The Bank of Italy's Cost Autopsy: Stablecoin Settlement Is Cheap. Everything Around It Isn't."

"article": "Banca d'Italia has published a study the crypto industry will read as the moment central banks turned hostile to stablecoins. The headline conclusion: stablecoins do not offer a consistent cost advantage for cross-border remittances. That sentence will enter the policy bloodstream quietly, then resurface in digital euro impact assessments, MiCA implementation reviews, and international forums. The paper's cost decomposition contains a more unsettling truth for the industry than any regulatory threat: blockchain settlement is no longer the problem. The majority of the cost differential, the researchers find, comes from fiat conversion and payment infrastructure. Not gas. Not network fees. Not finality.\n\nThe ledger doesn't lie. Neither do the toll booths that surround it. Every anomaly is a story the data forgot to tell, and this story is about layers. The cost of a stablecoin transfer is now the cost of reaching the chain, not the cost of using it. The core has been optimized to near-zero. The boundaries — the on-ramps and off-ramps where digital tokens touch fiat money — still carry the full freight of legacy finance. Banca d'Italia has performed an autopsy. The cause of death is not what the coroner announced.\n\nContext: A Central Bank with a Policy Stake\n\nBanca d'Italia is not a detached academic institution; it is a founding pillar of the Eurosystem, the monetary architecture that operates the euro. Its research staff feeds directly into European Central Bank deliberations, and its publications carry institutional weight far beyond the academy. When it speaks about payment instruments, European regulators listen.\n\nThe timing is deliberate. MiCA, the European Union's landmark crypto-asset regulation, began phased implementation in 2024 and has rolled out stablecoin-specific requirements through 2025. European authorities now possess a regulatory framework but insufficient empirical evidence about what stablecoins actually do in the real economy. This study is an early, serious attempt to generate that evidence, and its policy orientation is legible in every line.\n\nThe market context matters equally. Stablecoins have become a monetary category of their own: hundreds of billions in circulating supply, trillions of dollars in annual settlement volume, and a growing role in dollar access beyond the United States. But settlement volume is not payment volume. The dominant share of stablecoin transaction flow remains exchange-internal — collateral rotations, market-making activity, institutional transfers, DeFi positions. The retail remittance corridor — a migrant worker in Turin sending €300 monthly to Dakar — accounts for a sliver. This study targets that sliver, and concludes that the stablecoin instrument, measured end to end, does not consistently beat traditional transfer rails on price.\n\nThe study is thin in public details. It names no specific stablecoins, discloses no corridor sample, and publishes no raw dataset. This absence of data is itself a data point. The Bank of Italy chose to publish a conclusion without the underlying evidence trail, in sharp contrast to the open-data conventions of academic economics. That choice signals the intended audience: not peer reviewers, but policymakers already familiar with the contours of the European stablecoin debate. The study is best read as a positioning document with economic furniture. Yet the analytical spine — that cost differentials trace to fiat conversion and payment infrastructure rather than blockchain fees — is precise enough to anchor a decade of policy debate.\n\nCore: The Cost Stack Is a Three-Layer Sandwich\n\nEvery stablecoin transaction carries three distinct prices. The first is the entry price, paid when fiat converts into tokens at an on-ramp. The second is the settlement price, paid to the blockchain for processing and finalizing the transfer. The third is the exit price, paid when tokens convert back into local currency on the receiver's side. The end-to-end cost is the sum of all three.\n\nReduced to arithmetic, Banca d'Italia's analysis states that the middle term has collapsed while the boundary terms still demand rent. On a modern layer-2, a USDC transfer settles for fractions of a cent. On a congested day, a small retail transfer might cost two or three cents. Structurally negligible. The on-ramp, by contrast, typically extracts two to three percent on a card-funded purchase before spread. The off-ramp takes another bite on the return journey — spread, liquidity fees, local payment network surcharges. Between the boundaries sits compliance: KYC verification, sanctions screening, correspondent bank charges, and the operational cost of maintaining licensed fiat access.\n\nAdd the layers, and the end-to-end result returns to a range that resembles the legacy system it was supposed to replace. The study's conclusion — no consistent cost advantage — is therefore not a critique of blockchain technology. It is an audit of the interface layer, and it exposes a structural reality the industry spent five years avoiding: stablecoins are a high-efficiency settlement core wrapped inside a legacy-compatible shell.\n\nCore: The Term the Industry Optimized Was Never the Binding Constraint\n\nThis is the uncomfortable insight. The industry devoted massive engineering resources to reducing the settlement term. Layer-2 scaling, cheaper consensus, compressed calldata, specialized payment chains — all of it aimed at driving the middle of the cost stack toward zero. The Bank of Italy's finding is that the middle is already negligible while the final cost remains high. The majority of the price sits in fiat conversion and infrastructure, categories that no consensus mechanism change can fix.\n\nMy own work taught me this pattern. During DeFi Summer in 2020, I built a Python-based backtesting engine to simulate yield farming across Compound and Uniswap, tracking more than ten thousand swap events to quantify slippage behavior under volatility. Initial results suggested persistent arbitrage opportunities in early Aave deployments. The apparent edge vanished once I modeled MEV activity — bots extracting value at exactly the boundaries where naive accounting failed to look. The lesson reapplies directly: the visible cost is rarely the real cost. The hidden cost lives in the interfaces between systems, not inside the systems themselves.\n\nThe same logic governs stablecoin payments. The visible cost — the chain fee — is a rounding error. The hidden cost — the fiat bridge — is the entire price. The central bank has stated, in institutional prose, what every active stablecoin user already knows: the on-ramp is the fee.\n\nCore: The Recurring Lesson of the Terra Ledger\n\nI have seen this boundary-level reconciliation failure before. In 2022, I monitored TerraUSD's reserve ratios daily, applying statistical frameworks to the distance between on-chain stablecoin supply and the real-world collateral backing it. The model flagged a divergence weeks before the collapse: the ledger's internal accounting had separated from outside reality, and every protocol built on the assumption of their equivalence was borrowing against air. The generalization holds across the ecosystem: the chain records what happens inside the system with perfect fidelity, and it is silent about the system's boundaries.\n\nThe Bank of Italy study performs the same kind of boundary-level reconciliation for payments. The blockchain records the settlement leg faithfully and cheaply. But the invoice that reaches the end user includes entries no blockchain can record — bank interface fees, compliance overhead, currency spreads, liquidity provider margins. The total cost of the transfer lives outside the network's audit trail. The central bank simply read the complete invoice. That conclusion should surprise no one who has ever reconciled the difference between a token balance and a bank balance.\n\nCore: Speed Is Real, but It Is Not Enough\n\nBanca d'Italia's conclusion targets cost, not speed. This is a distinction the market should respect. Stablecoins settle in minutes on public networks, at any hour, on any day of the week. Traditional cross-border rails operate on banking hours, batch processing, and correspondent relationships that can delay settlement for days. The study does not dispute this speed advantage; its cost analysis implicitly accepts it. That timing gap is a genuine product feature, and it does not disappear merely because the price advantage evaporates.\n\nBut speed alone is not a business model. Settlement speed only matters if the transaction can complete the full journey, including the fiat gates at both ends. A transfer that reaches the blockchain in ninety seconds can still sit for hours inside an off-ramp's compliance queue before the receiver sees local currency. The end-to-end time, like the end-to-end cost, is dominated by the boundary layers. The same structural pattern the study identified in money applies to time.\n\nThe industry's response should be clear-eyed. Speed is a necessary condition for premium use cases — urgent large-value transfers, interbank liquidity management, time-sensitive commercial settlements. It is not sufficient for the mass-market remittance business, where price dominates. The value proposition narrows to high-urgency corridors. That is a smaller market than the narrative claimed, but it is a real one.\n\nCore: The Remittance Pie and the Stablecoin Slice\n\nThe global remittance market is vast. The World Bank tracks cross-border worker transfers in the hundreds of billions of dollars annually, with a global average cost around six percent. Sub-Saharan Africa averages higher, sometimes above seven percent, with specific corridors reaching double digits. The industry's pitch has always been simple: replace a six percent instrument with a one percent instrument, and the savings flow directly to the world's poorest households.\n\nThe traditional baseline itself is heterogeneous. Bank-to-bank transfers through correspondent networks can cost three to five percent for the sender, but a growing share of remittances now moves through digital-first fintechs such as Wise and Remitly, which have compressed advertised prices to under one percent in many corridors. The competition is therefore not simply stablecoin versus an archaic bank wire; it is stablecoin versus an already-optimized fintech stack that also benefits from modern compliance automation. The competitive bar has risen while the industry narrative keeps fighting an older, weaker opponent.\n\nThe Bank of Italy's study challenges the one percent premise. The actual cost, measured at the user level, includes the fully loaded fiat conversion at both ends. A worker with a bank account can fund a purchase through a regulated exchange at reduced rates. A worker without a bank card faces cash-based on-ramps with far higher fees. The profile of the typical remittance sender — often unbanked or underbanked, often operating in cash — is exactly the profile that faces the most expensive fiat conversion. The people the narrative claims to serve are the people least able to access the cheap middle layer.\n\nThis is the cruel arithmetic at the core of the study. The low cost of blockchain settlement is captured mainly by users who already have access to the traditional financial system. The unbanked, who were supposed to be the primary beneficiaries of the payment revolution, must pay the highest prices to reach the chain. The inclusion story and the cost story are in tension. The central bank has quantified one side of that tension.\n\nCore: A Quiet Technological Validation\n\nHere is the counterintuitive read that most market commentary will miss. When a central bank reports that blockchain fees are not the source of cost differentials, it is certifying the blockchain layer's economic efficiency. The statement implies that the settlement component of stablecoin infrastructure has become cheap enough to remove from the end-user cost equation. That is a technological endorsement, embedded inside a policy-oriented critique.\n\nThis matters for the ongoing infrastructure war among layer-2 platforms. The competition between OP Stack and ZK Stack variants, and between general-purpose networks and payment-specific chains, has long centered on transaction prices. Marketing materials contrast sub-cent fees. Battle lines are drawn over gas schedules. Banca d'Italia's finding suggests the war over the cheapest settlement is being fought over a term that no longer determines the price a user pays. The real differentiator was never purely technical; the platform that convinces more projects to deploy on its stack, and connects them to the fiat rails that matter, wins the actual market. The research reinforces that view. The binding constraint for payment adoption sits outside the settlement race entirely.\n\nCore: The Policy Game Theory Beneath the Prose\n\nCentral banks do not publish research in a vacuum. Banca d'Italia is a Eurosystem member, and the European Central Bank is actively designing the digital euro. In that context, a study concluding that private stablecoins lack a payment cost advantage performs an implicit policy function. It weakens the economic case for stablecoin-based payment adoption inside the European Union, and it strengthens the positioning of a central bank digital currency engineered for zero-friction deposit conversion.\n\nThe MiCA connection is equally direct. The regulatory framework imposes substantial compliance obligations on stablecoin issuers and on the fiat channels connecting them to the traditional financial system. Those obligations are not abstract policy texts; they materialize as KYC verification costs, licensing fees, and reporting overhead — precisely the categories Banca d'Italia identifies as the dominant cost drivers. The logic becomes circular in a way that favors the regulator: compliance burdens raise the cost of stablecoin payments, the elevated cost is cited as evidence that stablecoin payments lack economic value, and that evidence justifies further regulatory stringency.\n\nThe digital euro angle deserves explicit attention. A retail digital euro designed as a direct central bank liability, with conversion at par and no commercial intermediary margin, would eliminate both fiat boundary costs in a single stroke. The study's quiet implication: the public sector can settle at zero and connect to existing bank accounts at zero, while private stablecoin issuers must pay for both privileges.\n\nTraditional banks can read the same evidence differently. For incumbent financial institutions, the study is a comfort: stablecoins do not pose an existential threat to the correspondent banking model. That relief could accelerate bank-crypto partnerships on terms set by banks — stablecoins integrated as settlement layers inside existing rails, rather than replacing them. The competitive threat shifts from substitution to assimilation.\n\nThese dynamics belong to game theory rather than economics. The research hands the European policy complex a citable anchor. It can be deployed in digital euro impact assessments, in MiCA implementation reviews, and in international forums where the Bank for International Settlements and the International Monetary Fund coordinate central bank perspectives. A single study is not a consensus. It is the opening move in a sequence that could become one.\n\nCore: Market Consequences — Who Pays and Who Profits\n\nAt the market level, the finding lands unevenly. The most exposed assets are payment-token names — projects such as Stellar and Ripple, whose valuation narratives depend on the claim that they displace traditional remittance infrastructure through lower cost. A central bank with a credible empirical study is a heavier adversary than a skeptical blogger. If mainstream financial media amplifies the finding, those tokens carry a psychological overhang unrelated to their technical roadmaps.\n\nStablecoin issuers themselves face a subtler exposure. If the payment use case stalls, the issuer business model narrows to reserve management — interest earned on Treasury bills and money-market instruments backing the token supply. The issuer becomes, in economic substance, a bond fund with a blockchain wrapper. That business remains profitable in a high-rate environment, but it no longer constitutes a payment revolution. The growth narrative for stablecoin supply shifts from the cheap rails of global payments to the digital representation of dollar yields. The research accelerates that shift and grants it institutional legitimacy.\n\nThe same caution applies to subsidized adoption. Payment programs that offer cashback or token incentives to users resemble liquidity mining in one crucial respect: activity disappears when the subsidy does. If stablecoin remittance flows require continuous incentives to overcome fiat gate costs, the measured volume is a function of the marketing budget, not a user preference. Compounding errors are just debt in disguise.\n\nLiquidity is the oxygen; volatility is the breath. The market's immediate reaction to this paper will be quiet — central bank research does not trigger liquidations. But the long-run repricing of payment narratives will unfold through precisely the slow institutional attention that never appears on a price chart.\n\nCore: The Interface Economy Becomes the Investment Case\n\nEvery study has a beneficiary, and here the beneficiaries are fiat-ramp providers. If the cost bottleneck sits at the boundary between fiat and crypto, economic value across the entire payment stack concentrates at that boundary. Companies building compliant on-ramps and off-ramps, licensed fiat-crypto gateways, stablecoin-compatible payment cards, and bank-integrated settlement APIs become the essential infrastructure of any stablecoin payment future.\n\nVenture flows will follow the cost line. A research conclusion identifying fiat conversion as the dominant cost is, in effect, a map of where innovation can create the most value. Expect increased investment in regulated stablecoin banking infrastructure, cross-border ramp networks, and compliance middleware. Expect the payment-chain marketing narrative to shift from cheapest transaction to deepest bank connectivity. The layers that move money between the traditional and tokenized economies will capture the premium.\n\nThe ecosystem's real bottleneck has moved from on-chain to off-chain, and the industry should reallocate accordingly. Trust is a variable, not a constant. The chain produces settlement trust at near-zero marginal cost. The fiat gates, however, remain permissioned, slow, and expensive. They are the new battleground.\n\nContrarian: The Forensic Caveat — Correlation Is Not Causation\n\nNow the forensic caveat. Correlation is the ghost; causation is the corpse. Banca d'Italia observed that stablecoin remittances do not consistently undercut traditional channels on price. Observation is not explanation, and the study's thin public methodology invites skepticism about its scope. If the sample corridors were predominantly within Europe or between high-banking-penetration economies — where SEPA transfers settle for free in seconds — then the comparison loads the dice against stablecoins by construction. In corridors where legacy infrastructure is expensive or absent, the result could invert.\n\nConsider the West Africa corridor. Traditional agent-banking remittance costs routinely exceed seven percent and can climb toward double digits at rural collection points. A stablecoin transfer can bypass correspondent banking entirely, settling the intermediate leg on-chain at near-zero cost. The residual friction is the local cash-out network, and that infrastructure is improving from both directions — informal peer-to-peer markets in Lagos and Accra are already quoting stablecoin-to-naira rates at a fraction of agent-banking margins. The conclusion no consistent cost advantage is fully compatible with a massive, corridor-specific advantage in precisely the markets international remittances serve most.\n\nThree structural risks weaken the study's evidentiary weight. First, its public summary lacks a methodology appendix — no sample design, no stablecoin selection criteria, no corridor list. Second, the research has not been subjected to conventional peer review. Third, the institutional position of the authors invites a conflict-of-interest question: a central bank evaluating a private instrument that competes with its own planned digital currency has a stake in the answer. None of this invalidates the finding. All of it argues for treating the conclusion as directional rather than settled.\n\nThere is also a self-fulfilling policy risk