Talk to us
All articles Insights

Two banks were fined for trade they could not see. The exporter will pay in paperwork.

Abstract 3D render: a stream of golden spheres flows behind a translucent cyan glass wall, only a few visible through a narrow slit, on a deep blue-black background

On 8 October, the US Federal Reserve and the Office of the Comptroller of the Currency fined American Express National Bank US$350 million for anti-money-laundering failures. The OCC’s finding was specific: from roughly June 2014 to May 2025 the bank processed about US$13 billion in suspected trade-based money laundering, largely suspicious card charges and the repayments that followed them, in some cases through accounts linked to the bank’s own insiders. Five days later, on 13 October, FinCEN assessed a US$1.2 billion penalty against Bank of America, the third-largest in its history, with a record US$750 million OCC penalty counted inside it. The regulators said the bank missed or delayed reports on more than US$5 billion of suspicious activity between January 2018 and July 2025, and that about US$1 billion of it related to national-security risks, including potential trade-based money laundering across the US south-west border.

Two banks, one week, the same phrase. Trade-based money laundering is the name for moving value by misdescribing a trade: an invoice for more than the goods are worth, a shipment that never sailed, a repayment that matches nothing that shipped. Regulators are telling banks they must see it. Banks will respond the way they always have, by asking the customer for more documents. The customer who feels that is not the launderer. It is the exporter with real goods on a real ship.

What the two cases have in common.

Neither bank was fined for failing to read paper. American Express processed card charges and repayments; there was no bill of lading to inspect. Bank of America, FinCEN said, had built a monitoring feature for trade-based laundering by early 2024 and then used it only on activity already under investigation. In both cases the bank saw money moving and had no way to know whether a trade stood behind it.

That is the shape of the problem. From its own seat, a bank can inspect a description of a trade. It cannot see the trade.

Why the exporter pays.

When a bank is told to see more, its compliance team reaches for the only tool it has: documents from the customer. Another copy of the invoice, more proof of shipment, more questions about the buyer, longer holds on incoming payments. For a large exporter with a trade finance desk, this is friction. For a small one, it is a reason to be declined.

The Asian Development Bank’s latest trade finance gap survey, published in January, puts unmet demand at US$2.5 trillion in 2025, unchanged from 2023, and finds 41% of requests from small and medium-sized firms rejected, now close to the 40% rate for large companies. The survey’s own authors say that narrowing needs more research, because firms that are refused repeatedly stop applying. Every new fine for unseen trade pushes that number the wrong way.

More paper is the launderer’s home ground.

Trade-based money laundering works precisely because paper is cheap and detached from the goods. A launderer produces a plausible invoice. An honest exporter produces a plausible invoice. The bank reads both the same way. Asking for a third copy does not tell them apart; it only raises the cost of being honest.

What tells them apart is whether the events the paper describes actually happened, and whether the bank can see them happen. Did control of the goods change hands? Did the payment go to the party that gave up that control, at that moment, for that amount? Those are questions about events, not about documents.

This is the layer we are building

Our work sits where title and money meet. In the digitrad vault, an electronic bill of lading and a payment are exchanged in one step: the title moves only as the money moves, and both leave a record of who, when and how much. Together with a tokenised-deposit platform and a bank, we are working on settling open-account trades in a bank-issued deposit token that waits for the electronic bill of lading. On the stablecoin side, our end-to-end run with BitTrade showed a cross-border payment received, converted and withdrawable as ordinary yen within the same business day, with every step on record.

For a bank, this changes the question from “does this invoice look right” to “did this exchange happen”. A trade settled that way carries its own evidence. The honest exporter’s payment arrives together with the title that justifies it. The launderer’s invoice has nothing to attach to.

The litmus test

Ask your bank one question: when it reviews a trade payment, does it see the shipment, or a description of the shipment?

If the answer is a list of documents, it sees paper. Paper is what the two fines were about, and paper is what you will be asked for next.

Related reading

Leave a Reply

Discover more from STANDAGE

Subscribe now to keep reading and get access to the full archive.

Continue reading