ECB Executive Board member: Central banks should also "go blockchain-based".

On August 28th local time, at the 2026 Jackson Hole Economic Policy Symposium, Isabel Schnabel, a member of the Executive Board of the European Central Bank, delivered a speech entitled "Central Banks on-chain." Unlike past discussions by central banks on blockchain and stablecoins that focused more on regulatory and financial stability risks, Schnabel pushed the issue further to the level of financial infrastructure: as securities, deposits, collateral, and other financial assets gradually enter distributed ledgers, should central bank money also go on-chain? Her answer was very clear: central banks should also embrace distributed ledger technology and go "on-chain" themselves. Schnabel believes that tokenization may become one of the most important infrastructure changes in the future wholesale financial markets. It can not only improve the efficiency of securities, repurchase agreements, collateral, and cross-border transactions through atomic settlement and programmability, but may also help Europe address the long-standing problem of financial market fragmentation. However, she also emphasizes that private stablecoins cannot replace central bank money as the ultimate settlement asset in the financial system because even if stablecoins are sufficiently secure, they lack the ability of central banks to flexibly create liquidity during periods of financial stress. Therefore, the future question is no longer just "whether traditional finance should adopt blockchain," but rather how central banks should enter the tokenized financial system and whether central bank reserves themselves should become natively programmable assets on the blockchain. This is also the question the European Central Bank is exploring through its Pontes and Appia projects. Tokenization truly changes more than just settlement speed. According to Schnabel, wholesale finance is one of the most promising application areas for distributed ledger technology. This isn't just because blockchain makes transactions "faster," but because tokenization brings two more fundamental capabilities: atomicity and programmability. Atomicity means that different stages of a transaction either complete simultaneously or none occur, thus eliminating settlement risk. In fact, traditional finance can already achieve partial atomic settlement through mechanisms such as Delivery vs. Payment (DvP), like TARGET2-Securities in the European Central Bank system. However, tokenization further extends this capability to the entire lifecycle of financial products. Taking repurchase agreements as an example, a repurchase involves not only the initial exchange of cash and collateral, but also subsequent collateral replacement, margin calls, and maturity repayment. Traditional models require repeated coordination among multiple institutions, systems, and messages, while on a programmable ledger, these processes can be automatically executed by smart contracts. This is particularly important for cross-border finance. Due to differences in transaction times, time zones, and settlement systems, cross-border transactions often require large amounts of funds and collateral to be prepared in advance. If tokenization can achieve continuous operation and atomic settlement, it could significantly reduce this pre-financing requirement and improve the liquidity and efficiency of collateral utilization. Schnabel particularly emphasizes that this is of particular significance for Europe. European financial markets have long been fragmented along national borders, with different countries possessing different securities registration, trading, and settlement systems. For example, investing in sovereign bonds of multiple Eurozone countries often requires connecting to multiple central securities depositories, which increases operating costs and naturally favors large institutions capable of bearing the costs of complex infrastructure. Therefore, in her view, tokenization is not merely an upgrade to payment technology, but also an opportunity for Europe to reintegrate its capital market infrastructure: instead of continuing to "patch together" existing national systems, it could establish a financial network designed from the outset with cross-border characteristics. Stablecoins can exist, but they cannot replace central bank money. The truly crucial question then arises: if all assets are on-chain, what will be used for settlement? In the crypto market, stablecoins seem like the most natural answer. However, Schnabel clearly argues that even with extremely secure design, stablecoins still cannot become the ultimate settlement asset in the financial system. She provides two criteria for this judgment. First, the final settlement asset must be sufficiently safe, with virtually no credit risk, liquidity risk, or redemption risk. Second, its supply must be elastic, especially capable of rapid expansion during market liquidity crises. A stablecoin fully backed by government bonds and with strict control over maturity risk could theoretically solve the first problem, but would struggle to address the second. Private issuers cannot create new final liquidity out of thin air during market crises, while central banks can. Schnabel, in addressing this, reviewed US monetary history. Before the establishment of the Federal Reserve, the money supply of the US banking system was closely linked to the size of eligible government bonds, which prevented the financial system from expanding liquidity in a timely manner during seasonal surges in funding demand or panic. The banking crisis of 1907 was one of the key factors that led to the enactment of the Federal Reserve Act of 1913. The real problem that modern central bank systems solve is not simply issuing a more credible currency, but creating an institution capable of flexibly providing liquidity during periods of financial stress. Therefore, Schnabel does not deny the value of stablecoins. She believes that, if properly designed and regulated, stablecoins can enrich the payment tools available to residents and businesses and serve new digital economic activities; however, their role should be a supplement to central bank money, not a replacement. Even if the future financial system is highly tokenized, the underlying "settlement anchor" should still be central bank money. Since central bank money remains irreplaceable, the next question is: how should it enter the tokenized financial system? Schnabel summarizes three models. The first is for the central bank to directly issue tokenized reserves on a programmable ledger; the second is to maintain the existing central bank payment system and connect traditional payment systems and DLT platforms through technological bridges; the third is for private intermediaries to open integrated accounts at the central bank and then issue corresponding on-chain tokens based on the reserves in the accounts. In the third model, what circulates on-chain is not actual central bank liabilities, but tokens issued by private institutions and 100% backed by central bank reserves. While this structure enables on-chain settlement, it reintroduces the operational, legal, and financial risks of intermediaries. If different platforms create their own "central bank money wrappers," it could potentially lead to renewed liquidity fragmentation. The bridging model also presents similar problems. Although assets are traded on DLT, cash still returns to the traditional system for settlement, and the central bank itself does not truly utilize the programmability brought by tokenization. Schnabel argues that if central bank reserves themselves become on-chain assets, the situation is entirely different. The central bank could even directly implement monetary policy operations using smart contracts. For example, when central bank reserves and collateral exist in the same programmable environment, a repurchase operation can be completed atomically; smart contracts can also add collateral, replace securities, or automatically apply different interest rates in real time. Project Pine, conducted by the New York Fed and the Bank for International Settlements Innovation Center, has demonstrated that smart contracts can be used for central bank open market operations. More importantly, future tokenized finance itself may require central banks to provide faster liquidity tools. When smart contracts automatically trigger margin calls based on market prices, asset price changes can quickly translate into asset sales and liquidity demands; more frequent, non-nettable payments may also increase banks' intraday reserve requirements. Traditional monetary policy operating frameworks operating on fixed time windows may not be able to adapt to this speed. One unified ledger or multiple chains? The European Central Bank (ECB) is currently exploring various possible tokenized financial architectures through Project Appia. One approach is to establish a unified European ledger, placing central bank money, commercial bank money, and financial assets on the same infrastructure. A second approach involves the ECB system operating its own DLT (Digitalized Ledger), issuing tokenized central bank reserves, and connecting to private DLT networks. A third approach involves multiple interconnected ledgers forming the future financial system. The biggest advantage of a unified ledger is interoperability. If funds and assets reside on the same ledger, transactions can achieve true atomic settlement, and different assets can be programmed according to a unified protocol. South Korea's Project Hangang has begun testing a similar architecture. However, Schnabel does not conclude that a "single ledger" is the final answer. A unified ledger also implies a highly centralized critical financial infrastructure and could potentially lead to technology lock-in. Governance issues are equally thorny: Who is responsible if a smart contract malfunctions? Who decides which institutions can access the system? How is the software upgraded? Who sets the data confidentiality rules? How much control should the central bank actually have? Conversely, multiple interconnected ledgers can distribute system load, isolate faults, and allow for competition among different technological approaches, but at the cost of higher interoperability costs and potential liquidity fragmentation. Therefore, Schnabel ultimately did not offer a simple "one-chain" solution. She believes that future financial infrastructure needs to strike a balance between interoperability, financial resilience, innovation capabilities, and governance structure. If different networks can achieve sufficiently smooth interoperability, then establishing a super unified ledger covering all financial activities may not be necessary. The European Central Bank has begun to advance these discussions to the level of actual infrastructure. Project Pontes connects the existing TARGET system with DLT platforms on the market, and also plans to establish a DLT settlement platform operated by the Eurosystem itself, gradually incorporating smart contracts and 24/7 operational capabilities. The longer-term Project Appia is responsible for researching the overall architecture that a tokenized financial system in Europe might adopt.