quality reason: "application data from merchant cash advance providers would not produce data comparable to other transactions, which would limit their value as part of the dataset." It also acknowledged the cost, stating that "small businesses will experience a cost in the form of a reduction in fair lending and community development benefits related to these types of transactions." The practical consequence for a business owner is that the public dataset which will eventually show who gets small business loans and on what terms will contain nothing about this product.
Whether it is a loan is genuinely unsettled, and the regulator says so in both directions. This is the question everything else about the product turns on, because usury laws, licensing requirements and most consumer protections attach to lending. Providers have long maintained that an advance is a purchase of future receivables rather than a loan. The Bureau addressed the argument directly in 2026 and refused both extremes:
"The Bureau disagrees with the assertions of certain commenters that MCAs are categorically not credit. The Bureau also disagrees with categorical attempts to exclude MCAs from the definition of credit, including on the grounds that they should be treated as simply a purchase of future receivables. There is evidence provided by commenters that in certain instances, MCAs in practice do involve debt, confer a right to payment, and are loans."
And, in the same passage, the other half: "The Bureau also disagrees with the assertion that all MCAs should be covered as credit under ECOA. The Bureau believes that certain MCAs may have some features resembling factoring in certain circumstances."
So the answer is that it depends on the agreement. The features that push a particular arrangement toward being treated as a loan are whether repayment is genuinely contingent on sales or effectively absolute, whether the business can have its remittance reconciled downward when sales fall, and whether the provider has recourse against the owner personally. The Bureau noted that last point as a practical matter: "Commenters also provided evidence that in many instances MCA providers are seeking recourse against the natural person owners of a small business that no longer has revenue." A business signing an advance should read the reconciliation clause and the personal guarantee clause first, because between them they largely determine what the arrangement actually is.
The pricing mechanism, and the part that behaves backwards. An advance is quoted as a total: a business receives a sum and agrees that a fixed larger sum will be collected. Because the total is fixed at signing and does not vary with time, the cost of the money is not a rate, and converting it into one produces a result that runs against intuition.
On a loan, repaying early saves interest. On an advance, repaying early saves nothing, because the amount owed was fixed at the start. The business simply pays the same total over a shorter period, which means the annualized cost goes up. A business whose sales exceed expectations therefore pays a higher effective rate than one whose sales disappoint, and the strongest trading month is the most expensive one. This is the single most important structural fact about the product and it is the one no quoted figure conveys, because the quoted figure is a multiplier rather than a rate.
A second consequence follows from the same structure. Because collection is a percentage of daily or weekly receipts, the payment falls when sales fall, which is presented as flexibility and genuinely is. But it also means a slow month extends the term rather than reducing the cost, and a business that takes a second advance while the first is outstanding is having two percentages of the same daily receipts collected at once, which is where the arrangement most often becomes unsustainable.
What law reaches it, and the honest answer is very little. Regulation Z, the source of the disclosures a consumer borrower expects, exempts credit "primarily for a business, commercial or agricultural purpose" at 12 CFR 1026.3(a)(1) and separately exempts credit "to other than a natural person" at 1026.3(a)(2). So there is no required annual percentage rate, no periodic statement and no billing-error procedure, whatever the arrangement turns out to be legally.
Some states have legislated in the gap, and California's approach is worth understanding because of how it handles the loan question: it brackets it rather than answering it. An advance falls inside the statute's "commercial financing" because it is an "accounts receivable purchase transaction," a term the statute defines to reach "cash receipts that are owed to the recipient or are collected by the recipient during a specified period or in a specified amount." Once inside, section 22802 requires the provider to disclose six items before the transaction closes, including "the total cost of the financing expressed as an annualized rate." So the statute requires an annualized rate on a product it never calls a loan, and it says elsewhere, in section 22800(m), that arranging financing on a bank's behalf "shall not be construed to mean that the provider engaged in lending or originated that loan or financing."
Two limits, and they matter. Section 22801 exempts depository institutions from the whole division, and the law reaches only offers of $500,000 or less. And which other states have such laws, and what those laws cover, varies.
How the product is sold is part of what a buyer needs to know. Because nothing requires a rate to be quoted on it, an advance is marketed on speed and approval likelihood rather than on cost, which is a combination that reliably produces expensive outcomes for the businesses least able to shop. And where an advance is arranged through a broker rather than bought direct, the broker's fee may come out of the advance itself, so the business can receive less than the headline figure while still owing the full ceiling. The amount actually disbursed, rather than the amount agreed, is the number to check on the funding statement. The Federal Reserve's 2026 report on employer firms measured the visible end of that: among firms that borrowed, 60 percent of those that borrowed from online lenders reported that actual borrowing costs were higher than expected, against 4 percent who found them lower, and the comparable figures for small and large banks were 37 and 32 percent. The same report found that "high interest rates and unfavorable repayment terms were the most common challenges at online lenders." Those are findings about a broader category than merchant cash advances alone, but they describe the channel this product is sold through.
