The European Central Bank is no longer studying tokenization from the sidelines. On October 1, 2026, ECB Executive Board member Isabel Schnabel presented three distinct architectural models for placing central bank money directly onto distributed-ledger-technology (DLT) platforms, speaking at the...
"Central banks could modernise monetary policy implementation by going on-chain." — Isabel Schnabel, Member of the ECB Executive Board
The European Central Bank is no longer studying tokenization from the sidelines. On October 1, 2026, ECB Executive Board member Isabel Schnabel presented three distinct architectural models for placing central bank money directly onto distributed-ledger-technology (DLT) platforms, speaking at the Bank of England's Future of Money conference in London. The speech arrived ten days after the Eurosystem launched Pontes, a live wholesale settlement rail that lets 13 financial institutions — including Deutsche Bank, Santander, Société Générale, and the European Investment Bank — settle tokenized assets in central-bank money.
Schnabel's framework addresses a structural question that private stablecoins and tokenized deposits have left unanswered: what serves as the risk-free settlement layer when financial markets move onchain. The ECB's answer is that central bank reserves must follow. Each of the three proposed models preserves the two-tier monetary system while adapting it for programmable infrastructure. The initiative puts the ECB ahead of most peer central banks in operational deployment, though Switzerland's Project Helvetia has settled CHF 750 million in wholesale CBDC on SIX Digital Exchange since 2023, and the BIS's Project Agorá demonstrated atomic cross-border settlement across seven central banks and 40+ financial institutions in May 2026.
Schnabel's October 1 presentation laid out three architectures for bringing central bank money onto DLT platforms. Each model addresses a different trade-off between integration depth, operational risk, and central bank control.
Model 1: Direct Issuance. The central bank issues reserves directly as native tokens on a programmable ledger. This is the deepest form of integration. Reserves exist on the DLT platform itself, enabling atomic settlement — where the asset leg and cash leg of a transaction either complete together or not at all. The model maximizes programmability: smart contracts can automate collateral substitution, margin calls, and repo settlement without intermediary steps. The trade-off is operational: the central bank must operate or co-operate DLT infrastructure, extending its technology footprint.
Model 2: Interoperability Bridge. Central bank reserves remain in existing real-time gross settlement (RTGS) systems. A synchronization layer connects those systems to external DLT platforms via cryptographic linking (hash-based verification). Settlement finality still occurs in the RTGS system, but the DLT platform can trigger and confirm settlement events. This model minimizes change to existing central bank operations but sacrifices full atomicity — settlement across the bridge introduces timing dependencies.
Model 3: Reserve-Backed Private Settlement Tokens. Private intermediaries deposit reserves at the central bank and issue tokens on DLT networks, backed one-to-one by those reserves. The tokens are private claims on the intermediary, not direct central bank liabilities. This model distributes operational complexity to the private sector but introduces credit risk on the token issuer and requires supervisory oversight of reserve backing — functionally similar to the requirements imposed on stablecoins under regimes like MiCA or the GENIUS Act.
Schnabel framed the core argument in economic terms: tokenization enables "faster, safer, and smarter" settlement through programmability and atomicity. She argued that stablecoins lack the independent capacity to expand liquidity during financial stress, making central bank reserves structurally superior as the settlement layer. The policy trade-off, she noted, lies between unified ledgers (which maximize atomicity but concentrate risk) and interconnected platforms (which encourage competition but risk liquidity fragmentation).
Pontes went live on September 21, 2026, marking the transition from the Eurosystem's 2024 DLT testing program to production infrastructure. The system links DLT platforms to the ECB's TARGET Services, enabling tokenized wholesale transactions to settle against risk-free central-bank reserves rather than commercial bank deposits or stablecoins.
Participants at launch:
| Category | Institutions | |----------|-------------| | Banks | ABANCA, BayernLB, Caisse des Dépôts et Consignations, Cecabank, Deutsche Bank, Deka Bank, DZ Bank, KfW, Memo Bank, NRW.BANK, Santander, Société Générale | | Supranational | European Investment Bank | | Central Bank | Deutsche Bundesbank | | DLT Operators | Axiology (XRPL-based), Cashlink, Clearstream, SWIAT |
The system currently operates between 8:00 a.m. and 4:00 p.m. CET on business days. The ECB plans to extend operating hours and add enhanced features incrementally, with full implementation targeted for 2028.
ECB President Christine Lagarde stated at the launch: "The Eurosystem is working to enable a more integrated, innovative and resilient European financial market." Executive Board member Piero Cipollone added that "Pontes brings the stability and trust of central bank money to the European tokenised finance."
Pontes bundles the full asset lifecycle — issuance, trading, settlement, custody, and servicing — on DLT infrastructure while anchoring the cash leg in central bank money. The design emerged from the 2024 Eurosystem DLT trials, during which participants identified access to a risk-free settlement asset as the critical prerequisite for institutional adoption of tokenized markets.
If Pontes is the operational rail, Appia is the blueprint. The ECB has published a consultation paper for Appia, a long-term initiative to design the governance, standards, and architecture of a tokenized wholesale financial ecosystem.
Appia is evaluating three architectural options:
The blueprint is due in the second half of 2028. The ECB has explicitly linked Appia to the Capital Markets Union agenda, framing tokenized infrastructure as a potential delivery mechanism for the integrated European capital market that conventional regulatory harmonization has sought but not achieved.
On September 21, concurrent with the Pontes launch, the ECB announced preparatory work to invest a portion of its own funds in tokenized securities. Initial investments will focus on euro-denominated euro area public sector and European supranational securities.
No tokenized purchases have been executed to date. The ECB's Executive Board will determine operational details and timing after completing preparatory work, contingent on developments in tokenized issuances and the broader tokenized ecosystem in Europe. Settlement will occur in central-bank money via Pontes.
This positions the ECB not only as infrastructure operator but as market participant — a dual role intended to generate institutional expertise in DLT-based financial markets. The approach mirrors central bank practice in conventional bond markets, where the ECB holds portfolios for reserve management and monetary policy purposes.
The ECB's framework does not exist in isolation. Multiple central banks are pursuing wholesale CBDC or tokenized settlement infrastructure.
Switzerland — Project Helvetia. The Swiss National Bank has operated the furthest-advanced wholesale CBDC pilot globally. Running on SIX Digital Exchange (SDX), the project has settled approximately CHF 750 million of the CHF 1.4 billion in digital bonds issued on SDX using wholesale CBDC. The pilot runs until at least June 2027. In June 2025, the SNB expanded access to BX Digital, a subsidiary of Germany's Boerse Stuttgart. The SNB has characterized stablecoins as structurally deficient for monetary policy transmission, citing their inability to transmit central bank rate changes.
BIS — Project Agorá. Convened by the BIS and the Institute of International Finance, Project Agorá demonstrated in May 2026 that atomic settlement of cross-border wholesale transactions is technically feasible using tokenized central bank reserves and commercial bank deposits. The project involves eight central banks (Bank of England, Federal Reserve Bank of New York, Bank of France, Bank of Japan, Bank of Korea, Bank of Mexico, Swiss National Bank, Bank of Canada) and more than 40 private financial institutions. In July 2026, the project completed real-value testing: 28 institutions settled approximately CHF 800,000 across 17 transaction scenarios involving six currencies, completing in an average of 80 seconds. The platform proved compatible with ISO 20022 messaging standards.
United Kingdom. The Bank of England has maintained an exploratory stance on wholesale CBDC, hosting the Future of Money conference where Schnabel delivered her presentation. The UK has not launched a production wholesale settlement rail comparable to Pontes.
The infrastructure the ECB is building serves a market that is small today but growing rapidly.
The gap between current market size (~$35 billion) and 2030 projections ($2.7–8.2 trillion) is where infrastructure like Pontes, Appia, and Project Agorá must prove their utility. Settlement in central-bank money addresses one of the core institutional adoption barriers: counterparty and credit risk on the cash leg of tokenized transactions.
Fragmentation. The proliferation of central bank settlement models — Pontes in Europe, Helvetia in Switzerland, Agorá across seven jurisdictions — risks creating incompatible silos. Interoperability between these systems remains unproven at production scale. Project Agorá's ISO 20022 compatibility is a positive signal, but operational interoperability between Pontes and non-Eurosystem DLT platforms has not been demonstrated.
Operating hours. Pontes currently operates 8 hours on business days. DLT markets are designed for 24/7 operation. Until Pontes extends to continuous availability, settlement in central bank money will be constrained to a fraction of the time window that private stablecoins cover.
Model selection. The ECB has not chosen among the three Schnabel models. Each implies different cost structures, risk profiles, and levels of central bank operational exposure. Model 1 (direct issuance) maximizes the benefits but extends the central bank's technology surface area. Model 3 (reserve-backed tokens) effectively recreates the supervisory challenge of stablecoins within the central bank perimeter.
Appia timeline. The 2028 blueprint target means the full tokenized ecosystem architecture will not be defined for at least two years. In the interim, market participants must build on Pontes without clarity on the long-term infrastructure shape.
Liquidity. Schnabel identified the core structural argument: stablecoins cannot expand liquidity supply during financial stress. Whether tokenized central bank reserves can fulfill this function on DLT platforms — particularly during periods of market dislocation when demand spikes — remains untested.
The ECB's three-model framework represents the most comprehensive articulation by a major central bank of how sovereign money should adapt to tokenized financial infrastructure. The question is no longer whether central bank money belongs onchain — Schnabel's speech and the Pontes launch have settled that within the Eurosystem. The question is which model prevails, and whether European infrastructure can achieve interoperability with parallel systems in Switzerland, the UK, and the cross-border BIS framework before fragmentation calcifies.
The economic logic is straightforward: if securities settlement moves to DLT, the cash leg must follow, and the cash leg must be risk-free. Stablecoins and tokenized deposits introduce counterparty risk. Central bank reserves do not. The challenge is operational — extending central bank infrastructure to programmable platforms without introducing new systemic vulnerabilities. Pontes is the first production test of that proposition. Its operating constraints (limited hours, business days only) and the two-year timeline for Appia suggest the full answer is years away. In the interim, the ECB has staked a position: central bank money will be onchain, and the Eurosystem intends to operate the rail.