Western Union and MoneyGram turn remittances into stablecoin cards

Four money transfer incumbents have launched stablecoin cards in ten months. The card is where the transfer stops being a transfer.

MoneyGram put a stablecoin-backed Visa card into Colombia on September 10. Five weeks earlier, Western Union launched one in 37 markets. Both run on Visa, both were built with the same enabler, and both do the same thing to the business model they came from: they keep the money inside the sender’s app after the transfer lands, instead of handing it over at a counter.

That is a larger change than it sounds. A remittance company has always been paid for the moment money crosses a border. A card gets paid for what happens next, every time the receiver buys groceries. For these companies, the card extends the product past the point where their revenue used to stop.

What Western Union built

Stablecard, announced on August 4, pairs a wallet with a Visa card inside the Western Union app. Transfers arrive directly in the wallet as USDPT, a dollar token issued by Anchorage Digital Bank on Solana and redeemable one for one. The card spends that balance at Visa merchants and works with Apple Pay and Google Pay. It went live in 37 markets with Rain, with more than 60 targeted by the end of the year.

Two design choices stand out. The card is structured as a secured credit card rather than a debit card, so the balance backs the line instead of being drawn down directly. And the token is Western Union’s own, which matters for reasons that have nothing to do with technology and everything to do with who holds the reserves.

The company describes the audience as receivers in markets with unstable currencies, plus customers who want to hold and control digital dollars. Chief executive Devin McGranahan said the product gives consumers “a new way to hold value, move money and spend confidently across borders.” Rain’s chief executive, Farooq Malik, was blunter about the design goal. The card is for “people who have never thought about onchain money and never need to.”

What MoneyGram built

The MoneyGram card launch came five weeks later and starts smaller, in one country. Customers hold USDC inside the card account, spend it by tapping through the MoneyGram app, and can still move the balance to local currency for cash pickup at a retail agent. A physical card arrives late in 2026 and brings ATM withdrawals with it. Support for MGUSD, the stablecoin MoneyGram launched on Stellar in June with Bridge as issuer, comes after that.

The retail tie is the difference worth noting. MoneyGram serves more than 60 million active customers across upwards of 200 countries and reaches nearly 500,000 agent locations, and the card does not ask anyone to give that up. The digital balance and the cash counter sit in the same product, which is a sensible hedge in markets where card acceptance is thinner than the map suggests.

The Stellar choice is not new here. MoneyGram launched Access on Stellar in June 2022, a cash-in and cash-out service for digital wallets that used USDC and started in Canada, Kenya, the Philippines and the United States. Four years later, the same rail carries a card, and the agent network that made Access work is the part a competitor cannot copy in a quarter.

MoneyGram’s framing is almost identical to Western Union’s. A spokesperson said the company had “taken a fundamentally different approach,” pairing a stablecoin balance with ordinary spending, which is the same claim in different words: the token is plumbing, the product is the point.

Read more: Mastercard’s Yellow Card bet awaits its first priced corridor • Rain’s $250 million round prices stablecoin card plumbing

The same enabler underneath both

Rain built both programs. It also sits behind a long list of other card programs, holds Visa principal membership, processes more than $3 billion a year, and has drawn about $2 billion through a settlement credit facility to fund its daily obligations to the network. Two of the largest remittance brands in the world now depend on one private company for authorization, conversion and settlement on their newest product.

That concentration is the story underneath the launches, and it is not unique to Rain. Zepz, which owns WorldRemit and Sendwave, announced stablecoin-linked Visa cards in December 2025 powered by Bridge, the Stripe subsidiary. Remitly’s Global Card, announced in August, holds balances in fiat or USDC. Four incumbents, two enablers, one network. Every announced program here runs on Visa, which matches Visa’s own count of more than 160 stablecoin-linked card programs live on its rails.

Anyone sizing the risk here should read the card enabler layer as its own market. When the same firm runs the conversion window for competing brands, a bad week at that firm is a bad week for all of them, and none of the brands control the fix.

“A new way to hold value, move money and spend confidently across borders.”

Farooq Malik, chief executive of rain

Why a card, when the whole business is moving money

The obvious question is why a company that already moves money to 200 countries needs a piece of plastic.

The answer starts with where the customer goes after the money arrives. In the old flow, the receiver collects cash, and the relationship ends until the next transfer. FXC Intelligence, writing about money transfer providers in August, put it plainly: cards are a tool to capture customers inside the provider’s network so it can sell them more. Every purchase on the card earns interchange, the balance stays on the provider’s books, and the receiver becomes a user rather than a pickup.

The second reason is the state of the core business. Western Union’s second quarter revenue came in at $1.0 billion, down 1%, with Americas retail weaker than the company expected and margins under pressure. Branded digital transactions grew 25% in the same quarter. When retail drags and digital compounds, the products that get funded are the digital ones.

The third reason is demand that has nothing to do with crypto. In markets where the local currency loses value between payday and the weekend, a dollar balance is a savings product. Western Union says so in its own announcement, and MoneyGram’s choice of Colombia for the first launch fits: the country took in more than $6 billion of remittances in the first half of 2026 and has now passed $1 billion a month for 24 consecutive months.

What the card does not fix

It does not make sending cheaper. The World Bank put the global average cost of sending $200 at 6.36% in the third quarter of 2025, with digital transfers at 4.59% and everything else at 7.30%. Latin America and the Caribbean averaged 5.64%. The UN target is 3% by 2030. None of the card launches touch that number, because the fee is charged when the money is sent and the card sits on the other end.

What the card can remove is the second cost, the one nobody quotes: the trip to the agent, the cash-out fee where one applies, and the local-currency conversion the receiver takes whether they want it or not. Hold dollars and spend them directly and that leg disappears, which is a real saving for the receiver even though it never shows up in the headline price of the transfer.

It also relocates the foreign exchange spread rather than removing it. Somebody still prices the conversion when a dollar balance pays a peso merchant, and neither company has published what that costs. Until one of them does, the honest position is that the card moves the margin around, and the disclosure needed to say more is missing.

For the receiver the question is narrower than the policy debate. What does it cost to turn $200 of a dollar balance into groceries, once the merchant conversion, any ATM fee and the gaps in local acceptance are counted? None of the four companies has published that number, and it is the one that decides whether a balance stays put or gets emptied on the day it arrives.

The receiver has to actually want dollars

Every one of these products rests on a behavioral bet: that the person receiving money will hold a dollar balance instead of converting it the moment it lands.

In Argentina, Venezuela and parts of Central America, that bet looks safe, since holding dollars is what people already do through informal channels. Elsewhere it is genuinely open. Rent, school fees, and the neighborhood market run on local currency, card acceptance outside cities is patchy, and a balance that cannot buy what a household needs today is a balance that gets cashed out today.

MoneyGram’s design takes that seriously by keeping the cash counter in the product. Western Union’s keeps the wallet and the retail network in the same app for the same reason. Both companies are betting they can move receivers up the curve gradually, and both have built the exit ramp in case they cannot.

Two bets on which dollar

The token choice is where the strategies diverge, and it is worth reading carefully.

Western Union commissioned its own, USDPT, issued by a federally chartered trust bank. MoneyGram started with USDC, the largest regulated dollar token, and will add its own MGUSD, issued by Bridge. Both companies are heading toward a proprietary token while using a familiar one to get the product live.

The reason is reserves. A dollar token is backed by short-term assets that earn interest, and under the GENIUS Act, the issuer cannot pass that interest to holders. So the float belongs to whoever issues the token, and a remittance company that owns its token captures the yield on balances its own customers hold. On a large enough float, that income can matter as much as the interchange. It also explains why a company with a payout network in 200 countries would rather mint a dollar than rent one.

The risk sits on the other side of the same trade. A proprietary token has no market outside the app that issues it. A customer holding USDC can move it anywhere. A customer holding USDPT or MGUSD has a balance that is only as liquid as its issuer’s redemption promise, and that difference will be tested the first time someone needs their money during a bad week.

What everyone else is doing

The wave is wider than these two. Zepz is going first in Brazil with cards for Sendwave wallet customers, serving a base of more than 9 million across about 5,000 corridors. Remitly started a phased rollout in August to invited customers, with wider regional coverage promised through the rest of 2026, and lets customers hold balances in USDC or fiat. Four of the biggest names in consumer cross-border payments have now shipped or announced the same product within a year.

For scale, FXC Intelligence estimates stablecoins accounted for just under 1% of consumer-to-consumer cross-border flows in 2025, while those flows grew 44% year over year. The incumbents are not chasing a market that already exists. They are positioning for one they expect, at a cost that is small against their digital budgets, with the enabler carrying most of the build.

What would tell us whether it is working

Nobody has published a number that matters yet. Not activations, not cards funded, not spend per card, not the share of transfers that stay in a wallet instead of going to a counter. Until one company does, the launches are announcements about strategy, not evidence about behavior.

Four things to watch into next year. Whether Western Union hits its 60-market target by December, since a 37 to 60 expansion in five months would say the compliance work travels well. Whether MoneyGram’s physical card and ATM withdrawals land on schedule late this year, which is the feature that makes a dollar balance useful in a cash economy. Whether any of the four discloses how much gets spent rather than cashed out. And the January 18 effective date of the GENIUS Act, which will decide how proprietary tokens can be marketed to American senders and their families abroad.

The wider corridor economics, including what tokenized payouts do to pricing in markets like Nigeria, are a separate question that Mastercard’s Africa corridor work is already testing. For now, the card is the incumbents’ answer to a problem they have had for decades: they own the transfer and somebody else owns everything that happens after it. Two of them have just decided to stop handing that part away.