Bank of Italy test finds USDC remittances lack a consistent end-to-end edge
Across 200 transfers, costs ran from 0.3% to nearly 9% and speed depended on local instant-payment rails.
A Bank of Italy experiment running 200 USDC remittances across 10 bidirectional corridors found stablecoins did not deliver a systematic cost or settlement-time advantage once fiat conversion and local payment rails were included. The results put most of the “cheap and fast” narrative pressure where users actually feel it: exchange fees, FX spreads, and whether the recipient country has instant payments.
Key Takeaways
- A Bank of Italy research team executed 200 USDC remittances across 10 bidirectional corridors linking Italy with Brazil, Argentina, Japan, the United Arab Emirates, and South Africa.
- End-to-end results showed no consistent cost or speed advantage versus traditional remittance channels once on-ramp/off-ramp steps and domestic payment-rail constraints were counted.
- Total costs varied sharply by corridor, from 0.3% to nearly 9%, with exchange fees and currency conversion doing most of the damage while blockchain transaction fees were a small share.
- Transfers cleared in under 20 minutes where instant payment systems were available, but stretched to one to two business days where they were not, and stablecoin routes beat Wise on cost in only three of seven comparable corridors.
Bank of Italy’s USDC Corridor Test Challenges the “Cheaper and Faster” Pitch
The Bank of Italy’s latest stablecoin remittance experiment lands on an uncomfortable conclusion for one of crypto’s most repeated adoption claims: stablecoin-based remittances did not show a systematic end-to-end advantage on either cost or settlement time once the fiat boundary and local payment infrastructure were treated as part of the product.
Researchers ran 200 transfers using USDC, a US dollar-pegged stablecoin, across 10 bidirectional payment corridors linking Italy with Brazil, Argentina, Japan, the United Arab Emirates, and South Africa. The study compared total costs and settlement times against traditional remittance services, then attributed the variance to where it actually occurred.
The core finding is procedural, not philosophical. On-chain transfer costs were not the binding constraint. The binding constraint was everything required to make a stablecoin transfer usable for a recipient who ultimately lives in a local currency system: converting in and out of fiat, and moving value through domestic rails that may or may not settle instantly.
The Numbers Traders Will Quote: 0.3%–9% Costs and <20 Minutes vs 1–2 Days
The headline range is wide enough to break most one-line narratives. Across the stablecoin remittances examined, total costs ranged from 0.3% to nearly 9% depending on the corridor.
That spread matters because it reframes what “stablecoin remittances” even means in practice. A payment corridor is not an abstract route on a blockchain. It is a specific cross-border path between two countries, with its own banking access, FX liquidity, compliance friction, and local payout options. The same stablecoin can look efficient in one corridor and uncompetitive in another, without anything changing on-chain.
The time results were equally conditional. Settlement times were under 20 minutes where instant payment systems were available, and one to two business days where they were not. Instant payment systems, in this context, are domestic rails that allow near-real-time bank transfers. When they exist and are reachable by the off-ramp, the “crypto leg” can be the fastest part of the journey. When they do not, the stablecoin transfer can arrive quickly on-chain and still wait on a domestic payout process that runs on business-day cadence.
The study’s mechanics also clarify what traders should and should not infer from the data. On-ramp and off-ramp steps are the conversion points between fiat money and crypto assets like stablecoins, and they are where fees, spreads, and operational constraints accumulate. The experiment’s results treat those steps as first-order variables, not as incidental implementation details.
Benchmarks That Matter: World Bank’s 6.65% vs Wise in Comparable Corridors
The Bank of Italy results can be made to look flattering or harsh depending on which benchmark a reader reaches for, and that’s the trap.
Using the World Bank’s reported global average remittance cost of 6.65% as a reference point, the study found stablecoin transfers were cheaper in most corridors examined. That comparison is useful as a sanity check against the global mean, but it is not the bar that determines whether stablecoin remittances win share in liquid, competitive routes.
The more relevant comparison in developed corridors is best-in-class fintech pricing, because that is what a user actually sees when they have access to modern remittance apps. On that measure, the stablecoin path looked less dominant. The study found stablecoin transfers were less expensive than Wise in only three of seven comparable corridors.
Two caveats sit inside that sentence. First, the comparison set is only “comparable corridors,” and the excerpt does not specify which corridors were included in that Wise match-up or how the comparability criteria were defined. Second, “cheaper than Wise” is not the same as “cheaper end-to-end for the median user,” because access to specific on-ramps, off-ramps, and payout rails can differ by user segment.
Still, the directional message is hard to ignore. Stablecoins can beat the global average while failing to beat the best price discovery already available in the corridors that matter most for mainstream adoption.
Where the Edge Disappears: FX, Exchange Fees, and Local Rails — Not Gas
What stands out in the Bank of Italy’s attribution is how little of the end-to-end outcome is explained by blockchain fees. The study found exchange fees and currency conversion made up most of the cost, while blockchain transaction fees represented only a small share.
That is the practical rebuttal to the “gas is cheap now” argument. Even if the on-chain leg is near-free, the user still pays to enter and exit the system, and those costs are not uniform. FX spreads widen in thinner corridors. Exchange fee schedules differ. Payout options can force additional steps. The corridor, not the chain, becomes the dominant variable.
The time side has the same structure. The study concluded that investment in domestic instant payment infrastructure could improve the competitiveness of stablecoin-based cross-border payments, because settlement times depended heavily on the quality of local payment rails. In other words, the stablecoin can move in minutes, but the remittance only “settles” when the recipient can spend or withdraw it through local infrastructure.
The authors also point to the only obvious path to a step-change: reduce the number of times value has to cross back into fiat. They wrote, “If stablecoins could be spent directly in the real economy, for goods and services, rents, or school fees, without reconversion into local fiat currency, the economic advantages of stablecoin-based transfers would be substantially higher.”
That quote is doing more work than it looks like. It implies the current remittance pitch is often benchmarking the wrong product. The real product is not “USDC transfer,” it is “USDC transfer plus conversion plus payout,” and the conversion plus payout leg is where the edge disappears.
Regulation as a Cost Variable: MiCA, the GENIUS Act, and the Offshore Leakage Problem
The study frames regulation as part of the efficiency equation, not as an external constraint. It found regulatory design materially affects transfer efficiency: prohibitionist regimes did not fully suppress stablecoin demand and pushed users toward offshore platforms and other unregulated channels, while overly restrictive frameworks increased operational complexity for retail users.
That matters for two reasons. First, it suggests measured “official” efficiency can deteriorate even when underlying demand is stable, because users route around constraints in ways that are harder to observe and often more expensive or riskier. Second, it implies that compliance-heavy frameworks can add friction at the exact points the study identifies as the bottleneck: on-ramps, off-ramps, and payout rails.
The policy backdrop is not hypothetical. The findings were positioned alongside the European Union’s Markets in Crypto-Assets (MiCA) framework and the United States’ GENIUS Act, which the study referenced as governing payment stablecoins. Both regimes shape who can intermediate stablecoin flows, what disclosures and controls are required, and how easily stablecoins can integrate with regulated payment infrastructure.
This is also happening against a market that is still expanding. The stablecoin market was cited at about $307 billion, up roughly 16% over the past year, according to DefiLlama data. Growth in supply and usage can coexist with weak remittance economics in specific corridors, particularly if stablecoins are being used for trading, settlement, and treasury functions that do not require retail off-ramps.
What Comes Next for Bank of Italy tests USDC remittance
The immediate missing piece is granularity. The excerpt does not specify which corridors produced the lowest costs near 0.3% and which pushed toward nearly 9%, and it does not name the three corridors where stablecoin transfers beat Wise on cost. It also does not detail the specific on-ramp and off-ramp providers used, or the study’s full methodology for selecting “traditional remittance services” for comparison.
Those details are not academic. Corridor-by-corridor breakdowns would tell traders whether the low-cost outcomes are concentrated in a small set of highly liquid routes, or whether they generalize across different banking and FX environments. Provider disclosure would clarify whether the results are driven by a particular exchange fee schedule, a particular payout method, or a broader structural constraint.
The second forward signal is whether “direct spend” acceptance expands in the markets that currently force reconversion to fiat. The study explicitly ties larger efficiency gains to recipients being able to spend stablecoins for real-economy payments without off-ramping, which makes merchant and payment integrations a more important variable than marginal improvements in on-chain fees.
Third, domestic instant-payment rail rollouts and upgrades are likely to be the cleanest way to compress settlement times in corridors that currently take one to two business days. The study’s time split is effectively a proxy for whether the last mile is modern.
Finally, implementation details and enforcement posture under MiCA and the GENIUS Act will determine whether regulated stablecoin rails become simpler for retail users or whether activity continues to leak toward offshore and unregulated channels, which the study flags as a predictable response to prohibitionist or overly restrictive regimes.
My Read: Stablecoin Growth Can Coexist With Weak Remittance Economics Until Spendability Improves
The Bank of Italy experiment is being read as a referendum on stablecoins, and it’s more useful as a referendum on where the stablecoin story still breaks. The study doesn’t argue that USDC can’t move value cheaply on-chain. It argues that the remittance product people actually buy includes two conversions and a payout, and those legs are priced and timed by exchanges, FX conditions, and domestic rails that crypto does not control.
The threshold that matters is whether stablecoins can reduce the number of fiat crossings for the median recipient. If the user path remains “buy USDC, send USDC, sell USDC, withdraw fiat,” the study’s cost range of 0.3% to nearly 9% is the right mental model, because it is dominated by exchange fees and currency conversion, and the speed outcome will keep flipping between minutes and business days depending on instant-payment availability. In that world, stablecoins can still look good against the World Bank’s 6.65% global average, but they will keep struggling to beat best-in-class fintech pricing consistently, which the Wise comparison already hints at.
If, instead, the recipient can spend stablecoins directly for the categories the authors name, “goods and services, rents, or school fees,” then the economics change because the off-ramp stops being mandatory. That is the only scenario in the study that reads like a structural advantage rather than a corridor-specific optimization. Regulation then becomes a second-order cost variable: frameworks that make compliant on-ramps and merchant acceptance easier can compress friction, while regimes that are prohibitionist or operationally heavy can push users into offshore routes that look efficient on paper until you price in the real-world detours.
So the clean takeaway is not that stablecoins fail at remittances. It’s that remittances remain a fiat-rail product until stablecoins are spendable enough that the off-ramp becomes optional, and confirmation will look like corridor-level costs clustering near the low end without relying on fiat payout rails to finish the job.
