A Bank of Italy research paper has found no systematic cost or speed advantage for stablecoin remittances, with the economics of each transfer determined mainly by exchanges, currency conversion and the domestic payment systems at either end.
The authors conducted a mystery-shopping exercise using 200 USDC transfers across ten bidirectional corridors linking Italy with Argentina, Brazil, South Africa, the United Arab Emirates and Japan. In the eight corridors included in the main comparable analysis, total costs ranged from 0.30% to 8.96% of the transferred amount.
The paper was published on July 30 in the central bank’s Markets, Infrastructures, Payment Systems series. Its findings are those of authors Alberto Di Iorio, Enrica Di Stefano, Michele Mascioli and Giorgio Trebeschi and do not necessarily represent the views of the Bank of Italy.
The blockchain leg was rarely the bottleneck
The experiment followed the complete path from local fiat currency into USDC, across a blockchain and back into the recipient’s local bank account. That structure allowed the researchers to separate funding, stablecoin purchase, onchain transfer, stablecoin sale and fiat withdrawal.
The onchain leg averaged 0.4% across the main corridors and was generally the least expensive component. By contrast, exchange trading fees, conversion rates and fixed charges for funding or withdrawal produced most of the variation in the all-in price.
The UAE-to-Italy transfer illustrates the dependence on the surrounding payment infrastructure. Because the sender used a card to fund the purchase, the transaction incurred a surcharge that helped push the total cost to about 9%. In other corridors, bank-transfer funding was free or cheaper.
The lowest measured cost, 0.30% from Italy to Argentina, also requires qualification. The study calculated the result against Argentina’s official retail exchange rate. The authors said using a market-based conversion rate would reduce the apparent advantage, showing how foreign-exchange conditions can influence a stablecoin comparison independently of the blockchain.
Results varied against traditional providers
Stablecoin transfers looked cheaper than broad World Bank country-level remittance averages in most of the study’s comparisons, but the authors warned that those averages were not like-for-like corridor benchmarks. Most stablecoin results were based on a single transaction in a specific bilateral corridor, while the World Bank figures aggregate outbound costs across destinations.
A narrower comparison with Wise produced mixed results. USDC was cheaper in three of seven usable corridors and more expensive in four. The stablecoin transactions took place on March 24 and 26, while the Wise simulations were conducted on April 14, so market timing may also account for part of the difference.
The speed results followed a similar pattern. The onchain transfer completed in less than 15 minutes in seven of eight main corridors and took about 30 minutes in the South Africa-to-Italy direction. End-to-end performance depended on the fiat steps.
Where domestic instant-payment systems supported funding and withdrawal, the complete process generally finished in less than 20 minutes. Where standard bank transfers were required, as in the South African corridors, it stretched to one or two business days. One UAE withdrawal time was estimated from exchange documentation because the researchers could not complete that step.
Implications for payment providers
For remittance companies and payment processors, the study shifts attention from headline blockchain fees to the full service chain. A low-cost token transfer does not guarantee a low-cost payment if users face expensive acquisition, foreign-exchange spreads, withdrawal charges or slow bank settlement.
That makes exchange selection, access to instant payments and local liquidity central design decisions. Providers evaluating stablecoin corridors need to measure the customer outcome from initial funding through final spendable balance, rather than report only blockchain confirmation time or network fees.
The findings also suggest that stablecoins and domestic instant-payment systems can be complements rather than substitutes. Faster local rails improve the on-ramp and off-ramp that a fiat-to-fiat stablecoin transfer still needs. Direct stablecoin spending could remove the final conversion step, but that is a different operating model with its own acceptance, regulatory and liquidity requirements.
A useful experiment with limited scope
The research is a corridor-level experiment, not a universal ranking of stablecoins against conventional remittances. It covers one token, a limited set of jurisdictions and particular exchanges. The paper notes that different stablecoins, regulatory frameworks, market structures and providers could produce different results.
Japan was discussed separately because restrictions required an unhosted wallet and other methodological changes that reduced comparability. The main cost and speed tables therefore focus on eight corridors, even though the broader exercise covered ten.
Within those limits, the central operational finding is consistent: the blockchain is only one component of a cross-border payment. For stablecoin remittances that begin and end in fiat currency, conversion, exchange pricing and domestic settlement infrastructure remain decisive.