TL;DR
Most stablecoin business cases describe a large supplier payment into an emerging market, where a percentage saving on a six-figure invoice makes the arithmetic obvious. That case is real, but it is not the one most finance teams deal with monthly.
Operating expenses are. A company headquartered in one country runs on software licensed in a second, cloud capacity billed in a third, contractors in a fourth, and an office with rent and utilities in a fifth. None of those invoices is large. All of them are regular, dated, and denominated in a currency the company does not hold. EY-Parthenon's post-GENIUS Act survey found that 77 percent of corporate stablecoin users cite supplier payments as their primary use case, and that 41 percent reported cost savings of at least 10 percent [1]. The operating-expense slice of that spend is the part nobody models, because each individual payment looks too small to bother with.
It is worth bothering with, for reasons that have little to do with the size of any single payment.
Operating costs that cross a border
The category is broader than most teams assume once they list it out.
Software and cloud. Licences, seats and infrastructure billed by vendors incorporated elsewhere, usually in dollars, usually monthly, usually on a fixed date.
Contractors and freelancers. Developers, designers, analysts and agencies engaged in other countries, invoicing in their own currency or in dollars, expecting payment within days rather than weeks.
Premises and utilities. Rent, serviced-office fees, power and connectivity for a local presence in a market where the company has no entity and no local bank account.
Local professional services. Accountants, legal counsel, compliance advisors and filing agents retained in each market the company operates in.
Platform and marketplace fees. Advertising spend, app-store commissions, listing fees and other charges deducted or invoiced by platforms based abroad.
What unites them is not the amount. It is the shape: low value, high frequency, hard deadline, wrong currency.
Why small, frequent payments cost the most
Cross-border cost does not scale down neatly. A wire carries a sending fee, intermediary deductions along the correspondent chain, a receiving fee at the far end, and an FX spread buried in the rate. Several of those are close to fixed, so on a small payment they stop being a percentage and start being a tax. The World Bank's data puts the global average cost of sending money across borders at around 6 percent, with banks the most expensive channel at close to 15 percent [2]. Those figures cover remittances rather than corporate treasury, and enterprise pricing is better, but the direction holds: the smaller the payment, the worse the ratio.
Frequency compounds it. A payment made once is an annoyance. The same payment made every month, across eight currencies, is a standing line item plus the operational cost of initiating, chasing and reconciling each one. And because these bills carry hard dates, a two or three day settlement window forces teams to pay early, which means holding balances in currencies they would rather not hold, or accepting the risk that a vendor suspends service over a payment that was sent on time but landed late.
Prefunding is the hidden cost underneath all of this. To pay reliably in eight currencies you hold balances in eight currencies, sized for the worst case and sitting idle the rest of the month. That is working capital doing nothing, exposed to the FX moves you were trying to avoid in the first place.
How the payment flow works
The important thing to understand is that your payee almost certainly still receives local fiat. A landlord in Manila, a utility company in Berlin and a contractor in São Paulo are not going to invoice in USDC or accept it, and they do not need to. The stablecoin leg is internal to the payment, not something the recipient sees.
The mechanism is the same one covered in depth in our stablecoin payments guide for global businesses: the company funds a payment in USDC, the provider moves that value across the settlement layer in minutes rather than days, and the destination leg converts to local currency and pays into the recipient's ordinary bank account. What changes is the middle. The correspondent chain, where the intermediary deductions and the multi-day wait live, is the part that gets removed.
Two consequences matter for operating expenses specifically. First, funding happens per payment rather than per currency, so the idle balances across eight markets are released back into the business. Second, because the settlement window collapses from days to minutes, payments can be initiated close to the due date instead of days ahead, which is the difference between paying on time and paying early to be safe.
The Federal Reserve's analysis is a useful corrective on where the cost actually sits: moving the stablecoin itself is close to free, and the real cost structure lives at the on-ramp and off-ramp, shaped by regulation, liquidity depth and how competitive the provider market is in each destination [3]. So the saving is genuine but it is not unlimited, and it varies sharply by market. A euro payment is not where this shines. A payment into a currency where banking rails are slow and the FX spread is wide is.
What stablecoins do not solve
Four limits matter.
Accounting treatment is unsettled. Stablecoins fall outside the scope of the fair-value rules introduced by FASB ASU 2023-08, because a fiat-backed token carries a redemption right against the issuer's reserves and therefore represents an enforceable claim. That leaves classification to judgement, somewhere between cash equivalents, financial instruments and intangible assets, and US GAAP does not yet give a definitive answer [4]. Agree the treatment with your auditor and write it into policy before the first payment, not after the first audit question.
Holding is not the point. Under the GENIUS Act, issuers of payment stablecoins cannot pay yield simply for holding a balance [5]. There is no treasury return to chase here. Value comes from settlement speed and released working capital, and a balance held longer than it needs to be is just an unhedged dollar position.
Not every market is worth it. Where local rails are already fast and cheap, the gain is marginal. The honest test is corridor by corridor, not company-wide.
It is a payment method, not a policy. Approval thresholds, who can initiate, how counterparties are verified and how the whole thing reconciles into the ledger still have to be designed. The treasury-side controls are covered in our CFO implementation guide.
How to test it on your own corridors
Operating expenses are a good first use case precisely because they are unglamorous. The amounts are small enough to pilot without board-level risk, the payments repeat every month so you get a clean read on cost and settlement time within one cycle, and the counterparties are ones you already know and have already onboarded.
A provider handling this properly gives you stablecoin settlement into local payout rails, so the USDC leg is internal and the recipient is paid in their own currency into their own bank account, with the funding, conversion and payout visible in one place rather than reconstructed from three systems. Start with the two or three markets where your bank rails are slowest and the spread is widest, run a month of real bills through them, and compare the landed cost and the settlement time against the same payments a month earlier. That comparison, on your own corridors and your own volumes, is a more useful answer than any industry average.
Sources
[1] EY-Parthenon. "Cost Savings and Speed Drive Stablecoin Adoption." 2025.
[2] World Bank. "Remittance Prices Worldwide, Issue 54." September 2025.
[3] Federal Reserve Board. "Payment Stablecoins and Cross-Border Payments." FEDS Notes, March 2026.
[4] Forvis Mazars. "Accounting for Stablecoins: Navigating Uncertainty Within US GAAP." November 2025.
[5] K&L Gates. "Crypto in 2026: The Democratization of Digital Assets." January 2026.
.png)
