Slow cross-border payments push small businesses into borrowing, but the interest saved through faster settlement is often a rounding error beside the fees and exchange-rate costs that separate payment rails. That is the arithmetic behind a November 11 analysis built on new survey data from Morph, an Ethereum layer-2 network — a network built on top of Ethereum to process transactions more quickly and cheaply than the base chain — that also runs a stablecoin payments platform for small businesses, carried out with Studio by Informa TechTarget. Stablecoins, the payment rail at the center of the analysis, are blockchain tokens pegged to a reference currency, most commonly the US dollar.

The topline results: 37% of surveyed firms borrowed during payment delays, three days' interest on a $40,000 gap works out to about $26 at bank lending rates, fintech transfers already cost $40–$60 per $10,000 moved, and only 10% of respondents already use stablecoins.

What 150 Businesses Said — and What the Survey Leaves Out

The research summary covered 150 business owners, independent operators and payments decision-makers in Singapore, Hong Kong, the Philippines, Indonesia and Vietnam. Some 83% of respondents said international invoice payments usually took at least two business days to land in a local bank account. Separately, 37% had drawn on credit lines, overdrafts or loans to bridge cash-flow gaps they attributed to payment delays, and 10% of those aware of stablecoins already used them for business transactions or payouts. The summary does not say how far those groups overlap, so it cannot show whether the stablecoin users borrowed less than anyone else.

Three gaps limit what the numbers can carry. The summary gives no fieldwork dates, no sampling method and no loan amounts, so the 37% figure describes an experience, not a measured cost. It also folds every delay into a single category: a wire that takes two days to settle and a customer who pays two weeks late both count as "payment delays," but only the first changes when the payment rail does. And Morph sells the product under discussion, which does not make the answers wrong but does mean the questions were framed around the problem stablecoins claim to solve.

Three Days of Interest Comes to About $26

Take a company owed $100,000 by an overseas customer. It holds $60,000 in cash and must pay $40,000 in wages before the money lands, so it borrows that $40,000 for the three calendar days the payment is in transit.

The rate is the test. Bangko Sentral ng Pilipinas data reported by the Philippine Daily Inquirer put the average bank lending rate at roughly 7.7% a year in November 2025, so 8% is a fair benchmark for a creditworthy Philippine business. Smaller firms on overdrafts or non-bank credit can pay far more, so a 20% case sits alongside it.

At 8%, three days on $40,000 costs about $26 ($40,000 × 8% × 3/365). At 20% it is $66. Those figures are the ceilings on what faster settlement can save in interest, and they apply only if the stablecoin route delivers spendable pesos before payday. If conversion and bank payout leave a one-day gap, the saving shrinks; if the final bank credit arrives on the same day as a conventional transfer, nothing is saved.

Calendar days are the right unit. A payment initiated on Thursday that takes two business days lands on Monday, and the loan charges for Saturday and Sunday. Both routes have to be timed from the same initiation moment and counted to the same event: the hour the local bank account can fund the bill.

Put in daily terms, a $40,000 shortfall carries about $9 a day of interest at 8% and $22 a day at 20%, on simple interest with a 365-day year — a Coindoo calculation for a hypothetical three-calendar-day gap, rounded to the nearest dollar. Any extra fee on the stablecoin route must be subtracted from those figures: at $50, the entire 8% column turns negative.

Fees, Not Interest, Decide the Race

Set those figures against the cost of moving the money. A BIS working paper by Wenxin Du, Catherine Huang and David Scharfstein finds that fintech transfer operators charge a median of roughly $40–$60 to send $10,000 from the United States to developed-market currencies, while bank wires in the authors' sample carried foreign-exchange markups of 2% to 4.7% before explicit fees. On a $100,000 payment, a 2% markup is $2,000 — around 75 times the interest on the three-day gap at 8%. Whichever rail wins on fees and exchange rate wins the comparison; the interest is a rounding error beside it. The same paper notes that stablecoin routes add exchange, conversion and withdrawal charges of their own, and that recipients may still wait for conversion and bank processing before the funds are usable.

The loan's terms can push the interest saving lower still. An origination fee, a minimum interest charge or a bar on early repayment means three days of interest cannot be clawed back by repaying on day one. A business funding the gap from its own cash saves no interest at all; it gets its buffer back sooner, which is worth whatever that cash would otherwise earn on deposit — less than the lending-rate figures above.

Where the Saving Stops Being Small

Three things change the picture: the size of the shortfall, the length of the delay and the price of the credit. A distributor short $500,000 for ten calendar days — a payment sent before a long weekend that clears late the following week — pays about $1,096 at 8% and $2,740 at 20%. At that scale, a faster rail that costs an extra $50 covers its fee many times over, provided the local currency actually arrives.

The larger prize is structural. A firm that cannot predict when foreign receipts will land keeps a credit line open for the bad weeks and pays for that facility whether or not it draws on it. Settlement that arrives before payroll every time, not on average, lets the firm shrink or close the line. Settlement that is usually fast but occasionally stuck leaves the facility in place and saves only the odd day of interest. Morph's survey did not ask about consistency, so it has nothing to say about this benefit either way.

The Clock Stops at the Bank, Not the Wallet

Stablecoins move between wallets in minutes. Converting them into pesos, dong or rupiah and crediting a bank account is a separate service with its own timetable — which is how a token can arrive on Monday while the payroll loan runs to Wednesday. Coindoo's piece on the conversion step in stablecoin payments covers that leg in detail.

Stripe's documentation shows the split as one provider has built it. Its stablecoin wallet payouts, in public preview for eligible U.S. businesses outside New York, list typical receipt within minutes, subject to network conditions. Paying a stablecoin balance out to a bank account is a separate feature in private preview: Stripe converts the balance to the payout currency and hands it to the destination's banking network, with Philippine pesos routed over SWIFT, the global messaging network banks use to send cross-border payment instructions, and bank arrival depending on the destination and network. The converted amount can also move with the exchange rate between the two steps. A U.S.-only preview does not describe what a Vietnamese exporter can buy today; it is cited here because few providers document the two stages so plainly.

Once pesos reach a Philippine institution, the domestic leg is not where the time goes. BPI's published transfer features describe InstaPay, the country's real-time retail transfer system, as available around the clock within applicable limits, while PESONet, the batch system, settles on banking days with cutoffs. A business weighing a stablecoin provider needs to know which of those rails the provider uses for the final credit, and at what hour, before it books any saving on the loan.

That is the arithmetic the survey leaves out. For a $40,000 three-day gap at bank rates, the interest at stake is $26, and the fee difference between rails decides the outcome. For a $500,000 ten-day gap it is $1,096, and the deciding factor becomes whether spendable local currency reliably lands before the bill — evidence Morph's 150 respondents were not asked to provide.

The full analysis is published on Coindoo.