Written by: Vaidik Mandloi
Compiled by: Luffy, Foresight News
Did you know? In July, spending on cryptocurrency credit cards exceeded $759 million, with over 9 million transactions. That's nearly two and a half times the volume from the same period last year. And yet, over 90% of that transaction value is still processed via Visa cards.
Almost all crypto card projects tell the same story: using stablecoin payment rails to bypass card network fees and pass the savings back to merchants. We previously explored this logic when analyzing Stripe's cross-border payment infrastructure built on stablecoins.
I delved deep into a core question: what actually happens when you try to cut out the traditional card networks? Can bypassing Visa and Mastercard truly save merchants money? Which layer of the payment infrastructure can stablecoins actually replace?
The conclusion I reached was completely unexpected.
How the Payment System Works
To find the answer, we must first understand where fees go when a consumer swipes a card to pay. My first realization was a common misconception: many, including practitioners in the crypto industry, believe card networks like Visa take the largest cut of the fees.
That's not the case. When a merchant accepts a $100 premium credit card payment, they pay a Merchant Discount Rate of about 2.2%, totaling $2.2. The key point is that this $2.2 is not pocketed by Visa but is distributed among three parties, and not equally.
- The largest share, about $1.75, goes to the issuing bank—the institution that issued the credit card to the consumer. This fee is called the Interchange Fee and accounts for 70%–80% of the merchant's total cost.
- Next, the merchant's payment processor (acquirer) takes a service fee of $0.30–$0.70.
- Lastly, the card networks—Visa/Mastercard—which everyone aims to disrupt, collect only a $0.13–$0.18 Assessment Fee, roughly 7%–9% of the merchant's total outlay.
This means that merely removing Visa only eliminates the smallest fee in the entire chain. There's a fundamental reason why Visa's fee is so low.
Visa does not issue credit to anyone, does not bear credit risk, chargeback disputes, or fraud losses. In fact, Visa doesn't even move money. It's simply an information transmission network that activates when a user swipes at a merchant terminal. Visa's job is to relay authorization messages between the merchant terminal and the issuer and set the rules for the entire system. The heaviest responsibilities fall on the issuer.
The issuer extends credit to consumers, bears bad debt risk, carries the cost of funds between the purchase date and the bill payment date, and uses interchange fees to subsidize reward programs to attract card usage.
This creates Visa's highly attractive business model. In 2025, Visa processed $14.2 trillion in payment volume across 257.5 billion transactions, generating $40 billion in net revenue with a net profit margin close to 50%. It makes about $0.13 per transaction on average—that's the entirety of its profit. Visa ranks among the world's highest market cap companies not because it charges high fees per swipe, but because it processes hundreds of billions of transactions annually with near-zero marginal cost and absolutely no credit risk.
Now, let's address the most challenging reality for stablecoin cards.
All stablecoin cards on the market are debit products. Funds—USDC, USDT—are already held in the user's wallet before a purchase occurs. There is no funding period and no revolving credit generating interest income. This places these cards in a completely different economic model.
Furthermore, the U.S. Congress's Durbin Amendment in 2010 capped interchange fees for debit cards issued by banks with assets over $100 billion at $0.21 + 0.05% per transaction. Most stablecoin card solutions partner with small banks (digital banks) with assets below the $100 billion threshold, exempt from the Durbin Amendment—a common FinTech partner bank model.
The average interchange fee for an unrestricted, dual-message network debit card is currently about $0.62 per transaction. A clear comparison: a $100 premium credit card transaction generates $2.20 in total revenue; a stablecoin debit card, even applying the higher, exempt rate, yields only $0.62 in total revenue. The project must then pay card network fees, processor fees, partner bank costs from these 62 cents, cover fraud losses and operational expenses, and only the remaining sliver can be considered for merchant rebates.
What Can Stablecoins Actually Replace?
As seen above, the savings from removing Visa are minimal, and the profit margin for operating a stablecoin card based on debit interchange fees is extremely narrow. However, the value of stablecoin payments might not lie in saving negligible card network fees but in potentially replacing more core parts of the payment stack.
To investigate, I revisited our previous article "Stripe Building Its Own Blockchain," which broke down a cross-border payment into seven fee layers: Acceptance, Orchestration, License/Compliance, Custody, FX, Issuance, and Settlement/Clearing. I mapped an in-country card transaction against these seven layers to see which ones actually change.
From the merchant's perspective, the Acceptance layer remains completely unchanged. The process stays the same: the merchant has a terminal, the user swipes, and the merchant still pays the Merchant Discount Rate to the acquirer. The merchant may not even be aware the funding source is USDC. On the merchant's statement, this transaction is identical to any other Visa transaction. Similarly, the Orchestration layer still goes through Visa or Mastercard, and the Issuance layer still relies on a partner bank and the card network's BIN. FinTechs were already using this model long before stablecoins—no fundamental innovation.
The real change happens in the backend layers, the ones most easily overlooked.
The Settlement/Clearing layer is the only area where stablecoins can bring fundamental change. In the traditional model, settlement between the issuer and card network operates on a T+2 cycle, further delayed by weekends and batch processing. To address this pain point, service providers like Rain support end-of-day stablecoin settlement with Visa; Mastercard has also begun accepting USDC, PYUSD, RLUSD, etc., for intraday settlement. This significantly compresses the traditional T+2 settlement, approaching real-time clearing and freeing up working capital previously tied up at the issuer.
However, the benefits from optimized working capital flow entirely to the issuing bank. The merchant's discount rate does not drop due to faster settlement, and the consumer's checkout experience feels no difference.
The sole beneficiary of T+0 stablecoin settlement is the card operator, who no longer needs to float funds for the two-day settlement period. The innovation from stablecoins essentially helps issuers optimize treasury management. The gains can be substantial at scale but offer no direct benefit to merchants or consumers. Meanwhile, Visa has no incentive to lower its fees—the cost of settlement float was never borne by Visa; the pressure was always on the issuer. Even if stablecoin settlement lowers the issuer's funding cost, Visa's own costs haven't changed, so merchant fees naturally remain the same.
Notably, Visa and Mastercard are not resisting stablecoin settlement; they are actively integrating it into their networks. Visa has supported USDC settlement on Ethereum and Solana since 2021, with annualized settlement volume reaching $7 billion. Mastercard acquired BVNK months ago to expand its stablecoin infrastructure. Both networks have even stated: "Stablecoins won't disrupt the existing payment landscape; instead, they will entrench this system."
The major card networks aren't being disrupted or replaced by stablecoins; on the contrary, they are absorbing stablecoins as an upgrade to their own settlement layer. Every stablecoin card running on Visa, while proclaiming to disrupt it, contributes to its transaction volume and continues to pay its fees.
As an additional note, in our Stripe article, we mentioned that stablecoins can indeed reduce cross-border payment costs by removing multiple correspondent banks—that argument holds. But the context matters. The core pain point of cross-border payments is FX and multiple intermediaries, while in-country consumer spending has a completely different fee structure. The industry is applying a logic proven for cross-border payments directly to in-country spending scenarios, which doesn't match the real cost structure.
The logic of stablecoins empowering cross-border payments is solid, but once applied to in-country spending, analyzing the fee flow reveals this narrative struggles to hold up.
So, what can stablecoins ultimately replace? Objectively, the answer is the clearing/settlement layer and cross-border FX conversion. They transform the underlying funds movement channel behind a transaction. However, the largest cost drivers in an in-country card transaction were never in the settlement infrastructure.
