The European Central Bank has laid out three possible models for putting central bank money onchain, a framework that could ultimately place euro-system reserves directly alongside tokenized securities, bank deposits and stablecoins on distributed ledger technology. Executive Board member Isabel Schnabel presented the options at the Bank of England’s Future of Money conference in London, offering the clearest picture yet of how one of the world’s most powerful central banks intends to answer the rise of tokenized finance.
Three paths to onchain settlement
The first model under consideration would have the central bank issue reserves directly on a programmable platform. In this scenario, central bank money itself would live on the same DLT infrastructure where tokenized assets trade, enabling so-called atomic settlement, the simultaneous exchange of an asset and its payment, eliminating the settlement risk that sits between the legs of a trade in today’s sequential systems.
The second option is more conservative. The ECB’s existing real-time gross settlement system, TARGET Services, would remain exactly where it is, and an interoperability layer would connect it to DLT platforms. Under this approach, the reserves would not themselves be tokenized. Instead, the legacy system and the new tokenized platforms would be linked by hash, synchronizing the two worlds without forcing the central bank to migrate its core plumbing.
The third approach would tokenize reserves held at the central bank and issue settlement tokens fully backed by those reserves. Crucially, those tokens would be private claims rather than claims issued directly by the central bank. In other words, supervised intermediaries would issue tokenized money backed one-for-one by reserves parked at the ECB, a structure that mirrors how regulated stablecoins are designed, but anchored to central bank money rather than commercial bank deposits or government bonds.
Defending the two-tier system
Across all three models, one red line is constant: the two-tier monetary system must survive. Central bank money would remain at the core of settlement, while commercial banks would continue to provide money and financial services to customers. That framing matters politically as much as technically. European policymakers have watched the stablecoin boom with growing unease, and Schnabel herself argued in June that a digital euro is key to countering stablecoin risks. Putting central bank money onchain, in a controlled fashion, is partly a defensive play to keep euro settlement at the heart of tokenized markets.
Schnabel argued that tokenization can make financial transactions more programmable and atomic, allowing the transfer of an asset and its payment to occur together. Tokenized infrastructure could also allow financial assets and money to interact directly on the same or connected DLT networks, a step change from today’s world where securities databases and payment systems speak different languages and reconcile overnight.
Pontes and Appia already in motion
The framework is not merely theoretical. The ECB’s Pontes project launched in September to provide tokenized central bank money for DLT-based transactions, giving banks a live rails experiment for settling tokenized asset trades in central bank money. In parallel, the Appia project is examining different architectures for tokenized markets, considering a unified ledger, interconnected networks, and multiple shared ledgers. Together, Pontes and Appia function as the ECB’s real-world test kitchens for the three models Schnabel described in London.
The timing reflects how quickly institutional tokenization has moved from white papers to balance sheets. Some 71 percent of senior decision-makers at the UK’s largest financial institutions expect tokenization to reshape financial services, according to Lloyds’ 10th annual Financial Institutions Sentiment Survey. Faster payments and settlement were cited as the biggest potential benefit by 60 percent of respondents, followed by collateral and liquidity management at 41 percent, while 77 percent said investment in new and emerging technologies is a growth priority, up from just 41 percent in 2025.
“The real opportunity is to make financial markets work faster, more efficiently and with greater flexibility for clients,” said Rob Hale, co-head of global markets at Lloyds, citing faster settlement, more efficient use of collateral and better movement of liquidity as tangible benefits already within reach.
What it means for crypto markets
For crypto-native observers, the ECB’s framing carries a quiet irony. The design of the third model, fully backed settlement tokens redeemable against central bank reserves, borrows heavily from the stablecoin playbook that European regulators have spent two years constraining under MiCA. The difference is issuer and anchor: the ECB’s version would be born inside the regulated two-tier system rather than emerging from it, as Circle’s USDC or Tether’s USDT did.
It also arrives as the MiCA review kicks off, with Circle and the Hyperliquid Policy Center already pressing Brussels on stablecoin reserve rules and perpetual futures treatment. If central bank money becomes directly available onchain for tokenized market participants, the competitive calculus for euro stablecoin issuers changes fundamentally: their product would no longer be the only tokenized euro in town, and the premium on being the safest settlement asset would shift toward whatever the ECB ultimately sanctions.
None of the three models is a digital euro for retail users, and the ECB has been careful to keep the projects separate. Pontes and Appia are wholesale infrastructure plays, aimed at banks and market infrastructure operators, while the digital euro pilot with 36 payment providers continues on its own track. But the direction is unmistakable. The institution that once dismissed crypto assets is now designing its own onchain money, on its own terms, and the architecture it chooses will shape how tokenized European finance settles for a generation.
model one only survives capped to tokenized securities settlement. full reserves on a programmable platform dies in committee the second anyone whispers retail cbdc access
fair on model one, but model 3 sidesteps that whole fight. supervised intermediaries issue tokens backed 1:1 by reserves at the ECB, basically a MiCA style stablecoin with a harder anchor. and Pontes is already live to test exactly that plumbing
The Pontes link is the tell though. September launch means model 3 is already being stress tested with real banks, not just sketched on a Schnabel slide. If those reserve backed tokens clear wholesale trades cleanly, the stablecoin comparison writes itself.
Funny detail from the Lloyds survey: 77 percent call new tech a growth priority, up from 41 percent the year before. That jump is what really forced the ECB to publish three concrete models instead of one vague roadmap.
Two-tier surviving all three models was the only realistic outcome. The ECB was never going to let non-banks hold reserves directly, no matter how many tokenized collateral pilots the BIS runs.
Schnabel presenting this at the BoE conference and not some EU crypto event tells you who this is actually for. tokenized securities settle atomic, reserves stay at the ECB, two tier survives yet again
schnabel presenting this at the BoE conference in london and not some EU event says a lot. they know tokenized settlement is coming with or without them
atomic settlement is the actual prize here. no more sitting on sequential legs while your counterparty can fail mid trade
The second option is the one they will pick. Nobody at the ECB is putting reserves directly on a DLT platform when a bridge to TARGET does the job.
the second model is the one that wins imo. keep TARGET exactly where it is, bolt on an interop layer, banks dont have to touch their core plumbing
dagmar wanting the interop model is way too optimistic. T2S took a decade to hook up EU CSDs and that was with willing partners. now try that across dlts with banks fighting over standards lol
Fair on T2S timelines, but Pontes is not connecting CSDs across borders. It is one settlement asset on a handful of DLT venues. Narrower scope, fewer chefs, and the banks involved actually want it this time.
sonnen_basis fair on T2S timelines but Pontes went live in September with actual banks on it. narrower scope than bridging CSDs, which is exactly why it ships while the big interop dream slides toward 2030
reply to dagmar: the interop layer is where the whole thing dies tbh. every project that promised bridging between old rails and DLT ended up with a settlement buffer and a support inbox
Eurobasis fair worry, but TARGET2 T2S already bridges old rails for real settlement and nobody calls it a corpse. The ECB did this plumbing dance before, DLT is just the new leg.
eurobasis every bridge project you are describing had a bank in the middle inventing a new fee. a hash link between TARGET and a DLT venue at least keeps the settlement asset at the central bank the whole time
still no date for anything. we get a consultation, then a pilot, then maybe wholesale settlement sometime before 2030 while stablecoins eat the volume meanwhile
Kwame has it right. The Bremen trials ran years ago and we are still at slide-deck stage. Meanwhile tokenized collateral settles at the BIS sandbox today, so the tech is not the blocker here.
model 3 is basically a MiCA stablecoin with a cbdc costume on. two tier survives again, and the Lloyds 77 percent stat is the real news here, banks actually want this plumbing now instead of tolerating it