Stablecoin remittances can be surprisingly expensive, according to a new experiment by the Bank of Italy, which found costs climbing as high as 9% in some payment corridors and offering no consistent advantage over traditional money transfer services.
The July 2026 study used “mystery shopper”‑style testing, sending real transfers of 200 USDC across ten corridors connecting Italy with Argentina, Brazil, South Africa, the United Arab Emirates and Japan. Researchers tracked the full journey: buying the stablecoin, moving it on-chain, converting it back into local currency and withdrawing the funds.
Total user costs varied dramatically by route, ranging from just 0.30% on the cheapest transfer to nearly 8.96% on the most expensive. The blockchain component itself was a minor part of the bill, averaging only 0.4% of the amount sent. Most of the expense came from the fiat “edges” of the process: exchange spreads, card or bank funding fees, withdrawal charges and foreign‑exchange conversion.
The authors concluded that, based on this sample, stablecoins offered “no systematic cost advantage” over established remittance channels. In other words, sometimes USDC was cheaper, sometimes more expensive, but there was no consistent pattern of savings.
Differences between corridors were stark. The Italy‑to‑Argentina payment delivered the best result, with an all‑in cost of just 0.30%. Yet sending money in the opposite direction, from Argentina to Italy, cost 8.96%, almost thirty times more. The study cautioned that the ultra‑low cost of outbound transfers from Italy to Argentina partly reflected Argentina’s dual exchange‑rate regime: the gap between official and market FX rates lowered the apparent cost, making that route look more efficient than it might be in a more transparent currency system.
Brazil showed more moderate spreads. The Brazil‑to‑Italy transfer cost 2.21%, while sending money from Italy to Brazil cost 2.70%. South African routes were more expensive: 5.44% from South Africa to Italy and 4.58% from Italy to South Africa. Transfers involving the United Arab Emirates were among the priciest, coming in at 7.20% and 8.95% depending on direction. In those Gulf corridors, funding the purchase of stablecoins with cards and then withdrawing funds locally significantly pushed up the total cost.
To benchmark performance, Banca d’Italia compared its USDC results with simulated quotes from Wise for the same nominal amount of 200 dollars. Stablecoin transfers came out cheaper in only three corridors: Italy to Argentina, Italy to South Africa and Brazil to Italy. In four others, including both UAE routes and Italy to Brazil, the Wise simulations appeared more economical.
The researchers stressed that these comparisons have limits. Transfers and Wise simulations were executed on different days, so FX conditions and fees may not have been identical. The World Bank remittance data used as another reference point was treated as an approximate benchmark, not as a direct like‑for‑like test. The experiment also focused on a single stablecoin and a small set of transactions, meaning the conclusions “cannot be readily generalized” to every token, service provider or remittance corridor worldwide.
Speed showed a similar pattern: the blockchain leg was fast, but the overall experience depended on domestic banking rails. On‑chain transfers were completed in under 15 minutes in seven of eight directly comparable corridors, reflecting the efficiency of the underlying network. However, users ultimately care about when the recipient can spend local currency, not when tokens settle on-chain.
Where advanced instant‑payment systems existed, the full end‑to‑end time was short. Transfers using Italy’s TIPS system, Brazil’s Pix and Argentina’s Transferencias 3.0 typically finished in under 20 minutes from initiation to local payout. In contrast, South African routes relied on standard bank transfers, stretching total settlement to one or two business days. The study’s key takeaway: stablecoin rails and domestic fast‑payment systems behave as complements, not substitutes. To deliver a truly rapid experience, both the blockchain and the local banking infrastructure need to be modern and interconnected.
Japan posed a separate set of challenges. The Japan‑to‑Italy transfer showed a relatively modest cost of 1.6%, but regulatory constraints required using an unhosted wallet and splitting the amount into several smaller transactions. That fragmented structure made it impossible to compare timing cleanly with other corridors. The reverse route, Italy to Japan, recorded an estimated cost of 1.3%, but the last conversion step-from crypto back into yen in a recipient’s bank account-was not fully completed, again limiting the strength of any speed conclusions.
Overall, the findings align with Bank of Italy Governor Fabio Panetta’s earlier comments in May. He argued that while stablecoins can be useful in specific corridors or niches, they are not a universal remedy for high remittance costs. Instead, he urged policymakers to upgrade domestic payment systems and interconnect fast‑payment networks across borders so that money can move quickly and cheaply in local currencies without relying solely on private crypto infrastructures.
A March 2026 paper from the Bank for International Settlements reached a similar diagnosis. It highlighted poor interoperability between systems, inconsistent technical standards and institutional frictions as the main brakes on cheaper cross‑border payments. From that perspective, simply placing blockchain in the middle of the process cannot resolve compliance checks, banking‑system delays, or the complexity of converting into and out of local currencies.
At the same time, industry pilots continue to explore the opposite thesis: that tightly designed stablecoin setups can outperform legacy rails in specific contexts. Some providers report competitive pricing across hundreds of business‑payment corridors, and corporate trials have demonstrated high‑value transfers-such as a 20,000‑dollar payment between North American and Latin American banks-settling in mere minutes. Others are building dedicated infrastructure for stablecoin‑based remittances in partnership with local financial institutions.
These case studies do not necessarily contradict the central bank’s findings. They often involve different transaction sizes, business customers instead of retail senders, negotiated fee schedules, or direct integration with payment processors and banks. In such controlled environments, many of the frictions seen in the Bank of Italy experiment-like card funding charges or retail exchange spreads-can be reduced or removed.
For consumers, the study underscores that the “headline” promise of cheap blockchain transfers can be misleading if one ignores the full journey from salary account to family member’s bank balance. The network fee on a stablecoin transaction may be a few cents, but if a user pays several percent in card funding costs, plus another few percent in hidden FX spreads when cashing out, the end result can rival or even exceed traditional remittance prices. Transparency on spreads, funding options and withdrawal fees matters as much as the choice of token.
The corridor‑by‑corridor differences also highlight how local regulation, FX regimes and financial infrastructure shape outcomes. In Argentina, divergent exchange rates influence apparent costs and create arbitrage‑like opportunities that might not exist elsewhere. In Japan, strict rules around custody and transfers complicate straightforward wallet‑to‑wallet comparisons. In countries without instant‑payment systems, the “last mile” can overshadow any speed gains achieved onchain.
For policymakers, the experiment offers a roadmap for where reforms could have the biggest impact. Connecting instant‑payment systems across borders, harmonizing technical standards, and clarifying regulatory treatment of both stablecoins and digital wallets could shrink the role of costly intermediaries. Conversely, leaving domestic banking rails slow and fragmented ensures that even the most efficient blockchain will confront bottlenecks at the entry and exit points.
For stablecoin issuers and remittance startups, the message is that competing on blockchain fees alone is not enough. The decisive battleground is the fiat on‑ and off‑ramp: integrating cheaper funding methods like bank transfers instead of cards where possible, negotiating lower withdrawal fees with local partners, and providing clear, upfront disclosure of FX rates and spreads. In some corridors, partnering directly with banks or payment institutions may matter more than using the “cheapest” chain.
User education will be crucial as well. Many migrants and overseas workers choose services based on brand familiarity, agent networks, or perceived safety rather than marginal cost differences. If stablecoin‑based offerings are to gain traction, they must not only undercut incumbents on price and speed in a consistent way, but also deliver intuitive interfaces, robust customer support and strong consumer protection-especially in jurisdictions where refunds and dispute resolution are less developed.
Finally, the Bank of Italy study illustrates why broader and more granular testing is still needed. The next wave of research will have to cover multiple stablecoins, a wider set of providers, different corridor types (including high‑volume routes like Europe‑to‑Asia or Gulf‑to‑South Asia), varied transaction sizes and a mix of retail and corporate use cases. Only with detailed disclosure of exchange spreads, funding and withdrawal costs, and precise payout times will it be possible to say in which situations stablecoins truly lower remittance costs-and where traditional rails, once upgraded, remain the more efficient option.