Bank of Italy Study: Stablecoin Remittances Cost More Than Traditional Methods in 40% of Routes

Woofun AI reports that a July 2026 research paper from the Bank of Italy challenges the prevailing assumption that stablecoin remittances are universally cheaper than traditional methods. Authored by Alberto Di Iorio, Enrica Di Stefano, Michele Mascioli, and Giorgio Trebeschi, the study demonstrates that while on-chain costs are minimal, ancillary fees frequently erode the economic advantage of using USDC for cross-border payments.

The methodology involved executing real transfers of 200 USDC across ten distinct routes connecting Italy with Argentina, Brazil, South Africa, the United Arab Emirates, and Japan. Rather than relying on advertised fee structures, the researchers conducted actual transactions to capture the true cost of transfer. This empirical approach was published in Banca d’Italia’s Markets, Infrastructures, Payment Systems series, with the authors noting that the conclusions reflect their individual views rather than an official institutional position. The selection of these specific corridors allowed for a granular analysis of how local financial infrastructures impact the final cost of remittances.

Structurally, the cost analysis revealed that the onchain stage averaged only 0.4% of the amount sent, a figure that is often cited as the primary benefit of stablecoins.

However, the largest charges appeared before USDC was transferred or after it reached the recipient, specifically through card surcharges, exchange spreads, and fixed withdrawal fees. Because each transfer was worth only $200, these fixed and percentage-based off-chain costs had a disproportionately large effect on the total expense. This finding underscores that the blockchain fee is merely one component of a much larger cost structure that includes funding and cash-out mechanisms.

In the UAE-to-Italy case study, the total cost reached 8.95%, significantly higher than many traditional alternatives. A credit-card surcharge of 3.8% contributed to a combined funding and USDC purchase cost of 6.17%, illustrating how fiat-onboarding mechanisms can negate the efficiency of the underlying asset. The high cost was driven by the friction between traditional banking instruments and crypto-exchange platforms, where users are penalized for using credit cards to purchase digital assets. This route highlights the vulnerability of stablecoin remittances to the pricing policies of centralized exchanges and payment processors.

The Argentina-to-Italy route was slightly more expensive at 8.96%, with 5.36% spent while purchasing USDC. This calculation was heavily influenced by the gap between Argentina’s official peso exchange rate and the rates available through crypto markets. Using the more market-oriented Dólar MEP rate would have reduced the estimated cost to approximately 8.5%, but the route would still have been expensive. This discrepancy demonstrates how local currency conditions and regulatory arbitrage can alter the headline result, making stablecoin transfers less attractive in economies with significant exchange rate distortions.

When compared with Wise on the same bilateral routes, USDC was cheaper in three corridors and more expensive in four. One additional route was unavailable for a complete comparison, leaving the data inconclusive for that specific path. The two Brazil transfers show why direction mattered: both transfers used the same stablecoin, but the platforms, funding methods, spreads, and withdrawal arrangements differed at each end. The UAE-to-Italy route showed an even larger gap, with USDC costing 8.95% compared with approximately 1.02%–1.03% through Wise, with the card-funding surcharge accounting for much of the difference. This comparison emphasizes that the relevant metric is the full route available to a specific sender and recipient, not the blockchain fee in isolation.

Speed analysis revealed that the blockchain stage took less than 15 minutes in seven of the eight directly comparable corridors. End-to-end transfers finished in under 20 minutes when both sides had access to fast local payment systems, including Brazil’s PIX and Argentina’s Transferencias 3.0.

However, transfers involving South Africa took one to two business days because deposits and withdrawals depended on standard bank processing. USDC moved quickly between the exchanges, but it could not shorten the time required to fund an account or withdraw money through slower banking rails, indicating that speed benefits are contingent on local infrastructure quality.

Woofun AI data shows that the researchers tested a complete "stablecoin sandwich," where the sender deposited fiat currency, bought USDC through a centralized exchange, transferred it to another platform, and the recipient converted it back into local currency. Binance and Kraken were used in Italy, while platforms in other countries included Ripio, Foxbit, BitOasis, and VALR. This approach captures the experience of customers using identifiable exchanges together with bank accounts or payment cards. It does not represent every way people use stablecoins for cross-border payments, as some users trade through local P2P markets, brokers, or agent networks. Others receive USDC and keep it as a dollar-denominated asset rather than paying for another conversion into local currency, a scenario the paper describes as an "open stablecoin sandwich."

The results are therefore specific to one stablecoin, a $200 transfer size, selected exchanges, and ten Italy-linked routes tested over two days in March 2026. They show why low blockchain fees alone are not enough to judge a remittance service: the meaningful figure is how much usable value reaches the recipient after the complete transfer. This marks a critical shift in evaluating digital asset utility, moving the focus from technical efficiency to holistic economic viability. The study suggests that without addressing off-chain friction, stablecoins may fail to deliver on their promise of cheaper global payments.