Tokenized money — stablecoins moving on public blockchains — settles a cross-border transfer in seconds, for a fee far below what a bank charges, and that is the part crypto points to when it calls remittances a test case it is passing. The test actually has two legs. The wire is solved. The cash-out — turning that balance into spendable local currency — still runs through the same slow, fee-taking network it always has, which is why a transfer that settles on-chain in seconds can still take a recipient days to actually receive.
Two legs, one word
“Remittance” describes the whole trip: money leaves a sender in one country and becomes usable money for a recipient in another. Historically one rail did both jobs at once — a bank wire or a Western Union counter moved the value and handed over something spendable at the other end, because there was no way to separate the two.
A stablecoin transfer separates them for the first time. It does the first job — moving value across the border — essentially for free and instantly. It does not, by itself, do the second job. A balance of tokenized dollars sitting in a wallet is not the same thing as cash in a recipient's hand or a deposit their bank will accept, and conflating the two is where the “crypto fixed remittances” claim overreaches.
What the wire actually proved
It is worth being precise about what did get solved, because it is real. The World Bank's Remittance Prices Worldwide database — which tracks pricing across 377 country corridors, from 48 sending to 111 receiving countries — puts the global average total cost of sending $200 at 6.36% of the amount sent, in its most recent published survey round (Q3 2025 data). A September 2026 Solana Foundation report on money movement, looking specifically at legacy-rail corridors from the US to Mexico, Brazil, Nigeria, the Philippines and India, cites an average fee of 6.49% on a $200 transfer and settlement times of three to five business days. Worth noting what that fee figure is: 6.49% is the World Bank's own global average for a $200 transfer one quarter earlier, so it corroborates nothing independently — it is the same series, read at a different point. (Solana Foundation is a blockchain-industry publisher with an evident interest in this comparison; its settlement-time range is reported here as its own claim, not independently verified.)
That gap — a fee measured in whole percentage points and days, against a fee measured in fractions of a cent and seconds — is genuinely the thing crypto changed. It is just not the whole remittance.
The leg nobody tokenized
Most remittance recipients cannot spend a stablecoin balance directly. They need local cash, or a deposit their own bank or mobile-money account will accept, because that is what pays rent and buys groceries where they live. Getting from a wallet balance to that point requires an off-ramp: a local agent, an exchange, or a payout partner with liquidity in the local currency and the compliance work — identity checks, sanctions screening — already done on the receiving end. None of that moved on-chain. Only the transfer between the two ends of it did, which is why the fee and the delay that used to sit on the whole trip now concentrate almost entirely on that last mile.
The clearest evidence of this is who is adopting the technology, and how. Both incumbents now issue their own stablecoin: Western Union launched USDPT on Solana in May 2026, and MoneyGram launched MGUSD on Stellar in June 2026. Both are positioned first as settlement assets — moving dollars across their own networks around the clock instead of waiting on correspondent banks — and both now put a dollar-denominated balance in front of customers too. MoneyGram's sits in a self-custodial wallet inside its own app, letting a recipient hold dollars and convert to local currency on demand.
That second part cuts against the lazy version of this argument, so it is worth stating plainly: the incumbents are not only using stablecoins out of sight. But look at what neither of them retired. MoneyGram's balance still converts on demand through its network of roughly 500,000 retail locations, because cash is what a recipient who needs cash actually needs. The wire between their own accounts got replaced. The payout network got kept — and the payout network, not the wire, is the expensive thing to run. That is a meaningful signal about where the actual cost still lives.
Where it's already a clean pass
None of this means the test case fails everywhere — it means the test case is conditional. When a recipient is willing and able to hold the stablecoin itself, there is no second leg to solve, because the transfer is the payment. A freelancer invoicing a client abroad in USDC, a small merchant settling with a cross-border supplier, a diaspora saver who already keeps a dollar-denominated balance rather than the local currency — for all of them, the wire is the whole trip, and tokenized money passes outright: instant, near-free, and finished the moment it arrives.
That version of a cross-border transfer — fund it once, let the recipient receive it directly into their own wallet rather than through an intermediary account — is closer to how a claim-link transfer works than how a traditional remittance works, and it is the shape the rest of this gets judged against as off-ramp networks slowly improve.
Where Swop sits in this
Swop is built for the leg that is already solved, and is honest about not being a cash-out network. Swop is fully self-custodial — keys are generated and held on your device; Swop never holds them — so a transfer that lands in a Swop wallet is immediately the recipient's, not a balance an intermediary is holding on their behalf. Transactions on Swop are gas-sponsored, so neither side needs to hold SOL or ETH just to move the money, and Swop runs on Solana, Ethereum, Base, and Polygon, so the wire itself isn't tied to a single network's fees or congestion.
That is one leg of the trip, stated plainly as one leg. Whether a given remittance also needs a cash-out still depends on whether the recipient can use a dollar-denominated balance directly — the question this piece spent its length on, not a question any wallet answers by existing.
FAQ
Do stablecoins actually make remittances cheaper?
For the transfer itself, yes — moving a stablecoin across a public blockchain costs a small, predictable network fee regardless of distance. But the cost most people measure is the whole trip: sender to recipient, in spendable local currency. The World Bank's most recent published round puts that whole-trip average at 6.36% of a $200 transfer, and it has sat above 6% for years, because the fee is mostly charged on the cash-out leg, not the wire.
Haven't Western Union and MoneyGram already solved this with their own stablecoins?
Not the cash-out leg. Western Union launched USDPT on Solana in May 2026 and MoneyGram launched MGUSD on Stellar in June 2026, and both go further than back-end settlement — MoneyGram's sits in a self-custodial wallet in its own app, so a recipient can hold dollars instead of converting immediately. But neither company retired its payout network, and MoneyGram's dollar balance still converts to local currency on demand across roughly 500,000 retail locations. The stablecoin replaced the wire between their own accounts; the agent network that turns a balance into spendable cash is still there, still staffed, and still priced.
If the wire is instant, why do remittances still take days?
Because the wire and the payout are two different jobs. A stablecoin transfer settles in seconds. Turning that balance into cash a recipient can spend at a shop, or a deposit their local bank will accept, requires an off-ramp — a local agent, exchange or payout partner with liquidity, compliance checks and operating hours in the receiving country. None of that moved on-chain; only the transfer between the two ends of it did.
When does a stablecoin remittance skip the cash-out problem entirely?
When the recipient is willing and able to hold the stablecoin itself, rather than needing it converted to local cash — a freelancer being paid in USDC, a merchant settling with a cross-border supplier, or anyone who already spends or saves in a dollar-denominated balance. In that case there is no second leg: the transfer that lands in the wallet is the payment, and the test case is a clean pass.
Written by the Swop product team. Editorial rules: a direct answer up front, no invented statistics, dates on everything, and links to primary sources.