Stablecoins appear to solve the most visible problem in a remittance: moving a digital claim across a border quickly. That achievement is real, but it covers only one segment of the transaction. A sender still has to fund a wallet, obtain the token, convert currencies when necessary and deliver money in a form the recipient can use. The price of those surrounding steps can outweigh the blockchain fee.
That is the useful lesson from a Banca d'Italia mystery-shopping exercise reported by CoinDesk. The research does not show that stablecoins are always expensive. It shows that they are not systematically cheaper. For investors, the distinction changes where to look for advantage: away from raw settlement speed and toward liquidity, distribution and the integration of regulated local rails.
Settlement is only the middle of the journey
A remittance begins before a token moves and ends after it has arrived on-chain. The sender may pay a bank-transfer or card charge to reach an exchange, a trading fee to acquire a dollar stablecoin and an embedded foreign-exchange spread. The recipient may then pay again to sell the token, transfer the proceeds and withdraw cash or local-currency bank money. Compliance checks and failed-payment support sit around this chain as operating costs even when they are not itemized on the screen.
Banca d'Italia's official account says its preliminary analysis found no systematic cost advantage and observed costs reaching 9% in some corridors. A separate Bank speech on stablecoins identifies on- and off-ramp fees charged by crypto intermediaries as the largest part of total cost in the study.
This does not invalidate a low network fee. It limits what that fee proves. A rail can settle for cents while the customer pays several percentage points to enter and exit it. The economically relevant comparison is therefore delivered local currency against the sender's total outlay, not the cost of the on-chain hop in isolation.
The corridor decides whether the shortcut exists
Cross-border payments are not one market. A liquid euro-dollar route with many banks, fintechs and exchanges bears little resemblance to a thin local-currency corridor with few payout agents. Token liquidity, the number of market makers, banking access at each endpoint and the recipient's need for cash all change the result.
The BIS analysis of cross-border technologies argues that limited interoperability and institutional differences are binding constraints. Technology can shorten a chain, but it cannot by itself harmonize customer-identification rules, open a local bank account, deepen a currency market or create a trusted dispute process. Those frictions explain why an apparently borderless asset still acquires local costs at both borders.
This corridor logic also prevents an overly bearish reading. In a route where a new exchange or wallet creates real competition, stablecoins can reduce prices. In another, a dominant cash-out agent may capture the saving. The same protocol can therefore produce different consumer economics without any change in its technical performance.
For companies building this infrastructure, transaction volume is an incomplete metric. A stronger indicator is how much of the journey a platform controls without worsening the exchange rate: funding options, deep conversion liquidity, compliant payout coverage and customer support. Integration can remove duplicated margins; it can also conceal them. Transparent all-in quotes matter more than a headline fee.
Dollar access can matter more than the fee
Price is not the only reason to choose a payment method. A stablecoin may be available outside banking hours, arrive faster or allow a recipient to retain dollar exposure instead of immediately converting into a fragile local currency. In places with unreliable banks or strict capital controls, that option can be valuable even when a conventional remittance is marginally cheaper.
That counterargument is strongest when the recipient wants the token itself. If both endpoints already operate in stablecoins, fewer conversions are needed and the last-mile problem shrinks. It weakens when wages begin in bank money and household spending ends in cash, because every transition restores an intermediary and a spread.
The BIS 2026 monetary-architecture review reaches a compatible conclusion: on- and off-ramp costs can make the all-in price as high as or higher than a bank transfer. The report's point is not that the technology has no use, but that settlement efficiency should not be confused with a complete retail service. Access, speed and asset preference must be evaluated separately from price.
Interoperability is the durable cost lever
The evidence that would strengthen the investment case is practical rather than promotional. Providers need standardized, repeatable measurements of total cost across many corridors; tighter quoted-versus-realized exchange rates; deeper local liquidity; fewer payout failures; and lower support and compliance costs per completed transfer. A falling network fee alone would not answer the Bank's criticism.
There is also a competitive response from incumbent rails. Banca d'Italia notes that payment systems are adopting common ISO 20022 messaging and expanding links between fast-payment systems. If regulated bank and fintech networks improve interoperability, stablecoins must win on the full service rather than against yesterday's slowest transfer.
The provisional conclusion is narrower than either side's slogan. Stablecoins can be the cheapest route in specific corridors and can deliver non-price benefits elsewhere. They are not a universal bypass around foreign exchange, distribution and regulation. The platforms most likely to create durable value are those that make the last mile cheaper and more reliable — not those that merely make the middle look fast.