← All insights

Market Structure/

Fraud in receivables finance: fictitious debtors, duplicate invoices, stacking, and the controls that catch them

Receivables fraud is rarely sophisticated. It is a document that looks right, submitted to a lender that did not check. The controls that stop it are equally unglamorous, and they work.

A receivable is an unusually easy asset to fake. It is a piece of paper, or a line in an accounting system, that says one business owes another a sum of money. There is no serial number, no warehouse, no title. A lender that advances against invoices is advancing against the borrower’s word that the invoices are real, unless it does something to check.

The International Factoring Association’s Commercial Factor magazine identified synthetic identity fraud and duplicate invoicing as the leading threats to US factors going into 2026, and its call for shared data and standardized verification reflects how often the same schemes recur across lenders. The Association of Certified Fraud Examiners’ 2026 Report to the Nations makes the broader point that the presence of anti-fraud controls is associated with lower losses and faster detection. In receivables finance the controls are specific, and each one maps to a specific scheme.

Fictitious debtors

The simplest scheme invents a customer. The borrower creates invoices to a company that does not exist, or exists but never bought anything, and submits them for funding. When verification calls come, they go to a phone number the borrower supplied, answered by a confederate. The control is verification through independent channels: the lender finds the debtor’s contact details itself, from public records, the debtor’s registered address or a switchboard, and confirms delivery, acceptance and amount with someone in the debtor’s accounts payable function, not with the name on the invoice. Debtor credit reports and business registrations are pulled directly, and a debtor that has no footprint outside the borrower’s paperwork is ineligible.

Duplicate and altered invoices

The second scheme funds the same receivable twice: the same invoice submitted to the same lender under two numbers, or to two lenders at once. The controls are sequencing and search. Within a facility, invoice numbers are tracked and gaps and repeats investigated. Across a lender’s book, invoices are matched by debtor, amount and date so the same receivable cannot be financed under two facilities. Across lenders, the UCC index does the work: a receivables lender files a financing statement over its client’s accounts, so a second lender’s search reveals the first, and a filing that appears mid-term reveals a second. Altered invoices, where amounts or dates are changed, are caught by the same direct verification that catches fictitious ones, and by reconciling invoices to shipping documents and, where a processor is involved, to settlement records.

Round-tripping and related parties

A more patient fraud does not fake the invoice but fakes the payment. The borrower routes its own money, or a lender’s advance, through an affiliate or a friendly customer so that it returns to the operating account looking like a collection. To a lender reading bank statements, the receivable was paid and the borrower is a good credit. To a lender reading permissioned transaction data with counterparty detail, the deposit came from an account under common ownership and never matched a sale. Related-party receivables, invoices to companies the borrower’s owners also control, are excluded from eligibility for the same reason, and the beneficial ownership review that identifies the owners at onboarding is what makes the cross-check possible.

Stacking

Stacking is specific to short-term and revenue-based finance, and it is the largest driver of default in that segment. A business takes an advance repaid from daily receipts, then a second and a third from other funders, each unaware of or indifferent to the rest, until the combined daily debits exceed what the revenue can carry. It is a fraud when the borrower conceals the existing positions, and a credit failure even when it does not. The controls are the ones a secured lender should be running anyway: a UCC search before funding and repeated through the term, and bank data screened for new recurring debits. A first-position lender that has filed its financing statement has also given every later funder notice that the receivables are already spoken for.

Fabricated statements

Behind all of these sits the oldest control failure: accepting a document the borrower prepared as if it were evidence. Altered bank statements are among the most common frauds in small business credit, and they are increasingly easy to produce. The remedy is to stop accepting statements. A permissioned data connection to the bank returns the transaction record from the institution itself, complete across accounts and impossible to edit in a PDF tool. Where a document must be accepted, forensic review of its metadata, formatting, fonts and arithmetic catches most alterations, but sourcing the data directly catches all of them.

The paper behind the controls

Controls detect. Documents recover. Three provisions matter. A personal guarantee puts the principal’s own assets behind the debt. A validity guarantee, standard in factoring, is narrower and sharper: the principal personally warrants that every invoice submitted is genuine, owed, and free of offsets and disputes. It does not guarantee that debtors will pay, but it makes submitting a false invoice a personal liability even in a non-recourse facility. Cross-default and cross-collateralization provisions ensure that a fraud discovered in one facility triggers default across every facility the lender has with the group, and that all the collateral is available against all the debt.

None of this requires unusual technology or unusual skepticism. It requires doing, on every facility, the checks that are obvious in retrospect on every fraud. The lenders that lose money in receivables finance are rarely the ones that were deceived by something clever. They are the ones that did not call the debtor.

Sources

  1. 01IFA Commercial Factor, Data-led fraud defense in US invoice factoring: a call for collaboration and standardization (Q4 2025)
  2. 02Association of Certified Fraud Examiners, Report to the Nations (2026 edition)
  3. 03Uniform Commercial Code § 9-322, Priorities among conflicting security interests (Cornell LII)
  4. 04Uniform Commercial Code § 9-406, Discharge of account debtor; notification of assignment (Cornell LII)
  5. 05Plaid, Credit and Underwriting documentation